Recommender system

929:

domain of citation recommender systems, users typically do not rate a citation or recommended article. In such cases, offline evaluations may use implicit measures of effectiveness. For instance, it may be assumed that a recommender system is effective that is able to recommend as many articles as possible that are contained in a research article's reference list. However, this kind of offline evaluations is seen critical by many researchers. For instance, it has been shown that results of offline evaluations have low correlation with results from user studies or A/B tests. A dataset popular for offline evaluation has been shown to contain duplicate data and thus to lead to wrong conclusions in the evaluation of algorithms. Often, results of so-called offline evaluations do not correlate with actually assessed user-satisfaction. This is probably because offline training is highly biased toward the highly reachable items, and offline testing data is highly influenced by the outputs of the online recommendation module. Researchers have concluded that the results of offline evaluations should be viewed critically.

1124:(AI) applications in recommendation systems are the advanced methodologies that leverage AI technologies, to enhance the performance recommendation engines. The AI-based recommender can analyze complex data sets, learning from user behavior, preferences, and interactions to generate highly accurate and personalized content or product suggestions. The integration of AI in recommendation systems has marked a significant evolution from traditional recommendation methods. Traditional methods often relied on inflexible algorithms that could suggest items based on general user trends or apparent similarities in content. In comparison, AI-powered systems have the capability to detect patterns and subtle distinctions that may be overlooked by traditional methods. These systems can adapt to specific individual preferences, thereby offering recommendations that are more aligned with individual user needs. This approach marks a shift towards more personalized, user-centric suggestions. 666:, content-based filtering, and other approaches. There is no reason why several different techniques of the same type could not be hybridized. Hybrid approaches can be implemented in several ways: by making content-based and collaborative-based predictions separately and then combining them; by adding content-based capabilities to a collaborative-based approach (and vice versa); or by unifying the approaches into one model. Several studies that empirically compared the performance of the hybrid with the pure collaborative and content-based methods and demonstrated that the hybrid methods can provide more accurate recommendations than pure approaches. These methods can also be used to overcome some of the common problems in recommender systems such as cold start and the sparsity problem, as well as the knowledge engineering bottleneck in 1046:, et al. criticized that "it is currently difficult to reproduce and extend recommender systems research results," and that evaluations are "not handled consistently". Konstan and Adomavicius conclude that "the Recommender Systems research community is facing a crisis where a significant number of papers present results that contribute little to collective knowledge often because the research lacks the evaluation to be properly judged and, hence, to provide meaningful contributions." As a consequence, much research about recommender systems can be considered as not reproducible. Hence, operators of recommender systems find little guidance in the current research for answering the question, which recommendation approaches to use in a recommender systems. 430:. Collaborative filtering is based on the assumption that people who agreed in the past will agree in the future, and that they will like similar kinds of items as they liked in the past. The system generates recommendations using only information about rating profiles for different users or items. By locating peer users/items with a rating history similar to the current user or item, they generate recommendations using this neighborhood. Collaborative filtering methods are classified as memory-based and model-based. A well-known example of memory-based approaches is the user-based algorithm, while that of model-based approaches is 1034:, IJCAI), has shown that on average less than 40% of articles could be reproduced by the authors of the survey, with as little as 14% in some conferences. The articles considers a number of potential problems in today's research scholarship and suggests improved scientific practices in that area. More recent work on benchmarking a set of the same methods came to qualitatively very different results whereby neural methods were found to be among the best performing methods. Deep learning and neural methods for recommender systems have been used in the winning solutions in several recent recommender system challenges, WSDM, 1055:

recommendation algorithms or scenarios led to strong changes in the effectiveness of a recommender system. They conclude that seven actions are necessary to improve the current situation: "(1) survey other research fields and learn from them, (2) find a common understanding of reproducibility, (3) identify and understand the determinants that affect reproducibility, (4) conduct more comprehensive experiments (5) modernize publication practices, (6) foster the development and use of recommendation frameworks, and (7) establish best-practice guidelines for recommender-systems research."

1150:(CF) is one of the most commonly used recommendation system algorithms. It generates personalized suggestions for users based on explicit or implicit behavioral patterns to form predictions. Specifically, it relies on external feedback such as star ratings, purchasing history and so on to make judgments. CF make predictions about users' preference based on similarity measurements. Essentially, the underlying theory is: "if user A is similar to user B, and if A likes item C, then it is likely that B also likes item C." 818:. From 2006 to 2009, Netflix sponsored a competition, offering a grand prize of $ 1,000,000 to the team that could take an offered dataset of over 100 million movie ratings and return recommendations that were 10% more accurate than those offered by the company's existing recommender system. This competition energized the search for new and more accurate algorithms. On 21 September 2009, the grand prize of US$ 1,000,000 was given to the BellKor's Pragmatic Chaos team using tiebreaking rules. 307:. Collaborative filtering approaches build a model from a user's past behavior (items previously purchased or selected and/or numerical ratings given to those items) as well as similar decisions made by other users. This model is then used to predict items (or ratings for items) that the user may have an interest in. Content-based filtering approaches utilize a series of discrete, pre-tagged characteristics of an item in order to recommend additional items with similar properties. 419: 1199:(ANN), is a deep learning model structure which aims to mimic a human brain. They comprise a series of neurons, each responsible for receiving and processing information transmitted from other interconnected neurons. Similar to a human brain, these neurons will change activation state based on incoming signals (training input and backpropagated output), allowing the system to adjust activation weights during the network learning phase. ANN is usually designed to be a 739:

learning that is of particular use in the area of recommender systems is the fact that the models or policies can be learned by providing a reward to the recommendation agent. This is in contrast to traditional learning techniques which rely on supervised learning approaches that are less flexible, reinforcement learning recommendation techniques allow to potentially train models that can be optimized directly on metrics of engagement, and user interest.

726:

current user session. Domains, where session-based recommendations are particularly relevant, include video, e-commerce, travel, music and more. Most instances of session-based recommender systems rely on the sequence of recent interactions within a session without requiring any additional details (historical, demographic) of the user. Techniques for session-based recommendations are mainly based on generative sequential models such as

1069: 625:

recommendation system is significantly less than when other content types from other services can be recommended. For example, recommending news articles based on news browsing is useful. Still, it would be much more useful when music, videos, products, discussions, etc., from different services, can be recommended based on news browsing. To overcome this, most content-based recommender systems now use some form of the hybrid system.

565:. Content-based filtering methods are based on a description of the item and a profile of the user's preferences. These methods are best suited to situations where there is known data on an item (name, location, description, etc.), but not on the user. Content-based recommenders treat recommendation as a user-specific classification problem and learn a classifier for the user's likes and dislikes based on an item's features. 1334:. Therefore, there is a risk that the market could become fragmented, leaving it to the viewer to visit various locations and find what they want to watch in a way that is time-consuming and complicated for them. By using a search and recommendation engine, viewers are provided with a central 'portal' from which to discover content from several sources in just one location. 748:

predict a rating for unexplored items of u by exploiting preference information on multiple criteria that affect this overall preference value. Several researchers approach MCRS as a multi-criteria decision making (MCDM) problem, and apply MCDM methods and techniques to implement MCRS systems. See this chapter for an extended introduction.

2834: 757:

pushing recommendations in certain circumstances, for instance, during a professional meeting, early morning, or late at night. Therefore, the performance of the recommender system depends in part on the degree to which it has incorporated the risk into the recommendation process. One option to manage this issue is

1139:. These advanced methods enhance system capabilities to predict user preferences and deliver personalized content more accurately. Each technique contributes uniquely. The following sections will introduce specific AI models utilized by a recommendation system by illustrating their theories and functionalities. 1236:

Natural language processing is a series of AI algorithms to make natural human language accessible and analyzable to a machine. It is a fairly modern technique inspired by the growing amount of textual information. For application in recommendation system, a common case is the Amazon customer review.

952:– In some situations, it is more effective to re-show recommendations, or let users re-rate items, than showing new items. There are several reasons for this. Users may ignore items when they are shown for the first time, for instance, because they had no time to inspect the recommendations carefully. 844:

A number of privacy issues arose around the dataset offered by Netflix for the Netflix Prize competition. Although the data sets were anonymized in order to preserve customer privacy, in 2007 two researchers from the University of Texas were able to identify individual users by matching the data sets

676:

is a good example of the use of hybrid recommender systems. The website makes recommendations by comparing the watching and searching habits of similar users (i.e., collaborative filtering) as well as by offering movies that share characteristics with films that a user has rated highly (content-based

290:

articles to television. As operators compete to be the gateway to home entertainment, personalized television is a key service differentiator. Academic content discovery has recently become another area of interest, with several companies being established to help academic researchers keep up to date

999:

is a measure of "how surprising the recommendations are". For instance, a recommender system that recommends milk to a customer in a grocery store might be perfectly accurate, but it is not a good recommendation because it is an obvious item for the customer to buy. " serves two purposes: First, the

905:

Evaluating the performance of a recommendation algorithm on a fixed test dataset will always be extremely challenging as it is impossible to accurately predict the reactions of real users to the recommendations. Hence any metric that computes the effectiveness of an algorithm in offline data will be

437:

A key advantage of the collaborative filtering approach is that it does not rely on machine analyzable content and therefore it is capable of accurately recommending complex items such as movies without requiring an "understanding" of the item itself. Many algorithms have been used in measuring user

330:) to seed a "station" that plays music with similar properties. User feedback is used to refine the station's results, deemphasizing certain attributes when a user "dislikes" a particular song and emphasizing other attributes when a user "likes" a song. This is an example of a content-based approach. 1029:

in recommender systems publications. The topic of reproducibility seems to be a recurrent issue in some Machine Learning publication venues, but does not have a considerable effect beyond the world of scientific publication. In the context of recommender systems a 2019 paper surveyed a small number

928:

The effectiveness of recommendation approaches is then measured based on how well a recommendation approach can predict the users' ratings in the dataset. While a rating is an explicit expression of whether a user liked a movie, such information is not available in all domains. For instance, in the

572:

is built to indicate the type of item this user likes. In other words, these algorithms try to recommend items similar to those that a user liked in the past or is examining in the present. It does not rely on a user sign-in mechanism to generate this often temporary profile. In particular, various

360:

created the first recommender system in 1979, called Grundy. She looked for a way to recommend users books they might like. Her idea was to create a system that asks users specific questions and classifies them into classes of preferences, or "stereotypes", depending on their answers. Depending on

250:

generators for video and music services, product recommenders for online stores, or content recommenders for social media platforms and open web content recommenders. These systems can operate using a single type of input, like music, or multiple inputs within and across platforms like news, books

783:

to offer personalized, context-sensitive recommendations. This is a particularly difficult area of research as mobile data is more complex than data that recommender systems often have to deal with. It is heterogeneous, noisy, requires spatial and temporal auto-correlation, and has validation and

756:

The majority of existing approaches to recommender systems focus on recommending the most relevant content to users using contextual information, yet do not take into account the risk of disturbing the user with unwanted notifications. It is important to consider the risk of upsetting the user by

738:

The recommendation problem can be seen as a special instance of a reinforcement learning problem whereby the user is the environment upon which the agent, the recommendation system acts upon in order to receive a reward, for instance, a click or engagement by the user. One aspect of reinforcement

1207:

ANN is widely used in recommendation systems for its power to utilize various data. Other than feedback data, ANN can incorporate non-feedback data which are too intricate for collaborative filtering to learn, and the unique structure allows ANN to identify extra signal from non-feedback data to

799:

to generate driving routes for taxi drivers in a city. This system uses GPS data of the routes that taxi drivers take while working, which includes location (latitude and longitude), time stamps, and operational status (with or without passengers). It uses this data to recommend a list of pickup

747:

Multi-criteria recommender systems (MCRS) can be defined as recommender systems that incorporate preference information upon multiple criteria. Instead of developing recommendation techniques based on a single criterion value, the overall preference of user u for the item i, these systems try to

624:

A key issue with content-based filtering is whether the system can learn user preferences from users' actions regarding one content source and use them across other content types. When the system is limited to recommending content of the same type as the user is already using, the value from the

322:

Last.fm creates a "station" of recommended songs by observing what bands and individual tracks the user has listened to on a regular basis and comparing those against the listening behavior of other users. Last.fm will play tracks that do not appear in the user's library, but are often played by

1262:

An emerging market for content discovery platforms is academic content. Approximately 6000 academic journal articles are published daily, making it increasingly difficult for researchers to balance time management with staying up to date with relevant research. Though traditional tools academic

974:

is particularly notable for the detailed personal information released in its dataset. Ramakrishnan et al. have conducted an extensive overview of the trade-offs between personalization and privacy and found that the combination of weak ties (an unexpected connection that provides serendipitous

787:

There are three factors that could affect the mobile recommender systems and the accuracy of prediction results: the context, the recommendation method and privacy. Additionally, mobile recommender systems suffer from a transplantation problem – recommendations may not apply in all regions (for

633:

of items, because as they also reflect aspects of the item like metadata, extracted features are widely concerned by the users. Sentiments extracted from the reviews can be seen as users' rating scores on the corresponding features. Popular approaches of opinion-based recommender system utilize

604:

representation (also called vector space representation). The system creates a content-based profile of users based on a weighted vector of item features. The weights denote the importance of each feature to the user and can be computed from individually rated content vectors using a variety of

725:

These recommender systems use the interactions of a user within a session to generate recommendations. Session-based recommender systems are used at YouTube and Amazon. These are particularly useful when history (such as past clicks, purchases) of a user is not available or not relevant in the

628:

Content-based recommender systems can also include opinion-based recommender systems. In some cases, users are allowed to leave text reviews or feedback on the items. These user-generated texts are implicit data for the recommender system because they are potentially rich resources of both

913:

In A/B tests, recommendations are shown to typically thousands of users of a real product, and the recommender system randomly picks at least two different recommendation approaches to generate recommendations. The effectiveness is measured with implicit measures of effectiveness such as

1054:

conducted a study of papers published in the field, as well as benchmarked some of the most popular frameworks for recommendation and found large inconsistencies in results, even when the same algorithms and data sets were used. Some researchers demonstrated that minor variations in the

3817:

Jannach, Dietmar; Lerche, Lukas; Gedikli, Fatih; Bonnin, Geoffray (June 10, 2013). "What Recommenders Recommend – an Analysis of Accuracy, Popularity, and Sales Diversity Effects". In Carberry, Sandra; Weibelzahl, Stephan; Micarelli, Alessandro; Semeraro, Giovanni (eds.).

1279:

that takes a researchers' authorized paper and citations as input. Whilst these recommendations have been noted to be extremely good, this poses a problem with early career researchers which may be lacking a sufficient body of work to produce accurate recommendations.

599:

Basically, these methods use an item profile (i.e., a set of discrete attributes and features) characterizing the item within the system. To abstract the features of the items in the system, an item presentation algorithm is applied. A widely used algorithm is the

2831: 238:

that provides suggestions for items that are most pertinent to a particular user. Recommender systems are particularly useful when an individual needs to choose an item from a potentially overwhelming number of items that a service may offer.

552:

originally used collaborative filtering to recommend new friends, groups, and other social connections by examining the network of connections between a user and their friends. Collaborative filtering is still used as part of hybrid systems.

1271:

provide a readily accessible database of journal articles, content recommendation in these cases are performed in a 'linear' fashion, with users setting 'alarms' for new publications based on keywords, journals or particular authors.

338:

problem, and is common in collaborative filtering systems. Whereas Pandora needs very little information to start, it is far more limited in scope (for example, it can only make recommendations that are similar to the original seed).

3045:, pp. 377-408, The Adaptive Web, Peter Brusilovsky, Alfred Kobsa, Wolfgang Nejdl (Ed.), Lecture Notes in Computer Science, Springer-Verlag, Berlin, Germany, Lecture Notes in Computer Science, Vol. 4321, May 2007, 978-3-540-72078-2. 3931:

Langer, Stefan (September 14, 2015). "A Comparison of Offline Evaluations, Online Evaluations, and User Studies in the Context of Research-Paper Recommender Systems". In Kapidakis, Sarantos; Mazurek, Cezary; Werla, Marcin (eds.).

1203:

model. Unlike regular machine learning where the underlying theoretical components are formal and rigid, the collaborative effects of neurons are not entirely clear, but modern experiments has shown the predictive power of ANN.

2483: 3998:

Beel, Joeran; Genzmehr, Marcel; Langer, Stefan; Nürnberger, Andreas; Gipp, Bela (January 1, 2013). "A comparative analysis of offline and online evaluations and discussion of research paper recommender system evaluation".

901:

are useful to assess the quality of a recommendation method. Diversity, novelty, and coverage are also considered as important aspects in evaluation. However, many of the classic evaluation measures are highly criticized.

1016:(CTR) for recommendations labeled as "Sponsored" were lower (CTR=5.93%) than CTR for identical recommendations labeled as "Organic" (CTR=8.86%). Recommendations with no label performed best (CTR=9.87%) in that study. 1006:– A recommender system is of little value for a user if the user does not trust the system. Trust can be built by a recommender system by explaining how it generates recommendations, and why it recommends an item. 1237:

Amazon will analyze the feedbacks comments from each customer and report relevant data to other customers for reference. The recent years have witnessed the development of various text analysis models, including

983:– Beel et al. found that user demographics may influence how satisfied users are with recommendations. In their paper they show that elderly users tend to be more interested in recommendations than younger users. 537:: The number of items sold on major e-commerce sites is extremely large. The most active users will only have rated a small subset of the overall database. Thus, even the most popular items have very few ratings. 909:

User studies are rather a small scale. A few dozens or hundreds of users are presented recommendations created by different recommendation approaches, and then the users judge which recommendations are best.

531:: There are millions of users and products in many of the environments in which these systems make recommendations. Thus, a large amount of computation power is often necessary to calculate recommendations. 5279: 3038: 334:

Each type of system has its strengths and weaknesses. In the above example, Last.fm requires a large amount of information about a user to make accurate recommendations. This is an example of the

4285: 833:

Many benefits accrued to the web due to the Netflix project. Some teams have taken their technology and applied it to other markets. Some members from the team that finished second place founded

246:, such as what product to purchase, what music to listen to, or what online news to read. Recommender systems are used in a variety of areas, with commonly recognised examples taking the form of 2325: 2520: 1038:. Moreover, neural and deep learning methods are widely used in industry where they are extensively tested. The topic of reproducibility is not new in recommender systems. By 2011, 1294:

In contrast to an engagement-based ranking system employed by social media and other digital platforms, a bridging-based ranking optimizes for content that is unifying instead of

4340: 937:

Typically, research on recommender systems is concerned with finding the most accurate recommendation algorithms. However, there are a number of factors that are also important.

399:

et al. discussed the problems of offline evaluations. Beel et al. have also provided literature surveys on available research paper recommender systems and existing challenges.

346:

since they help users discover items they might not have found otherwise. Of note, recommender systems are often implemented using search engines indexing non-traditional data.

274:

in order to discover and recommend appropriate content, whilst reducing ongoing maintenance and development costs. A content discovery platform delivers personalized content to

821:

The most accurate algorithm in 2007 used an ensemble method of 107 different algorithmic approaches, blended into a single prediction. As stated by the winners, Bell et al.:

4191: 4059:. 41st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2018). Ann Arbor, Michigan, USA: ACM. pp. 415–424. 1992:

Bobadilla, J.; Ortega, F.; Hernando, A.; Alcalá, J. (2011). "Improving collaborative filtering recommender system results and performance using genetic algorithms".

1326:

As the connected television landscape continues to evolve, search and recommendation are seen as having an even more pivotal role in the discovery of content. With

849:. As a result, in December 2009, an anonymous Netflix user sued Netflix in Doe v. Netflix, alleging that Netflix had violated United States fair trade laws and the 4895:

Konstan, Joseph A.; Adomavicius, Gediminas (January 1, 2013). "Toward identification and adoption of best practices in algorithmic recommender systems research".

392: 5283: 3084:

Chen, Minmin; Beutel, Alex; Covington, Paul; Jain, Sagar; Belletti, Francois; Chi, Ed (2018). "Top-K Off-Policy Correction for a REINFORCE Recommender System".

1051: 1000:

chance that users lose interest because the choice set is too uniform decreases. Second, these items are needed for algorithms to learn and improve themselves".

541:

One of the most famous examples of collaborative filtering is item-to-item collaborative filtering (people who buy x also buy y), an algorithm popularized by

1917:

Chen, Hung-Hsuan; Chen, Pu (January 9, 2019). "Differentiating Regularization Weights -- A Simple Mechanism to Alleviate Cold Start in Recommender Systems".

4760: 3740:. In: Proceedings of the 33rd International ACMSIGIR Conference on Research and Development in Information Retrieval, SIGIR 2010, pp. 210–217. ACM, New York 3058:

Hidasi, Balázs; Karatzoglou, Alexandros; Baltrunas, Linas; Tikk, Domonkos (March 29, 2016). "Session-based Recommendations with Recurrent Neural Networks".

3035: 395:

provided a new, alternate overview of recommender systems. Herlocker provides an additional overview of evaluation techniques for recommender systems, and

3711: 2233: 1030:

of hand-picked publications applying deep learning or neural methods to the top-k recommendation problem, published in top conferences (SIGIR, KDD, WWW,

605:

techniques. Simple approaches use the average values of the rated item vector while other sophisticated methods use machine learning techniques such as

4669:

Rendle, Steffen; Krichene, Walid; Zhang, Li; Anderson, John (September 22, 2020). "Neural Collaborative Filtering vs. Matrix Factorization Revisited".

1039: 1035: 3782:

Chen, Hung-Hsuan; Chung, Chu-An; Huang, Hsin-Chien; Tsui, Wen (September 1, 2017). "Common Pitfalls in Training and Evaluating Recommender Systems".

5538: 521:: For a new user or item, there is not enough data to make accurate recommendations. Note: one commonly implemented solution to this problem is the 5390: 3468: 827:

Our experience is that most efforts should be concentrated in deriving substantially different approaches, rather than refining a single technique.

2211:

BEEL, Joeran, et al. Paper recommender systems: a literature survey. International Journal on Digital Libraries, 2016, 17. Jg., Nr. 4, S. 305–338.

573:

candidate items are compared with items previously rated by the user, and the best-matching items are recommended. This approach has its roots in

4513: 2067: 1164:: Create a n-dimensional space where each axis represents a user's trait (ratings, purchases, etc.). Represent the user as a point in that space. 4854:

Ekstrand, Michael D.; Ludwig, Michael; Konstan, Joseph A.; Riedl, John T. (January 1, 2011). "Rethinking the recommender research ecosystem".

3616: 3548:

Pimenidis, Elias; Polatidis, Nikolaos; Mouratidis, Haralambos (August 3, 2018). "Mobile recommender systems: Identifying the major concepts".

2546: 1489: 323:

other users with similar interests. As this approach leverages the behavior of users, it is an example of a collaborative filtering technique.

3327:

Xin, Xin; Karatzoglou, Alexandros; Arapakis, Ioannis; Jose, Joemon (2020). "Self-Supervised Reinforcement Learning for Recommender Systems".

2315:." In Proceedings of the SIGCHI conference on Human factors in computing systems, pp. 194–201. ACM Press/Addison-Wesley Publishing Co., 1995. 2295:." In Proceedings of the SIGCHI conference on Human factors in computing systems, pp. 210–217. ACM Press/Addison-Wesley Publishing Co., 1995. 396: 197: 5306: 5147:

Wu, L. (May 2023). "A Survey on Accuracy-Oriented Neural Recommendation: From Collaborative Filtering to Information-Rich Recommendation".

2271: 5405: 3669: 2431:

Herlocker, J. L.; Konstan, J. A.; Terveen, L. G.; Riedl, J. T. (January 2004). "Evaluating collaborative filtering recommender systems".

1047: 629:

feature/aspects of the item and users' evaluation/sentiment to the item. Features extracted from the user-generated reviews are improved

310:

The differences between collaborative and content-based filtering can be demonstrated by comparing two early music recommender systems,

431: 94: 1564:

Gupta, Pankaj; Goel, Ashish; Lin, Jimmy; Sharma, Aneesh; Wang, Dong; Zadeh, Reza (2013). "WTF: the who to follow service at Twitter".

944:– Users tend to be more satisfied with recommendations when there is a higher intra-list diversity, e.g. items from different artists. 2308: 2971:

The Deep Learning–Based Recommender System "Pubmender" for Choosing a Biomedical Publication Venue: Development and Validation Study

2769: 1012:– User satisfaction with recommendations may be influenced by the labeling of the recommendations. For instance, in the cited study 5513: 1186:: The system will analyze the similar preference of the k neighbors. The system will make recommendations based on that similarity 925:

Offline evaluations are based on historic data, e.g. a dataset that contains information about how users previously rated movies.

4759:

Schifferer, Benedikt; Deotte, Chris; Puget, Jean-François; de Souza Pereira, Gabriel; Titericz, Gilberto; Liu, Jiwei; Ak, Ronay.

4438:"Do not blame it on the algorithm: an empirical assessment of multiple recommender systems and their impact on content diversity" 3348:

Ie, Eugene; Jain, Vihan; Narvekar, Sanmit; Agarwal, Ritesh; Wu, Rui; Cheng, Heng-Tze; Chandra, Tushar; Boutilier, Craig (2019).

962:

using collaborative filtering can be problematic from a privacy point of view. Many European countries have a strong culture of

255:. Recommender systems have also been developed to explore research articles and experts, collaborators, and financial services. 5641: 368:

at Columbia University, and implemented at scale and worked through in technical reports and publications from 1994 onwards by

1314:

planned to pilot in 2024. Aviv Ovadya also argues for implementing bridging-based algorithms in major platforms by empowering

1153:

There are many models available for collaborative filtering. For AI-applied collaborative filtering, a common model is called

958:– Recommender systems usually have to deal with privacy concerns because users have to reveal sensitive information. Building 711:: One recommendation technique is applied and produces some sort of model, which is then the input used by the next technique. 5548: 5523: 5052: 5011: 4912: 4871: 4817: 4735: 4698: 4638: 4315: 4174: 4026: 3982: 3949: 3907: 3853: 3398: 3303: 3258: 3182: 3127: 2895: 2859: 2808: 2571: 2506: 2292: 1976: 1840: 1472: 1349: 1031: 838: 171: 2975: 286:. A large range of content discovery platforms currently exist for various forms of content ranging from news articles and 3108:

Yifei, Ma; Narayanaswamy, Balakrishnan; Haibin, Lin; Hao, Ding (2020). "Temporal-Contextual Recommendation in Real-Time".

2253: 2163:

Herz, Frederick SM. "Customized electronic newspapers and advertisements." U.S. Patent 7,483,871, issued January 27, 2009.

970:

can result in a negative customer response. Much research has been conducted on ongoing privacy issues in this space. The

3526: 2484:"A comparative analysis of offline and online evaluations and discussion of research paper recommender system evaluation" 2074: 2064: 1086: 1025:

Recommender systems are notoriously difficult to evaluate offline, with some researchers claiming that this has led to a

2945: 1794: 705:: Recommenders are given strict priority, with the lower priority ones breaking ties in the scoring of the higher ones. 5431: 5365: 4841: 3878:

Proceedings of the 24th annual international ACM SIGIR conference on Research and development in information retrieval

3752:

Proceedings of the 24th annual international ACM SIGIR conference on Research and development in information retrieval

5577: 5506: 5236: 2086: 1715: 1683: 1581: 1108: 190: 89: 4051: 2199: 2173: 1249:(LDA), etc. Their uses have consistently aimed to provide customers with more precise and tailored recommendations. 4135:

Ziegler CN, McNee SM, Konstan JA, Lausen G (2005). "Improving recommendation lists through topic diversification".

1354: 335: 69: 1599:"A capable multimedia content discovery platform based on visual content analysis and intelligent data enrichment" 788:

instance, it would be unwise to recommend a recipe in an area where all of the ingredients may not be available).

4771: 4221: 967: 1180:: Based on the computed distances, find k nearest neighbors of the user to which we want to make recommendations 5631: 5076:

Khanal, S.S. (July 2020). "A systematic review: machine learning based recommendation systems for e-learning".

3715: 2230: 1330:-connected devices, consumers are projected to have access to content from linear broadcast sources as well as 1302:

and Remesh which have been used around the world to help find more consensus around specific political issues.

1090: 4897:

Proceedings of the International Workshop on Reproducibility and Replication in Recommender Systems Evaluation

4341:"The Impact of Demographics (Age and Gender) and Other User Characteristics on Evaluating Recommender Systems" 4001:

Proceedings of the International Workshop on Reproducibility and Replication in Recommender Systems Evaluation

3649: 2555:

Proceedings of the International Workshop on Reproducibility and Replication in Recommender Systems Evaluation

2491:

Proceedings of the International Workshop on Reproducibility and Replication in Recommender Systems Evaluation

4053:

Should I Follow the Crowd? A Probabilistic Analysis of the Effectiveness of Popularity in Recommender Systems

2383:"Toward the Next Generation of Recommender Systems: A Survey of the State-of-the-Art and Possible Extensions" 1154: 774: 647: 439: 303:

and content-based filtering (also known as the personality-based approach), as well as other systems such as

3422:

Lakiotaki, K.; Matsatsinis; Tsoukias, A (March 2011). "Multicriteria User Modeling in Recommender Systems".

2677:

Koren, Yehuda; Volinsky, Chris (August 1, 2009). "Matrix Factorization Techniques for Recommender Systems".

2338:

Montaner, M.; Lopez, B.; de la Rosa, J. L. (June 2003). "A Taxonomy of Recommender Agents on the Internet".

2328:." In Proceedings of the 1994 ACM conference on Computer supported cooperative work, pp. 175–186. ACM, 1994. 1817: 975:

recommendations) and other data sources can be used to uncover identities of users in an anonymized dataset.

364:

Another early recommender system, called a "digital bookshelf", was described in a 1990 technical report by

5636: 5626: 4716:"Are We Evaluating Rigorously? Benchmarking Recommendation for Reproducible Evaluation and Fair Comparison" 4499:

Proceedings of the first international joint conference on Autonomous agents and multiagent systems: part 1

1554:

Resnick, Paul, and Hal R. Varian. "Recommender systems." Communications of the ACM 40, no. 3 (1997): 56–58.

1384: 1318:

that are representative of the platform's users to control the design and implementation of the algorithm.

1242: 850: 235: 183: 5037:

2020 2nd International Conference on Advances in Computing, Communication Control and Networking (ICACCCN)

3936:. Lecture Notes in Computer Science. Vol. 9316. Springer International Publishing. pp. 153–168. 5554: 2152:

System and method for providing recommendation of goods and services based on recorded purchasing history

1708:

Proceedings of the 19th National Conference on Innovative Applications of Artificial Intelligence, vol. 2

1246: 1136: 120: 4521:

Proceedings of the 17th International Conference on Theory and Practice of Digital Libraries (TPDL 2013)

4462: 4348:

Proceedings of the 17th International Conference on Theory and Practice of Digital Libraries (TPDL 2013)

4199:

Proceedings of the 17th International Conference on Theory and Practice of Digital Libraries (TPDL 2013)

3750:

Turpin, Andrew H; Hersh, William (2001). "Why batch and user evaluations do not give the same results".

3597: 3475: 251:

and search queries. There are also popular recommender systems for specific topics like restaurants and

4519:. In Trond Aalberg, Milena Dobreva, Christos Papatheodorou, Giannis Tsakonas, Charles Farrugia (eds.). 4346:. In Trond Aalberg; Milena Dobreva; Christos Papatheodorou; Giannis Tsakonas; Charles Farrugia (eds.). 4197:. In Trond Aalberg; Milena Dobreva; Christos Papatheodorou; Giannis Tsakonas; Charles Farrugia (eds.). 2113: 1344: 898: 618: 385: 125: 4370: 4192:"Persistence in Recommender Systems: Giving the Same Recommendations to the Same Users Multiple Times" 4437: 4251: 3685: 3354:

Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence (IJCAI-19)

1196: 4601:"Are we really making much progress? A worrying analysis of recent neural recommendation approaches" 4009: 3890: 3436: 3373:

Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining

3288:

Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining

3110:

Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining

2989: 2691: 2399: 1289: 870:

Evaluation is important in assessing the effectiveness of recommendation algorithms. To measure the

4761:"Using Deep Learning to Win the Booking.com WSDM WebTour21 Challenge on Sequential Recommendations" 3623: 2445: 1497: 1394: 1238: 854: 727: 487: 450: 391:

Montaner provided the first overview of recommender systems from an intelligent agent perspective.

84: 79: 5457: 5335: 4298: 3836: 2782:

Felício, Crícia Z.; Paixão, Klérisson V.R.; Barcelos, Celia A.Z.; Preux, Philippe (July 9, 2017).

2638:. In Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence (UAI'98). 1295: 4542:

Ferrari Dacrema, Maurizio; Boglio, Simone; Cremonesi, Paolo; Jannach, Dietmar (January 8, 2021).

4271: 4154: 2151: 1364: 1359: 1147: 1121: 1079: 890: 663: 427: 413: 304: 300: 243: 130: 74: 38: 4986:

Said, Alan; Bellogín, Alejandro (October 1, 2014). "Comparative recommender system evaluation".

4272:

Naren Ramakrishnan; Benjamin J. Keller; Batul J. Mirza; Ananth Y. Grama; George Karypis (2001).

3881: 3873: 2220:

RICH, Elaine. User modeling via stereotypes. Cognitive science, 1979, 3. Jg., Nr. 4, S. 329–354.

2078: 2058: 2052: 1452: 699:: Recommendations from different recommenders are presented together to give the recommendation. 326:

Pandora uses the properties of a song or artist (a subset of the 400 attributes provided by the

4293: 4289: 4273: 4004: 3885: 3831: 3737: 3666: 3431: 2887: 2686: 2440: 2394: 1414: 1404: 1374: 1315: 1026: 606: 511: 449:

When building a model from a user's behavior, a distinction is often made between explicit and

99: 3827: 3512:

Yong Ge; Hui Xiong; Alexander Tuzhilin; Keli Xiao; Marco Gruteser; Michael J. Pazzani (2010).

893:, the latter having been used in the Netflix Prize. The information retrieval metrics such as 1389: 1221:: sequence of pages visited, time spent on different parts of a website, mouse movement, etc. 989:– When users can participate in the recommender system, the issue of fraud must be addressed. 639: 578: 574: 361:

users' stereotype membership, they would then get recommendations for books they might like.

48: 5607: 5253: 4252:"Evaluating recommender systems from the user's perspective: survey of the state of the art" 3819: 3290:. KDD '18. London, United Kingdom: Association for Computing Machinery. pp. 1831–1839. 3227:

Li, Jing; Ren, Pengjie; Chen, Zhumin; Ren, Zhaochun; Lian, Tao; Ma, Jun (November 6, 2017).

3157:

Proceedings of the 27th ACM International Conference on Information and Knowledge Management

2879: 5482: 4796:

Volkovs, Maksims; Rai, Himanshu; Cheng, Zhaoyue; Wu, Ga; Lu, Yichao; Sanner, Scott (2018).

4418: 3513: 3235:. CIKM '17. Singapore, Singapore: Association for Computing Machinery. pp. 1419–1428. 2787: 1668:

Proceedings of the 11th Annual International ACM/IEEE Joint Conference on Digital Libraries

1660: 1171: 894: 327: 135: 3826:. Lecture Notes in Computer Science. Vol. 7899. Springer Berlin Heidelberg. pp. 2969:

X.Y. Feng, H. Zhang, Y.J. Ren, P.H. Shang, Y. Zhu, Y.C. Liang, R.C. Guan, D. Xu, (2019), "

2305: 8: 5254:""Extending and Customizing Content Discovery for the Legal Academic Com" by Sima Mirkin" 4338: 4189: 3521:. Proceedings of the 16th ACM SIGKDD Int'l Conf. on Knowledge Discovery and Data Mining. 1956: 1409: 1399: 1331: 963: 791:

One example of a mobile recommender system are the approaches taken by companies such as

443: 5536: 5529: 4422: 2660: 2382: 2200:

Automated detection and exposure of behavior-based relationships between browsable items

5341: 5197: 5156: 5093: 5058: 5017: 4968: 4939:

Breitinger, Corinna; Langer, Stefan; Lommatzsch, Andreas; Gipp, Bela (March 12, 2016).

4918: 4877: 4823: 4741: 4674: 4644: 4608: 4581: 4555: 4475: 4395: 4321: 4232: 4117: 4032: 3913: 3799: 3575: 3557: 3449: 3404: 3376: 3350:"SlateQ: A Tractable Decomposition for Reinforcement Learning with Recommendation Sets" 3328: 3309: 3264: 3236: 3207: 3188: 3160: 3133: 3085: 3059: 2814: 2795:. UMAP '17. Bratislava, Slovakia: Association for Computing Machinery. pp. 32–40. 2704: 2639: 2616: 2577: 2512: 2458: 2412: 2378: 2355: 2133: 1934: 1896: 1775: 1628: 1013: 919: 886: 762: 643: 522: 267: 156: 5044: 4940: 4544:"A Troubling Analysis of Reproducibility and Progress in Recommender Systems Research" 4201:. Lecture Notes of Computer Science (LNCS). Vol. 8092. Springer. pp. 390–394 3206:

Kang, Wang-Cheng; McAuley, Julian (2018). "Self-Attentive Sequential Recommendation".

2770:

Discovery of Hidden Similarity on Collaborative Filtering to Overcome Sparsity Problem

2596: 2553: 5603: 5544: 5519: 5502: 5384: 5232: 5097: 5062: 5048: 5035:

Verma, P.; Sharma, S. (2020). "Artificial Intelligence based Recommendation System".

5007: 4960: 4908: 4867: 4813: 4745: 4731: 4694: 4648: 4634: 4585: 4494: 4479: 4467: 4311: 4170: 4121: 4022: 3978: 3945: 3903: 3849: 3820: 3394: 3369:"Reinforcement Learning to Optimize Long-term User Engagement in Recommender Systems" 3367:

Zou, Lixin; Xia, Long; Ding, Zhuoye; Song, Jiaxing; Liu, Weidong; Yin, Dawei (2019).

3299: 3254: 3178: 3137: 3123: 2891: 2880: 2875: 2855: 2804: 2620: 2567: 2502: 2462: 2416: 2137: 2082: 2023: 1972: 1900: 1888: 1836: 1767: 1711: 1679: 1620: 1577: 1468: 1276: 1043: 503:

Obtaining a list of items that a user has listened to or watched on his/her computer.

478: 104: 5021: 4827: 4714:

Sun, Zhu; Yu, Di; Fang, Hui; Yang, Jie; Qu, Xinghua; Zhang, Jie; Geng, Cong (2020).

3917: 3803: 3579: 3453: 3408: 3313: 3268: 2708: 2372: 2370: 2359: 1938: 1632: 765:. This system combines a content-based technique and a contextual bandit algorithm. 5575: 5280:"Mendeley, Elsevier and the importance of content discovery to academic publishers" 5207: 5166: 5127: 5085: 5040: 4999: 4991: 4952: 4900: 4881: 4859: 4805: 4723: 4684: 4626: 4618: 4573: 4565: 4514:"Sponsored vs. Organic (Research Paper) Recommendations and the Impact of Labeling" 4457: 4449: 4399: 4385: 4325: 4303: 4236: 4162: 4109: 4060: 4036: 4014: 3970: 3937: 3895: 3841: 3791: 3567: 3493: 3441: 3386: 3291: 3246: 3192: 3170: 3113: 3015: 2924: 2796: 2789:

Proceedings of the 25th Conference on User Modeling, Adaptation and Personalization

2696: 2608: 2581: 2559: 2516: 2494: 2450: 2404: 2347: 2324:

Resnick, Paul, Neophytos Iacovou, Mitesh Suchak, Peter Bergström, and John Riedl. "

2186: 2125: 2071: 2035: 2001: 1964: 1926: 1880: 1828: 1779: 1757: 1671: 1610: 1569: 1532: 1460: 1128: 610: 473:

Presenting two items to a user and asking him/her to choose the better one of them.

343: 287: 5186:"Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications" 4972: 4922: 4453: 4436:

Möller, Judith; Trilling, Damian; Helberger, Natali; van Es, Bram (July 3, 2018).

4339:

Joeran Beel; Stefan Langer; Andreas Nürnberger; Marcel Genzmehr (September 2013).

4190:

Joeran Beel; Stefan Langer; Marcel Genzmehr; Andreas Nürnberger (September 2013).

3159:. CIKM '18. Torino, Italy: Association for Computing Machinery. pp. 843–852. 2818: 2129: 1832: 841:. 4-Tell, Inc. created a Netflix project–derived solution for ecommerce websites. 4797: 4715: 4166: 3941: 3673: 3368: 3233:

Proceedings of the 2017 ACM on Conference on Information and Knowledge Management

3042: 2992:, Semantic Web – Interoperability, Usability, Applicability 1 (2010) 1, IOS Press 2929: 2912: 2838: 2367: 2312: 2257: 2237: 2039: 2005: 1968: 1721: 1307: 915: 825:

Predictive accuracy is substantially improved when blending multiple predictors.

549: 454: 5537:

Jannach, Dietmar; Markus Zanker; Alexander Felfernig; Gerhard Friedrich (2010).

4222:"Is seeing believing?: how recommender system interfaces affect users' opinions" 3845: 2547:"Research paper recommender system evaluation: A quantitative literature survey" 1700: 1464: 1215:: what specify time and date or a season that a user interacts with the platform 5212: 5185: 5132: 5115: 5089: 4113: 4092: 3974: 2250: 1537: 1520: 1419: 1264: 834: 800:

points along a route, with the goal of optimizing occupancy times and profits.

667: 614: 506:

Analyzing the user's social network and discovering similar likes and dislikes.

369: 365: 140: 5432:"YouTube Adding Experimental Community Notes Feature to Battle Misinformation" 5170: 4956: 4390: 3153:"Recurrent Neural Networks with Top-k Gains for Session-based Recommendations" 2612: 2351: 2070:

Conference on Research and Development in Information Retrieval (SIGIR 2002).

1884: 1615: 1598: 621:

in order to estimate the probability that the user is going to like the item.

5620: 5582: 4964: 4471: 3571: 3522: 3466: 2911:

Wang, Donghui; Liang, Yanchun; Xu, Dong; Feng, Xiaoyue; Guan, Renchu (2018).

2545:

Beel, J.; Langer, S.; Genzmehr, M.; Gipp, B.; Breitinger, C. (October 2013).

1892: 1624: 1379: 1132: 971: 959: 871: 815: 809: 651: 315: 283: 279: 252: 166: 5611:

Proceedings of the Eighteenth National Conference on Artificial Intelligence

4995: 4904: 4863: 4809: 4727: 4689: 4622: 4064: 4018: 3795: 3765: 3390: 3295: 3250: 3174: 3118: 2800: 2724:"Application of Dimensionality Reduction in Recommender System A Case Study" 2723: 2563: 2498: 2024:"A survey of active learning in collaborative filtering recommender systems" 1675: 1573: 1275:

Google Scholar provides an 'Updates' tool that suggests articles by using a

470:

Asking a user to rank a collection of items from favorite to least favorite.

5576:

Robert M. Bell; Jim Bennett; Yehuda Koren & Chris Volinsky (May 2009).

3511: 1771: 1369: 585: 569: 381: 5587: 5307:"Social media algorithms can be redesigned to bridge divides — here's how" 4229:

Proceedings of the SIGCHI conference on Human factors in computing systems

4219: 3899: 3598:"A $ 1 Million Research Bargain for Netflix, and Maybe a Model for Others" 3283: 3228: 3152: 2783: 2658: 2454: 418: 5462:

Belfer Center for Science and International Affairs at Harvard University

4630: 4577: 4543: 4068: 3349: 2784:"A Multi-Armed Bandit Model Selection for Cold-Start User Recommendation" 2408: 1424: 1058: 996: 879: 874:

of recommender systems, and compare different approaches, three types of

857:, led to the cancellation of a second Netflix Prize competition in 2010. 814:

One of the events that energized research in recommender systems was the

693:: Choosing among recommendation components and applying the selected one. 687:: Combining the score of different recommendation components numerically. 635: 377: 357: 291:

with relevant academic content and serendipitously discover new content.

5501:

Kim Falk (d 2019), Practical Recommender Systems, Manning Publications,

5003: 4307: 3445: 2700: 1858:

Content-based book recommendation using learning for text categorization

1699:

Felfernig, Alexander; Isak, Klaus; Szabo, Kalman; Zachar, Peter (2007).

1659:

Chen, Hung-Hsuan; Gou, Liang; Zhang, Xiaolong; Giles, Clyde Lee (2011).

5109: 5107: 4600: 4094: 3282:

Liu, Qiao; Zeng, Yifu; Mokhosi, Refuoe; Zhang, Haibin (July 19, 2018).

2661:

Empirical Analysis of Predictive Algorithms for Collaborative Filtering

2636:

Empirical analysis of predictive algorithms for collaborative filtering

2633: 2293:

Social information filtering: algorithms for automating "word of mouth"

2054: 1093: in this section. Unsourced material may be challenged and removed. 875: 780: 601: 542: 438:

similarity or item similarity in recommender systems. For example, the

426:

One approach to the design of recommender systems that has wide use is

4599:

Ferrari Dacrema, Maurizio; Cremonesi, Paolo; Jannach, Dietmar (2019).

2970: 2913:"A content-based recommender system for computer science publications" 2757:. AAAI Workshop in Semantic Web Personalization, San Jose, California. 2326:

GroupLens: an open architecture for collaborative filtering of netnews

2174:

System and method for providing access to data using customer profiles

1746:"How to tame the flood of literature : Nature News & Comment" 1647:

ExpertSeer: a Keyphrase Based Expert Recommender for Digital Libraries

5608:

Content-Boosted Collaborative Filtering for Improved Recommendations.

4541: 1521:"A systematic review and research perspective on recommender systems" 1327: 1200: 510:

Collaborative filtering approaches often suffer from three problems:

161: 53: 43: 5104: 4569: 3967:

Proceedings of the 2017 SIAM International Conference on Data Mining

3020: 3003: 2832:

Collaborative Recommendations Using Item-to-Item Similarity Mappings

2251:

Newsgroup Clustering Based On User Behavior-A Recommendation Algebra

1955:

Rubens, Neil; Elahi, Mehdi; Sugiyama, Masashi; Kaplan, Dain (2016).

1930: 1762: 1745: 1068: 349:

Recommender systems have been the focus of several granted patents.

5406:"YouTube's community notes feature rips a page out of X's playbook" 5202: 5161: 4679: 4613: 4560: 4220:

Cosley, D.; Lam, S.K.; Albert, I.; Konstan, J.A.; Riedl, J (2003).

3562: 3381: 3333: 3241: 3212: 3165: 3090: 3064: 2946:"Online Recommender Systems – How Does a Website Know What I Want?" 2187:

Playlist-based detection of similar digital works and work creators

1871:

Haupt, Jon (June 1, 2009). "Last.fm: People-Powered Online Radio".

1646: 630: 271: 263: 247: 5563:

Computing Taste: Algorithms and the Makers of Music Recommendation

4493:

Montaner, Miquel; López, Beatriz; de la Rosa, Josep Lluís (2002).

4137:

Proceedings of the 14th international conference on World Wide Web

4093:

Cañamares, Rocío; Castells, Pablo; Moffat, Alistair (March 2020).

2755:

Using viewing time to infer user preference in recommender systems

2644: 1566:

Proceedings of the 22nd International Conference on World Wide Web

5260:. Digital Commons @ American University Washington College of Law 4758: 4512:

Beel, Joeran, Langer, Stefan, Genzmehr, Marcel (September 2013).

2306:

Recommending and evaluating choices in a virtual community of use

1597:

Baran, Remigiusz; Dziech, Andrzej; Zeja, Andrzej (June 1, 2018).

1311: 1303: 1170:: 'Distance' measures how far apart users are in this space. See 733: 673: 311: 275: 3686:"Netflix Spilled Your Brokeback Mountain Secret, Lawsuit Claims" 2595:

Beel, J.; Gipp, B.; Langer, S.; Breitinger, C. (July 26, 2015).

595:

A history of the user's interaction with the recommender system.

4598: 4049: 1453:"Recommender Systems: Techniques, Applications, and Challenges" 1299: 1268: 568:

In this system, keywords are used to describe the items, and a

476:

Asking a user to create a list of items that he/she likes (see

4938: 4856:

Proceedings of the fifth ACM conference on Recommender systems

4798:"Two-stage Model for Automatic Playlist Continuation at Scale" 3749: 3057: 2304:

Hill, Will, Larry Stead, Mark Rosenstein, and George Furnas. "

853:

by releasing the datasets. This, as well as concerns from the

761:, a system which models the context-aware recommendation as a 662:

Most recommender systems now use a hybrid approach, combining

561:

Another common approach when designing recommender systems is

422:

An example of collaborative filtering based on a rating system

384:, also at MIT, whose work with GroupLens was awarded the 2010 4605:

Proceedings of the 13th ACM Conference on Recommender Systems

4435: 4157:. In Ricci, Francesco; Rokach, Lior; Shapira, Bracha (eds.). 3874:"Why batch and user evaluations do not give the same results" 2185:

Harbick, Andrew V., Ryan J. Snodgrass, and Joel R. Spiegel. "

1991: 1959:. In Ricci, Francesco; Rokach, Lior; Shapira, Bracha (eds.). 1455:. In Ricci, Francesco; Rokach, Lior; Shapira, Bracha (eds.). 4988:

Proceedings of the 8th ACM Conference on Recommender systems

3997: 3547: 3421: 3151:

Hidasi, Balázs; Karatzoglou, Alexandros (October 17, 2018).

2430: 2198:

Linden, Gregory D., Brent Russell Smith, and Nida K. Zada. "

2172:

Herz, Frederick, Lyle Ungar, Jian Zhang, and David Wachob. "

460:

Examples of explicit data collection include the following:

5613:(AAAI-2002), pp. 187–192, Edmonton, Canada, July 2002. 4668: 3112:. Association for Computing Machinery. pp. 2291–2299. 3107: 2272:"A digital bookshelf: original work on recommender systems" 846: 796: 792: 500:

Keeping a record of the items that a user purchases online.

373: 4853: 3816: 3614: 3326: 2781: 2112:

Bi, Xuan; Qu, Annie; Wang, Junhui; Shen, Xiaotong (2017).

1127:

Recommendation systems widely adopt AI techniques such as

829:

Consequently, our solution is an ensemble of many methods.

779:

Mobile recommender systems make use of internet-accessing

730:, Transformers, and other deep-learning-based approaches. 16:

Information filtering system to predict users' preferences

5511: 4941:"Towards reproducibility in recommender-systems research" 4802:

Proceedings of the ACM Recommender Systems Challenge 2018

4371:"Recommender systems: from algorithms to user experience" 4134: 2721: 2634:

John S. Breese; David Heckerman & Carl Kadie (1998).

2597:"Research Paper Recommender Systems: A Literature Survey" 1954: 1670:. Association for Computing Machinery. pp. 231–240. 1661:"CollabSeer: a search engine for collaboration discovery" 1568:. Association for Computing Machinery. pp. 505–514. 588:, the system mostly focuses on two types of information: 494:

Observing the items that a user views in an online store.

4990:. RecSys '14. New York, NY, USA: ACM. pp. 129–136. 4858:. RecSys '11. New York, NY, USA: ACM. pp. 133–140. 4412: 3965:

Basaran, Daniel; Ntoutsi, Eirini; Zimek, Arthur (2017).

3002:

Gomez-Uribe, Carlos A.; Hunt, Neil (December 28, 2015).

2722:

Sarwar, B.; Karypis, G.; Konstan, J.; Riedl, J. (2000).

2594: 1698: 1451:

Ricci, Francesco; Rokach, Lior; Shapira, Bracha (2022).

4492: 4153:

Castells, Pablo; Hurley, Neil J.; Vargas, Saúl (2015).

3467:

Gediminas Adomavicius; Nikos Manouselis; YoungOk Kwon.

3083: 2659:

Breese, John S.; Heckerman, David; Kadie, Carl (1998).

2337: 1795:"Netflix Revamps iPad App to Improve Content Discovery" 1701:"The VITA Financial Services Sales Support Environment" 299:

Recommender systems usually make use of either or both

4899:. RepSys '13. New York, NY, USA: ACM. pp. 23–28. 4768:

WSDM '21: ACM Conference on Web Search and Data Mining

4511: 3934:

Research and Advanced Technology for Digital Libraries

2544: 1059:

Artificial intelligence applications in recommendation

270:

which uses recommender system tools. It utilizes user

5258:

Articles in Law Reviews & Other Academic Journals

4003:. RepSys '13. New York, NY, USA: ACM. pp. 7–14. 3872:

Turpin, Andrew H.; Hersh, William (January 1, 2001).

3714:. Netflix Prize Forum. March 12, 2010. Archived from 3347: 2752: 2022:

Elahi, Mehdi; Ricci, Francesco; Rubens, Neil (2016).

742: 5366:"Elon Musk keeps Birdwatch alive — under a new name" 4249: 4152: 4095:"Offline Evaluation Options for Recommender Systems" 1208:

boost user experience. Following are some examples:

720: 657: 5149:

IEEE Transactions on Knowledge and Data Engineering

3964: 3281: 2753:Parsons, J.; Ralph, P.; Gallagher, K. (July 2004). 2387:

IEEE Transactions on Knowledge and Data Engineering

2376: 966:, and every attempt to introduce any level of user 5364:Smalley, Alex Mahadevan, Seth (November 8, 2022). 4894: 3958: 3781: 3736:Lathia, N., Hailes, S., Capra, L., Amatriain, X.: 3507: 3505: 3008:ACM Transactions on Management Information Systems 2176:." U.S. Patent 8,056,100, issued November 8, 2011. 2060:Methods and Metrics for Cold-Start Recommendations 1855: 1563: 1450: 1142: 5305:Thorburn, Luke; Ovadya, Aviv (October 31, 2023). 4795: 3498:(Ph.D.), Institut National des Télécommunications 3150: 2943: 2874: 2768:Sanghack Lee and Jihoon Yang and Sung-Yong Park, 2482:Beel, J.; Genzmehr, M.; Gipp, B. (October 2013). 1919:ACM Transactions on Knowledge Discovery from Data 1815: 1658: 1459:(3 ed.). New York: Springer. pp. 1–35. 751: 464:Asking a user to rate an item on a sliding scale. 5618: 5116:"Artificial intelligence in recommender systems" 4720:Fourteenth ACM Conference on Recommender Systems 4671:Fourteenth ACM Conference on Recommender Systems 4413:Ricci F, Rokach L, Shapira B, Kantor BP (2011). 2481: 2021: 1860:. In Workshop Recom. Sys.: Algo. and Evaluation. 1490:"How Computers Know What We Want — Before We Do" 342:Recommender systems are a useful alternative to 4265: 4148: 4146: 4050:Cañamares, Rocío; Castells, Pablo (July 2018). 3502: 3229:"Neural Attentive Session-based Recommendation" 3001: 2910: 2825: 2202:." U.S. Patent 9,070,156, issued June 30, 2015. 2189:." U.S. Patent 8,468,046, issued June 18, 2013. 2118:Journal of the American Statistical Association 2063:. Proceedings of the 25th Annual International 1809: 1596: 1550: 1548: 837:, a recommendation engine that's active in the 5304: 4155:"Novelty and Diversity in Recommender Systems" 3880:. SIGIR '01. New York, NY, USA: ACM. pp. 3822:User Modeling, Adaptation, and Personalization 3591: 3589: 3495:DRARS, A Dynamic Risk-Aware Recommender System 3485: 3366: 2843: 2154:." U.S. Patent 7,222,085, issued May 22, 2007. 1231: 734:Reinforcement learning for recommender systems 4713: 4161:(2 ed.). Springer US. pp. 881–918. 4086: 4043: 3515:An Energy-Efficient Mobile Recommender System 2676: 1963:(2 ed.). Springer US. pp. 809–846. 1950: 1948: 1823:. In Claude Sammut; Geoffrey I. Webb (eds.). 1257: 768: 191: 5389:: CS1 maint: multiple names: authors list ( 4985: 4486: 4463:11245.1/4242e2e0-3beb-40a0-a6cb-d8947a13efb4 4368: 4143: 3226: 2111: 2017: 2015: 1545: 1487: 1306:has also used this approach for manging its 242:Typically, the suggestions refer to various 5229:Introduction to natural language processing 5034: 4362: 4128: 3871: 3617:"The BellKor solution to the Netflix Prize" 3586: 3205: 1692: 5226: 4945:User Modeling and User-Adapted Interaction 4505: 4378:User Modeling and User-Adapted Interaction 4259:User Modeling and User-Adapted Interaction 4213: 3743: 3491: 2601:International Journal on Digital Libraries 1945: 1739: 1737: 1735: 1733: 1731: 1557: 1446: 1444: 1442: 1440: 556: 432:matrix factorization (recommender systems) 407: 198: 184: 5211: 5201: 5160: 5131: 4688: 4678: 4612: 4559: 4461: 4406: 4389: 4332: 4297: 4183: 4008: 3889: 3835: 3738:Temporal diversity in recommender systems 3561: 3435: 3380: 3332: 3240: 3211: 3164: 3117: 3089: 3063: 3019: 2928: 2690: 2643: 2444: 2398: 2107: 2105: 2012: 1816:Melville, Prem; Sindhwani, Vikas (2010). 1761: 1652: 1639: 1614: 1536: 1109:Learn how and when to remove this message 5403: 4495:"Developing trust in recommender agents" 4442:Information, Communication & Society 3704: 3650:"Mátrixfaktorizáció one million dollars" 2849: 2627: 2269: 2046: 1957:"Active Learning in Recommender Systems" 1916: 1849: 1792: 1518: 1252: 417: 5363: 4548:ACM Transactions on Information Systems 3615:R. Bell; Y. Koren; C. Volinsky (2007). 3103: 3101: 3028: 2762: 2742:. International J. Man-Machine Studies. 2291:Shardanand, Upendra, and Pattie Maes. " 2053:Andrew I. Schein; Alexandrin Popescul; 1728: 1645:H. Chen, A. G. Ororbia II, C. L. Giles 1437: 865: 680:Some hybridization techniques include: 5619: 5578:"The Million Dollar Programming Prize" 5455: 5333: 5329: 5327: 5251: 5075: 4934: 4932: 4844:, Deep Learning Re-Work SF Summit 2018 4286:IEEE Educational Activities Department 4274:"Privacy risks in recommender systems" 4243: 3930: 3647: 2990:The Knowledge Reengineering Bottleneck 2240:. Syslab Working Paper 179 (1990). " 2102: 5429: 5183: 5113: 4842:Deep Learning for Recommender Systems 4834: 4607:. RecSys '19. ACM. pp. 101–109. 3867: 3865: 3777: 3775: 3079: 3077: 3075: 2740:User Models: Theory, Method, Practice 2737: 2477: 2475: 2473: 2114:"A group-specific recommender system" 1912: 1910: 1870: 1350:ACM Conference on Recommender Systems 1290:Criticism of Google § Algorithms 1227:: information from outer social media 172:ACM Conference on Recommender Systems 5540:Recommender Systems: An Introduction 5512:Bharat Bhasker; K. Srikumar (2010). 3595: 3469:"Multi-Criteria Recommender Systems" 3098: 3053: 3051: 2976:Journal of Medical Internet Research 1743: 1519:Roy, Deepjyoti; Dutta, Mala (2022). 1091:adding citations to reliable sources 1062: 803: 5324: 4929: 3320: 380:at MIT, Will Hill at Bellcore, and 13: 5490: 5146: 3862: 3784:ACM SIGKDD Explorations Newsletter 3772: 3596:Lohr, Steve (September 22, 2009). 3360: 3341: 3072: 2944:Blanda, Stephanie (May 25, 2015). 2470: 1907: 1873:Music Reference Services Quarterly 1856:R. J. Mooney & L. Roy (1999). 1283: 1190: 1184:Forming Predictive Recommendations 1020: 932: 885:The commonly used metrics are the 743:Multi-criteria recommender systems 497:Analyzing item/user viewing times. 14: 5653: 5606:, and Ramadass Nagarajan. (2002) 5515:Recommender Systems in E-Commerce 5045:10.1109/ICACCCN51052.2020.9362962 4250:Pu, P.; Chen, L.; Hu, R. (2012). 3648:Bodoky, Thomas (August 6, 2009). 3048: 2852:Recommender Systems: The Textbook 1603:Multimedia Tools and Applications 721:Session-based recommender systems 658:Hybrid recommendations approaches 592:A model of the user's preference. 90:Item-item collaborative filtering 5476: 5449: 5423: 5404:Shanklin, Will (June 17, 2024). 5397: 5357: 5298: 5272: 5245: 5220: 5177: 5140: 5069: 5028: 4979: 4888: 4847: 4789: 4752: 4707: 4662: 4592: 4535: 3004:"The Netflix Recommender System" 2270:Karlgren, Jussi (October 2017). 1825:Encyclopedia of Machine Learning 1067: 5227:Eisenstein, J. (October 2019). 5120:Complex and Intelligent Systems 4429: 3991: 3924: 3810: 3758: 3730: 3678: 3660: 3641: 3608: 3541: 3460: 3415: 3375:. KDD '19. pp. 2810–2818. 3275: 3220: 3199: 3144: 2995: 2982: 2963: 2937: 2904: 2868: 2775: 2746: 2731: 2715: 2670: 2652: 2588: 2538: 2424: 2331: 2318: 2298: 2285: 2263: 2260:." SICS Research Report (1994). 2243: 2231:An Algebra for Recommendations. 2223: 2214: 2205: 2192: 2179: 2166: 2157: 2144: 1985: 1864: 1786: 1143:KNN-based collaborative filters 1078:needs additional citations for 715: 446:as first implemented by Allen. 5565:. University of Chicago Press. 4840:Yves Raimond, Justin Basilico 3550:Journal of Information Science 3036:Hybrid Web Recommender Systems 2340:Artificial Intelligence Review 1827:. Springer. pp. 829–838. 1793:Analysis (December 14, 2011). 1590: 1512: 1481: 880:online evaluations (A/B tests) 847:Internet Movie Database (IMDb) 752:Risk-aware recommender systems 1: 5642:Social information processing 5456:Ovadya, Aviv (May 17, 2022). 5430:Novak, Matt (June 17, 2024). 5334:Ovadya, Aviv (May 17, 2022). 5252:Mirkin, Sima (June 4, 2014). 4454:10.1080/1369118X.2018.1444076 2950:American Mathematical Society 2666:(Report). Microsoft Research. 2130:10.1080/01621459.2016.1219261 1833:10.1007/978-0-387-30164-8_705 1756:(7516). Nature.com: 129–130. 1706:. In William Cheetham (ed.). 1488:Lev Grossman (May 27, 2010). 1431: 1321: 878:are available: user studies, 860: 775:Location based recommendation 648:Multimodal sentiment analysis 634:various techniques including 514:, scalability, and sparsity. 482:or other similar techniques). 402: 376:, and research groups led by 4415:Recommender systems handbook 4369:Konstan JA, Riedl J (2012). 4350:. Springer. pp. 400–404 4167:10.1007/978-1-4899-7637-6_26 4159:Recommender Systems Handbook 3942:10.1007/978-3-319-24592-8_12 3492:Bouneffouf, Djallel (2013), 2930:10.1016/j.knosys.2018.05.001 2040:10.1016/j.cosrev.2016.05.002 2006:10.1016/j.knosys.2011.06.005 1969:10.1007/978-1-4899-7637-6_24 1961:Recommender Systems Handbook 1457:Recommender Systems Handbook 1385:Information filtering system 1243:singular value decomposition 1157:. The ideas are as follows: 851:Video Privacy Protection Act 523:multi-armed bandit algorithm 236:information filtering system 7: 5114:Zhang, Q. (February 2021). 3846:10.1007/978-3-642-38844-6_3 3667:Rise of the Netflix Hackers 2850:Aggarwal, Charu C. (2016). 2057:; David M. Pennock (2002). 1465:10.1007/978-1-0716-2197-4_1 1337: 1247:latent Dirichlet allocation 1232:Natural language processing 1137:natural language processing 882:, and offline evaluations. 294: 121:Collaborative search engine 10: 5658: 5213:10.1109/JPROC.2021.3060483 5133:10.1007/s40747-020-00212-w 5090:10.1007/s10639-019-10063-9 4114:10.1007/s10791-020-09371-3 3975:10.1137/1.9781611974973.44 2772:, Discovery Science, 2007. 2256:February 27, 2021, at the 1744:jobs (September 3, 2014). 1538:10.1186/s40537-022-00592-5 1345:Algorithmic radicalization 1287: 1258:Academic content discovery 807: 772: 769:Mobile recommender systems 619:artificial neural networks 411: 386:ACM Software Systems Award 352: 260:content discovery platform 126:Content discovery platform 5171:10.1109/TKDE.2022.3145690 4957:10.1007/s11257-016-9174-x 4391:10.1007/s11257-011-9112-x 3672:January 24, 2012, at the 2613:10.1007/s00799-015-0156-0 1885:10.1080/10588160902816702 1616:10.1007/s11042-017-5014-1 1197:artificial neural network 1174:for computational details 845:with film ratings on the 728:recurrent neural networks 244:decision-making processes 5458:"Bridging-Based Ranking" 5336:"Bridging-Based Ranking" 5184:Samek, W. (March 2021). 4108:(4). Springer: 387–410. 3572:10.1177/0165551518792213 3424:IEEE Intelligent Systems 1649:, in arXiv preprint 2015 1395:Media monitoring service 1239:latent semantic analysis 1219:User Navigation Patterns 855:Federal Trade Commission 488:implicit data collection 467:Asking a user to search. 442:(k-NN) approach and the 85:Implicit data collection 80:Dimensionality reduction 5190:Proceedings of the IEEE 4996:10.1145/2645710.2645746 4905:10.1145/2532508.2532513 4864:10.1145/2043932.2043958 4810:10.1145/3267471.3267480 4728:10.1145/3383313.3412489 4722:. ACM. pp. 23–32. 4690:10.1145/3383313.3412488 4623:10.1145/3298689.3347058 4278:IEEE Internet Computing 4065:10.1145/3209978.3210014 4019:10.1145/2532508.2532511 3796:10.1145/3137597.3137601 3523:New York City, New York 3391:10.1145/3292500.3330668 3296:10.1145/3219819.3219950 3251:10.1145/3132847.3132926 3175:10.1145/3269206.3271761 3119:10.1145/3394486.3403278 2917:Knowledge-Based Systems 2801:10.1145/3079628.3079681 2564:10.1145/2532508.2532512 2499:10.1145/2532508.2532511 2352:10.1023/A:1022850703159 2028:Computer Science Review 1994:Knowledge-Based Systems 1676:10.1145/1998076.1998121 1574:10.1145/2488388.2488433 1365:Collective intelligence 1360:Collaborative filtering 1148:Collaborative filtering 1122:Artificial intelligence 950:Recommender persistence 891:root mean squared error 664:collaborative filtering 563:content-based filtering 557:Content-based filtering 545:'s recommender system. 490:include the following: 428:collaborative filtering 414:Collaborative filtering 408:Collaborative filtering 305:knowledge-based systems 301:collaborative filtering 131:Decision support system 75:Collaborative filtering 39:Collective intelligence 3712:"Netflix Prize Update" 1710:. pp. 1692–1699. 1415:Preference elicitation 1405:Personalized marketing 1375:Enterprise bookmarking 1225:External Social Trends 1027:reproducibility crisis 831: 479:Rocchio classification 423: 100:Preference elicitation 62:Methods and challenges 5632:Mass media monitoring 5561:Seaver, Nick (2022). 5532:on September 1, 2010. 4804:. ACM. pp. 1–6. 4770:. ACM. Archived from 4284:(6). Piscataway, NJ: 4102:Information Retrieval 3900:10.1145/383952.383992 2455:10.1145/963770.963772 1818:"Recommender Systems" 1390:Information explosion 1263:search tools such as 1253:Specific applications 1178:Identifying Neighbors 823: 784:generality problems. 773:Further information: 640:information retrieval 579:information filtering 575:information retrieval 421: 218:(sometimes replacing 216:recommendation system 5286:on November 17, 2014 5039:. pp. 669–673. 4673:. pp. 240–248. 4231:. pp. 585–592. 3969:. pp. 390–398. 3768:. September 6, 2013. 3718:on November 27, 2011 2886:. Springer. p. 2738:Allen, R.B. (1990). 2433:ACM Trans. Inf. Syst 2409:10.1109/TKDE.2005.99 1213:Time and Seasonality 1172:statistical distance 1168:Statistical Distance 1087:improve this article 895:precision and recall 866:Performance measures 607:Bayesian Classifiers 328:Music Genome Project 234:), is a subclass of 136:Music Genome Project 95:Matrix factorization 5637:Recommender systems 5627:Information systems 5570:Scientific articles 5557:on August 31, 2015. 5345:. pp. 1, 14–28 4501:. pp. 304–305. 4423:2011rsh..book.....R 4308:10.1109/4236.968832 3766:"MovieLens dataset" 3754:. pp. 225–231. 3692:. December 17, 2009 3446:10.1109/mis.2011.33 2701:10.1109/MC.2009.263 1609:(11): 14077–14091. 1525:Journal of Big Data 1410:Personalized search 1400:Pattern recognition 1332:internet television 1316:deliberative groups 1298:. Examples include 1162:Data Representation 1155:K-nearest neighbors 444:Pearson Correlation 222:with terms such as 25:Recommender systems 5483:The New Face of TV 5342:Harvard University 4523:. pp. 395–399 3602:The New York Times 3529:. pp. 899–908 3041:2014-09-12 at the 2837:2015-03-16 at the 2558:. pp. 15–22. 2311:2018-12-21 at the 2249:Karlgren, Jussi. " 2236:2024-05-25 at the 2229:Karlgren, Jussi. " 2124:(519): 1344–1353. 1014:click-through rate 920:click-through rate 887:mean squared error 644:sentiment analysis 440:k-nearest neighbor 424: 262:is an implemented 212:recommender system 157:GroupLens Research 5604:Raymond J. Mooney 5550:978-0-521-49336-9 5525:978-0-07-068067-8 5340:Belfer Center at 5054:978-1-7281-8337-4 5013:978-1-4503-2668-1 4914:978-1-4503-2465-6 4873:978-1-4503-0683-6 4819:978-1-4503-6586-4 4777:on March 25, 2021 4737:978-1-4503-7583-2 4700:978-1-4503-7583-2 4640:978-1-4503-6243-6 4417:. pp. 1–35. 4317:978-1-58113-561-9 4176:978-1-4899-7637-6 4139:. pp. 22–32. 4074:on April 14, 2021 4028:978-1-4503-2465-6 3984:978-1-61197-497-3 3951:978-3-319-24591-1 3909:978-1-58113-331-8 3855:978-3-642-38843-9 3481:on June 30, 2014. 3400:978-1-4503-6201-6 3305:978-1-4503-5552-0 3260:978-1-4503-4918-5 3184:978-1-4503-6014-2 3129:978-1-4503-7998-4 2897:978-3-540-72078-2 2876:Peter Brusilovsky 2861:978-3-319-29657-9 2810:978-1-4503-4635-1 2573:978-1-4503-2465-6 2526:on April 17, 2016 2508:978-1-4503-2465-6 2493:. pp. 7–14. 2377:Adomavicius, G.; 2150:Stack, Charles. " 1978:978-1-4899-7637-6 1842:978-0-387-30164-8 1474:978-1-0716-2196-7 1277:statistical model 1119: 1118: 1111: 981:User demographics 804:The Netflix Prize 344:search algorithms 208: 207: 105:Similarity search 5649: 5599: 5597: 5595: 5586:. Archived from 5558: 5553:. Archived from 5533: 5528:. Archived from 5485: 5480: 5474: 5473: 5471: 5469: 5464:. pp. 21–23 5453: 5447: 5446: 5444: 5442: 5427: 5421: 5420: 5418: 5416: 5401: 5395: 5394: 5388: 5380: 5378: 5376: 5361: 5355: 5354: 5352: 5350: 5331: 5322: 5321: 5319: 5317: 5302: 5296: 5295: 5293: 5291: 5282:. Archived from 5276: 5270: 5269: 5267: 5265: 5249: 5243: 5242: 5224: 5218: 5217: 5215: 5205: 5181: 5175: 5174: 5164: 5155:(5): 4425–4445. 5144: 5138: 5137: 5135: 5111: 5102: 5101: 5084:(4): 2635–2664. 5078:Educ Inf Technol 5073: 5067: 5066: 5032: 5026: 5025: 4983: 4977: 4976: 4936: 4927: 4926: 4892: 4886: 4885: 4851: 4845: 4838: 4832: 4831: 4793: 4787: 4786: 4784: 4782: 4776: 4765: 4756: 4750: 4749: 4711: 4705: 4704: 4692: 4682: 4666: 4660: 4659: 4657: 4655: 4616: 4596: 4590: 4589: 4563: 4539: 4533: 4532: 4530: 4528: 4518: 4509: 4503: 4502: 4490: 4484: 4483: 4465: 4433: 4427: 4426: 4410: 4404: 4403: 4393: 4375: 4366: 4360: 4359: 4357: 4355: 4345: 4336: 4330: 4329: 4301: 4269: 4263: 4262: 4256: 4247: 4241: 4240: 4226: 4217: 4211: 4210: 4208: 4206: 4196: 4187: 4181: 4180: 4150: 4141: 4140: 4132: 4126: 4125: 4099: 4090: 4084: 4083: 4081: 4079: 4073: 4067:. Archived from 4058: 4047: 4041: 4040: 4012: 3995: 3989: 3988: 3962: 3956: 3955: 3928: 3922: 3921: 3893: 3869: 3860: 3859: 3839: 3825: 3814: 3808: 3807: 3779: 3770: 3769: 3762: 3756: 3755: 3747: 3741: 3734: 3728: 3727: 3725: 3723: 3708: 3702: 3701: 3699: 3697: 3682: 3676: 3664: 3658: 3657: 3645: 3639: 3638: 3636: 3634: 3629:on March 4, 2012 3628: 3622:. Archived from 3621: 3612: 3606: 3605: 3593: 3584: 3583: 3565: 3545: 3539: 3538: 3536: 3534: 3520: 3509: 3500: 3499: 3489: 3483: 3482: 3480: 3474:. Archived from 3473: 3464: 3458: 3457: 3439: 3419: 3413: 3412: 3384: 3364: 3358: 3357: 3345: 3339: 3338: 3336: 3324: 3318: 3317: 3279: 3273: 3272: 3244: 3224: 3218: 3217: 3215: 3203: 3197: 3196: 3168: 3148: 3142: 3141: 3121: 3105: 3096: 3095: 3093: 3081: 3070: 3069: 3067: 3055: 3046: 3032: 3026: 3025: 3023: 2999: 2993: 2988:Rinke Hoekstra, 2986: 2980: 2979:, 21 (5): e12957 2967: 2961: 2960: 2958: 2956: 2941: 2935: 2934: 2932: 2908: 2902: 2901: 2885: 2882:The Adaptive Web 2872: 2866: 2865: 2847: 2841: 2829: 2823: 2822: 2794: 2779: 2773: 2766: 2760: 2758: 2750: 2744: 2743: 2735: 2729: 2727: 2719: 2713: 2712: 2694: 2674: 2668: 2667: 2665: 2656: 2650: 2649: 2647: 2631: 2625: 2624: 2592: 2586: 2585: 2551: 2542: 2536: 2535: 2533: 2531: 2525: 2519:. Archived from 2488: 2479: 2468: 2466: 2448: 2428: 2422: 2420: 2402: 2374: 2365: 2363: 2335: 2329: 2322: 2316: 2302: 2296: 2289: 2283: 2282: 2280: 2278: 2267: 2261: 2247: 2241: 2227: 2221: 2218: 2212: 2209: 2203: 2196: 2190: 2183: 2177: 2170: 2164: 2161: 2155: 2148: 2142: 2141: 2109: 2100: 2099: 2097: 2095: 2050: 2044: 2043: 2019: 2010: 2009: 2000:(8): 1310–1316. 1989: 1983: 1982: 1952: 1943: 1942: 1914: 1905: 1904: 1868: 1862: 1861: 1853: 1847: 1846: 1822: 1813: 1807: 1806: 1804: 1802: 1790: 1784: 1783: 1765: 1741: 1726: 1725: 1705: 1696: 1690: 1689: 1665: 1656: 1650: 1643: 1637: 1636: 1618: 1594: 1588: 1587: 1561: 1555: 1552: 1543: 1542: 1540: 1516: 1510: 1509: 1507: 1505: 1496:. Archived from 1485: 1479: 1478: 1448: 1129:machine learning 1114: 1107: 1103: 1100: 1094: 1071: 1063: 1036:RecSys Challenge 839:RecSys community 611:cluster analysis 288:academic journal 200: 193: 186: 21: 20: 5657: 5656: 5652: 5651: 5650: 5648: 5647: 5646: 5617: 5616: 5602:Prem Melville, 5593: 5591: 5590:on May 11, 2009 5551: 5526: 5493: 5491:Further reading 5488: 5481: 5477: 5467: 5465: 5454: 5450: 5440: 5438: 5428: 5424: 5414: 5412: 5402: 5398: 5382: 5381: 5374: 5372: 5362: 5358: 5348: 5346: 5332: 5325: 5315: 5313: 5303: 5299: 5289: 5287: 5278: 5277: 5273: 5263: 5261: 5250: 5246: 5239: 5225: 5221: 5182: 5178: 5145: 5141: 5112: 5105: 5074: 5070: 5055: 5033: 5029: 5014: 4984: 4980: 4937: 4930: 4915: 4893: 4889: 4874: 4852: 4848: 4839: 4835: 4820: 4794: 4790: 4780: 4778: 4774: 4763: 4757: 4753: 4738: 4712: 4708: 4701: 4667: 4663: 4653: 4651: 4641: 4597: 4593: 4570:10.1145/3434185 4540: 4536: 4526: 4524: 4516: 4510: 4506: 4491: 4487: 4434: 4430: 4411: 4407: 4373: 4367: 4363: 4353: 4351: 4343: 4337: 4333: 4318: 4270: 4266: 4254: 4248: 4244: 4224: 4218: 4214: 4204: 4202: 4194: 4188: 4184: 4177: 4151: 4144: 4133: 4129: 4097: 4091: 4087: 4077: 4075: 4071: 4056: 4048: 4044: 4029: 4010:10.1.1.1031.973 3996: 3992: 3985: 3963: 3959: 3952: 3929: 3925: 3910: 3891:10.1.1.165.5800 3870: 3863: 3856: 3815: 3811: 3780: 3773: 3764: 3763: 3759: 3748: 3744: 3735: 3731: 3721: 3719: 3710: 3709: 3705: 3695: 3693: 3684: 3683: 3679: 3674:Wayback Machine 3665: 3661: 3646: 3642: 3632: 3630: 3626: 3619: 3613: 3609: 3594: 3587: 3546: 3542: 3532: 3530: 3518: 3510: 3503: 3490: 3486: 3478: 3471: 3465: 3461: 3437:10.1.1.476.6726 3420: 3416: 3401: 3365: 3361: 3346: 3342: 3325: 3321: 3306: 3280: 3276: 3261: 3225: 3221: 3204: 3200: 3185: 3149: 3145: 3130: 3106: 3099: 3082: 3073: 3056: 3049: 3043:Wayback Machine 3033: 3029: 3021:10.1145/2843948 3000: 2996: 2987: 2983: 2968: 2964: 2954: 2952: 2942: 2938: 2909: 2905: 2898: 2873: 2869: 2862: 2848: 2844: 2839:Wayback Machine 2830: 2826: 2811: 2792: 2780: 2776: 2767: 2763: 2751: 2747: 2736: 2732: 2720: 2716: 2692:10.1.1.147.8295 2675: 2671: 2663: 2657: 2653: 2632: 2628: 2593: 2589: 2574: 2549: 2543: 2539: 2529: 2527: 2523: 2509: 2486: 2480: 2471: 2429: 2425: 2400:10.1.1.107.2790 2375: 2368: 2336: 2332: 2323: 2319: 2313:Wayback Machine 2303: 2299: 2290: 2286: 2276: 2274: 2268: 2264: 2258:Wayback Machine 2248: 2244: 2238:Wayback Machine 2228: 2224: 2219: 2215: 2210: 2206: 2197: 2193: 2184: 2180: 2171: 2167: 2162: 2158: 2149: 2145: 2110: 2103: 2093: 2091: 2089: 2051: 2047: 2020: 2013: 1990: 1986: 1979: 1953: 1946: 1931:10.1145/3285954 1915: 1908: 1869: 1865: 1854: 1850: 1843: 1820: 1814: 1810: 1800: 1798: 1791: 1787: 1763:10.1038/513129a 1742: 1729: 1718: 1703: 1697: 1693: 1686: 1663: 1657: 1653: 1644: 1640: 1595: 1591: 1584: 1562: 1558: 1553: 1546: 1517: 1513: 1503: 1501: 1500:on May 30, 2010 1486: 1482: 1475: 1449: 1438: 1434: 1429: 1340: 1324: 1308:community notes 1292: 1286: 1284:Decision-making 1260: 1255: 1234: 1193: 1191:Neural networks 1145: 1115: 1104: 1098: 1095: 1084: 1072: 1061: 1023: 1021:Reproducibility 935: 933:Beyond accuracy 916:conversion rate 868: 863: 835:Gravity R&D 812: 806: 777: 771: 754: 745: 736: 723: 718: 668:knowledge-based 660: 559: 550:social networks 455:data collection 416: 410: 405: 355: 297: 266:recommendation 204: 113:Implementations 17: 12: 11: 5: 5655: 5645: 5644: 5639: 5634: 5629: 5615: 5614: 5600: 5572: 5571: 5567: 5566: 5559: 5549: 5534: 5524: 5509: 5498: 5497: 5492: 5489: 5487: 5486: 5475: 5448: 5422: 5396: 5356: 5323: 5297: 5271: 5244: 5237: 5219: 5196:(3): 247–278. 5176: 5139: 5103: 5068: 5053: 5027: 5012: 4978: 4928: 4913: 4887: 4872: 4846: 4833: 4818: 4788: 4751: 4736: 4706: 4699: 4661: 4639: 4591: 4534: 4504: 4485: 4448:(7): 959–977. 4428: 4405: 4361: 4331: 4316: 4264: 4242: 4212: 4182: 4175: 4142: 4127: 4085: 4042: 4027: 3990: 3983: 3957: 3950: 3923: 3908: 3861: 3854: 3809: 3771: 3757: 3742: 3729: 3703: 3677: 3659: 3640: 3607: 3585: 3556:(3): 387–397. 3540: 3501: 3484: 3459: 3414: 3399: 3359: 3340: 3319: 3304: 3274: 3259: 3219: 3198: 3183: 3143: 3128: 3097: 3071: 3047: 3027: 2994: 2981: 2962: 2936: 2903: 2896: 2867: 2860: 2842: 2824: 2809: 2774: 2761: 2745: 2730: 2714: 2669: 2651: 2626: 2607:(4): 305–338. 2587: 2572: 2537: 2507: 2469: 2446:10.1.1.78.8384 2423: 2393:(6): 734–749. 2366: 2346:(4): 285–330. 2330: 2317: 2297: 2284: 2262: 2242: 2222: 2213: 2204: 2191: 2178: 2165: 2156: 2143: 2101: 2087: 2045: 2011: 1984: 1977: 1944: 1906: 1879:(1–2): 23–24. 1863: 1848: 1841: 1808: 1785: 1727: 1716: 1691: 1684: 1651: 1638: 1589: 1582: 1556: 1544: 1511: 1480: 1473: 1435: 1433: 1430: 1428: 1427: 1422: 1420:Product finder 1417: 1412: 1407: 1402: 1397: 1392: 1387: 1382: 1377: 1372: 1367: 1362: 1357: 1352: 1347: 1341: 1339: 1336: 1323: 1320: 1285: 1282: 1265:Google Scholar 1259: 1256: 1254: 1251: 1233: 1230: 1229: 1228: 1222: 1216: 1192: 1189: 1188: 1187: 1181: 1175: 1165: 1144: 1141: 1117: 1116: 1075: 1073: 1066: 1060: 1057: 1022: 1019: 1018: 1017: 1007: 1001: 990: 984: 977: 976: 953: 946: 945: 934: 931: 867: 864: 862: 859: 808:Main article: 805: 802: 770: 767: 763:bandit problem 753: 750: 744: 741: 735: 732: 722: 719: 717: 714: 713: 712: 706: 700: 694: 688: 659: 656: 615:decision trees 597: 596: 593: 558: 555: 539: 538: 532: 526: 508: 507: 504: 501: 498: 495: 484: 483: 474: 471: 468: 465: 412:Main article: 409: 406: 404: 401: 370:Jussi Karlgren 366:Jussi Karlgren 354: 351: 332: 331: 324: 296: 293: 280:mobile devices 206: 205: 203: 202: 195: 188: 180: 177: 176: 175: 174: 169: 164: 159: 151: 150: 146: 145: 144: 143: 141:Product finder 138: 133: 128: 123: 115: 114: 110: 109: 108: 107: 102: 97: 92: 87: 82: 77: 72: 64: 63: 59: 58: 57: 56: 51: 46: 41: 33: 32: 28: 27: 15: 9: 6: 4: 3: 2: 5654: 5643: 5640: 5638: 5635: 5633: 5630: 5628: 5625: 5624: 5622: 5612: 5609: 5605: 5601: 5589: 5585: 5584: 5583:IEEE Spectrum 5579: 5574: 5573: 5569: 5568: 5564: 5560: 5556: 5552: 5546: 5542: 5541: 5535: 5531: 5527: 5521: 5517: 5516: 5510: 5508: 5507:9781617292705 5504: 5500: 5499: 5495: 5494: 5484: 5479: 5463: 5459: 5452: 5437: 5433: 5426: 5411: 5407: 5400: 5392: 5386: 5371: 5367: 5360: 5344: 5343: 5337: 5330: 5328: 5312: 5308: 5301: 5285: 5281: 5275: 5259: 5255: 5248: 5240: 5238:9780262042840 5234: 5231:. MIT press. 5230: 5223: 5214: 5209: 5204: 5199: 5195: 5191: 5187: 5180: 5172: 5168: 5163: 5158: 5154: 5150: 5143: 5134: 5129: 5125: 5121: 5117: 5110: 5108: 5099: 5095: 5091: 5087: 5083: 5079: 5072: 5064: 5060: 5056: 5050: 5046: 5042: 5038: 5031: 5023: 5019: 5015: 5009: 5005: 5001: 4997: 4993: 4989: 4982: 4974: 4970: 4966: 4962: 4958: 4954: 4951:(1): 69–101. 4950: 4946: 4942: 4935: 4933: 4924: 4920: 4916: 4910: 4906: 4902: 4898: 4891: 4883: 4879: 4875: 4869: 4865: 4861: 4857: 4850: 4843: 4837: 4829: 4825: 4821: 4815: 4811: 4807: 4803: 4799: 4792: 4773: 4769: 4762: 4755: 4747: 4743: 4739: 4733: 4729: 4725: 4721: 4717: 4710: 4702: 4696: 4691: 4686: 4681: 4676: 4672: 4665: 4650: 4646: 4642: 4636: 4632: 4631:11311/1108996 4628: 4624: 4620: 4615: 4610: 4606: 4602: 4595: 4587: 4583: 4579: 4578:11311/1164333 4575: 4571: 4567: 4562: 4557: 4553: 4549: 4545: 4538: 4522: 4515: 4508: 4500: 4496: 4489: 4481: 4477: 4473: 4469: 4464: 4459: 4455: 4451: 4447: 4443: 4439: 4432: 4424: 4420: 4416: 4409: 4401: 4397: 4392: 4387: 4384:(1–2): 1–23. 4383: 4379: 4372: 4365: 4349: 4342: 4335: 4327: 4323: 4319: 4313: 4309: 4305: 4300: 4299:10.1.1.2.2932 4295: 4291: 4287: 4283: 4279: 4275: 4268: 4260: 4253: 4246: 4238: 4234: 4230: 4223: 4216: 4200: 4193: 4186: 4178: 4172: 4168: 4164: 4160: 4156: 4149: 4147: 4138: 4131: 4123: 4119: 4115: 4111: 4107: 4103: 4096: 4089: 4070: 4066: 4062: 4055: 4054: 4046: 4038: 4034: 4030: 4024: 4020: 4016: 4011: 4006: 4002: 3994: 3986: 3980: 3976: 3972: 3968: 3961: 3953: 3947: 3943: 3939: 3935: 3927: 3919: 3915: 3911: 3905: 3901: 3897: 3892: 3887: 3883: 3879: 3875: 3868: 3866: 3857: 3851: 3847: 3843: 3838: 3837:10.1.1.465.96 3833: 3829: 3824: 3823: 3813: 3805: 3801: 3797: 3793: 3789: 3785: 3778: 3776: 3767: 3761: 3753: 3746: 3739: 3733: 3717: 3713: 3707: 3691: 3687: 3681: 3675: 3671: 3668: 3663: 3655: 3651: 3644: 3625: 3618: 3611: 3603: 3599: 3592: 3590: 3581: 3577: 3573: 3569: 3564: 3559: 3555: 3551: 3544: 3528: 3524: 3517: 3516: 3508: 3506: 3497: 3496: 3488: 3477: 3470: 3463: 3455: 3451: 3447: 3443: 3438: 3433: 3429: 3425: 3418: 3410: 3406: 3402: 3396: 3392: 3388: 3383: 3378: 3374: 3370: 3363: 3355: 3351: 3344: 3335: 3330: 3323: 3315: 3311: 3307: 3301: 3297: 3293: 3289: 3285: 3278: 3270: 3266: 3262: 3256: 3252: 3248: 3243: 3238: 3234: 3230: 3223: 3214: 3209: 3202: 3194: 3190: 3186: 3180: 3176: 3172: 3167: 3162: 3158: 3154: 3147: 3139: 3135: 3131: 3125: 3120: 3115: 3111: 3104: 3102: 3092: 3087: 3080: 3078: 3076: 3066: 3061: 3054: 3052: 3044: 3040: 3037: 3034:Robin Burke, 3031: 3022: 3017: 3013: 3009: 3005: 2998: 2991: 2985: 2978: 2977: 2972: 2966: 2951: 2947: 2940: 2931: 2926: 2922: 2918: 2914: 2907: 2899: 2893: 2889: 2884: 2883: 2877: 2871: 2863: 2857: 2853: 2846: 2840: 2836: 2833: 2828: 2820: 2816: 2812: 2806: 2802: 2798: 2791: 2790: 2785: 2778: 2771: 2765: 2756: 2749: 2741: 2734: 2725: 2718: 2710: 2706: 2702: 2698: 2693: 2688: 2684: 2680: 2673: 2662: 2655: 2646: 2641: 2637: 2630: 2622: 2618: 2614: 2610: 2606: 2602: 2598: 2591: 2583: 2579: 2575: 2569: 2565: 2561: 2557: 2556: 2548: 2541: 2522: 2518: 2514: 2510: 2504: 2500: 2496: 2492: 2485: 2478: 2476: 2474: 2464: 2460: 2456: 2452: 2447: 2442: 2438: 2434: 2427: 2418: 2414: 2410: 2406: 2401: 2396: 2392: 2388: 2384: 2381:(June 2005). 2380: 2373: 2371: 2361: 2357: 2353: 2349: 2345: 2341: 2334: 2327: 2321: 2314: 2310: 2307: 2301: 2294: 2288: 2273: 2266: 2259: 2255: 2252: 2246: 2239: 2235: 2232: 2226: 2217: 2208: 2201: 2195: 2188: 2182: 2175: 2169: 2160: 2153: 2147: 2139: 2135: 2131: 2127: 2123: 2119: 2115: 2108: 2106: 2090: 2088:1-58113-561-0 2084: 2080: 2076: 2072: 2069: 2066: 2062: 2061: 2056: 2055:Lyle H. Ungar 2049: 2041: 2037: 2033: 2029: 2025: 2018: 2016: 2007: 2003: 1999: 1995: 1988: 1980: 1974: 1970: 1966: 1962: 1958: 1951: 1949: 1940: 1936: 1932: 1928: 1924: 1920: 1913: 1911: 1902: 1898: 1894: 1890: 1886: 1882: 1878: 1874: 1867: 1859: 1852: 1844: 1838: 1834: 1830: 1826: 1819: 1812: 1796: 1789: 1781: 1777: 1773: 1769: 1764: 1759: 1755: 1751: 1747: 1740: 1738: 1736: 1734: 1732: 1723: 1719: 1717:9781577353232 1713: 1709: 1702: 1695: 1687: 1685:9781450307444 1681: 1677: 1673: 1669: 1662: 1655: 1648: 1642: 1634: 1630: 1626: 1622: 1617: 1612: 1608: 1604: 1600: 1593: 1585: 1583:9781450320351 1579: 1575: 1571: 1567: 1560: 1551: 1549: 1539: 1534: 1530: 1526: 1522: 1515: 1499: 1495: 1491: 1484: 1476: 1470: 1466: 1462: 1458: 1454: 1447: 1445: 1443: 1441: 1436: 1426: 1423: 1421: 1418: 1416: 1413: 1411: 1408: 1406: 1403: 1401: 1398: 1396: 1393: 1391: 1388: 1386: 1383: 1381: 1380:Filter bubble 1378: 1376: 1373: 1371: 1368: 1366: 1363: 1361: 1358: 1356: 1353: 1351: 1348: 1346: 1343: 1342: 1335: 1333: 1329: 1319: 1317: 1313: 1309: 1305: 1301: 1297: 1291: 1281: 1278: 1273: 1270: 1266: 1250: 1248: 1244: 1240: 1226: 1223: 1220: 1217: 1214: 1211: 1210: 1209: 1205: 1202: 1198: 1185: 1182: 1179: 1176: 1173: 1169: 1166: 1163: 1160: 1159: 1158: 1156: 1151: 1149: 1140: 1138: 1134: 1133:deep learning 1130: 1125: 1123: 1113: 1110: 1102: 1092: 1088: 1082: 1081: 1076:This section 1074: 1070: 1065: 1064: 1056: 1053: 1049: 1045: 1041: 1037: 1033: 1028: 1015: 1011: 1008: 1005: 1002: 998: 994: 991: 988: 985: 982: 979: 978: 973: 972:Netflix Prize 969: 965: 961: 960:user profiles 957: 954: 951: 948: 947: 943: 940: 939: 938: 930: 926: 923: 921: 917: 911: 907: 903: 900: 896: 892: 888: 883: 881: 877: 873: 872:effectiveness 858: 856: 852: 848: 842: 840: 836: 830: 828: 822: 819: 817: 816:Netflix Prize 811: 810:Netflix Prize 801: 798: 794: 789: 785: 782: 776: 766: 764: 760: 749: 740: 731: 729: 710: 707: 704: 701: 698: 695: 692: 689: 686: 683: 682: 681: 678: 675: 671: 669: 665: 655: 653: 652:deep learning 649: 645: 641: 637: 632: 626: 622: 620: 616: 612: 608: 603: 594: 591: 590: 589: 587: 582: 580: 576: 571: 566: 564: 554: 551: 546: 544: 536: 533: 530: 527: 524: 520: 517: 516: 515: 513: 505: 502: 499: 496: 493: 492: 491: 489: 481: 480: 475: 472: 469: 466: 463: 462: 461: 458: 456: 452: 447: 445: 441: 435: 433: 429: 420: 415: 400: 398: 394: 389: 387: 383: 379: 375: 371: 367: 362: 359: 350: 347: 345: 340: 337: 329: 325: 321: 320: 319: 317: 316:Pandora Radio 313: 308: 306: 302: 292: 289: 285: 284:set-top boxes 281: 277: 273: 269: 265: 261: 256: 254: 253:online dating 249: 245: 240: 237: 233: 229: 225: 221: 217: 213: 201: 196: 194: 189: 187: 182: 181: 179: 178: 173: 170: 168: 167:Netflix Prize 165: 163: 160: 158: 155: 154: 153: 152: 148: 147: 142: 139: 137: 134: 132: 129: 127: 124: 122: 119: 118: 117: 116: 112: 111: 106: 103: 101: 98: 96: 93: 91: 88: 86: 83: 81: 78: 76: 73: 71: 68: 67: 66: 65: 61: 60: 55: 52: 50: 47: 45: 42: 40: 37: 36: 35: 34: 30: 29: 26: 23: 22: 19: 5610: 5594:December 10, 5592:. Retrieved 5588:the original 5581: 5562: 5555:the original 5539: 5530:the original 5514: 5478: 5466:. Retrieved 5461: 5451: 5439:. Retrieved 5435: 5425: 5413:. Retrieved 5409: 5399: 5373:. Retrieved 5369: 5359: 5347:. Retrieved 5339: 5314:. Retrieved 5310: 5300: 5288:. Retrieved 5284:the original 5274: 5264:December 31, 5262:. Retrieved 5257: 5247: 5228: 5222: 5193: 5189: 5179: 5152: 5148: 5142: 5123: 5119: 5081: 5077: 5071: 5036: 5030: 5004:10486/665450 4987: 4981: 4948: 4944: 4896: 4890: 4855: 4849: 4836: 4801: 4791: 4779:. Retrieved 4772:the original 4767: 4754: 4719: 4709: 4670: 4664: 4652:. Retrieved 4604: 4594: 4551: 4547: 4537: 4525:. Retrieved 4520: 4507: 4498: 4488: 4445: 4441: 4431: 4414: 4408: 4381: 4377: 4364: 4352:. Retrieved 4347: 4334: 4281: 4277: 4267: 4258: 4245: 4228: 4215: 4203:. Retrieved 4198: 4185: 4158: 4136: 4130: 4105: 4101: 4088: 4076:. Retrieved 4069:the original 4052: 4045: 4000: 3993: 3966: 3960: 3933: 3926: 3877: 3821: 3812: 3787: 3783: 3760: 3751: 3745: 3732: 3722:December 14, 3720:. Retrieved 3716:the original 3706: 3694:. Retrieved 3689: 3680: 3662: 3653: 3643: 3631:. Retrieved 3624:the original 3610: 3601: 3553: 3549: 3543: 3533:November 17, 3531:. Retrieved 3514: 3494: 3487: 3476:the original 3462: 3430:(2): 64–76. 3427: 3423: 3417: 3372: 3362: 3356:: 2592–2599. 3353: 3343: 3322: 3287: 3277: 3232: 3222: 3201: 3156: 3146: 3109: 3030: 3011: 3007: 2997: 2984: 2974: 2965: 2953:. Retrieved 2949: 2939: 2920: 2916: 2906: 2881: 2870: 2854:. Springer. 2851: 2845: 2827: 2788: 2777: 2764: 2754: 2748: 2739: 2733: 2717: 2685:(8): 30–37. 2682: 2678: 2672: 2654: 2635: 2629: 2604: 2600: 2590: 2554: 2540: 2528:. Retrieved 2521:the original 2490: 2436: 2432: 2426: 2390: 2386: 2379:Tuzhilin, A. 2343: 2339: 2333: 2320: 2300: 2287: 2275:. Retrieved 2265: 2245: 2225: 2216: 2207: 2194: 2181: 2168: 2159: 2146: 2121: 2117: 2092:. Retrieved 2059: 2048: 2031: 2027: 1997: 1993: 1987: 1960: 1922: 1918: 1876: 1872: 1866: 1857: 1851: 1824: 1811: 1801:December 31, 1799:. Retrieved 1788: 1753: 1749: 1707: 1694: 1667: 1654: 1641: 1606: 1602: 1592: 1565: 1559: 1528: 1524: 1514: 1502:. Retrieved 1498:the original 1493: 1483: 1456: 1370:Configurator 1325: 1293: 1274: 1261: 1235: 1224: 1218: 1212: 1206: 1194: 1183: 1177: 1167: 1161: 1152: 1146: 1126: 1120: 1105: 1099:October 2023 1096: 1085:Please help 1080:verification 1077: 1024: 1009: 1003: 992: 986: 980: 964:data privacy 955: 949: 941: 936: 927: 924: 912: 908: 904: 884: 869: 843: 832: 826: 824: 820: 813: 790: 786: 778: 758: 755: 746: 737: 724: 716:Technologies 708: 702: 696: 690: 684: 679: 677:filtering). 672: 670:approaches. 661: 627: 623: 598: 586:user profile 584:To create a 583: 570:user profile 567: 562: 560: 547: 540: 534: 528: 518: 509: 486:Examples of 485: 477: 459: 448: 436: 425: 390: 382:Paul Resnick 363: 356: 348: 341: 333: 309: 298: 259: 257: 241: 231: 227: 223: 219: 215: 211: 209: 49:Star ratings 24: 18: 5290:December 8, 5126:: 439–457. 4654:October 16, 4554:(2): 1–49. 4527:December 2, 4354:November 1, 4205:November 1, 3014:(4): 1–19. 2955:October 31, 2530:October 22, 2439:(1): 5–53. 2277:October 27, 2094:February 2, 2077:. pp. 1425:Rating site 997:Serendipity 993:Serendipity 906:imprecise. 876:evaluations 781:smartphones 636:text mining 529:Scalability 393:Adomavicius 378:Pattie Maes 358:Elaine Rich 5621:Categories 5311:Nieman Lab 5203:2003.07631 5162:2104.13030 4680:2005.09683 4614:1907.06902 4561:1911.07698 3563:1805.02276 3382:1902.05570 3334:2006.05779 3242:1711.04725 3213:1808.09781 3166:1706.03847 3091:1812.02353 3065:1511.06939 1432:References 1355:Cold start 1322:Television 1296:polarizing 1288:See also: 987:Robustness 861:Evaluation 709:Meta-level 646:(see also 581:research. 543:Amazon.com 519:Cold start 512:cold start 403:Approaches 372:, then at 336:cold start 70:Cold start 5098:254475908 5063:232150789 4965:0924-1868 4746:221785064 4649:196831663 4586:208138060 4480:149344712 4472:1369-118X 4294:CiteSeerX 4122:213169978 4005:CiteSeerX 3886:CiteSeerX 3832:CiteSeerX 3790:: 37–45. 3633:April 30, 3432:CiteSeerX 3138:221191348 2687:CiteSeerX 2645:1301.7363 2621:207035184 2463:207731647 2441:CiteSeerX 2417:206742345 2395:CiteSeerX 2138:125187672 2034:: 29–50. 1901:161141937 1893:1058-8167 1625:1573-7721 1328:broadband 1201:black-box 1010:Labelling 968:profiling 942:Diversity 691:Switching 453:forms of 232:algorithm 162:MovieLens 54:Long tail 44:Relevance 5468:July 17, 5441:July 17, 5415:July 17, 5410:Engadget 5385:cite web 5375:July 17, 5349:July 17, 5316:July 17, 5022:15665277 4828:52942462 4781:April 3, 4078:March 5, 3918:18903114 3804:10651930 3670:Archived 3580:19209845 3454:16752808 3409:62903207 3314:50775765 3269:21066930 3039:Archived 2878:(2007). 2835:Archived 2709:58370896 2679:Computer 2360:16544257 2309:Archived 2254:Archived 2234:Archived 1939:59337456 1925:: 1–22. 1772:25186906 1722:ACM Copy 1633:36511631 1338:See also 1310:, which 1052:Bellogín 1040:Ekstrand 685:Weighted 631:metadata 535:Sparsity 451:implicit 295:Overview 276:websites 272:metadata 268:platform 264:software 248:playlist 224:platform 149:Research 31:Concepts 5543:. CUP. 5518:. CUP. 5436:Gizmodo 5370:Poynter 4882:2215419 4419:Bibcode 4400:8996665 4326:1977107 4261:: 1–39. 4237:8307833 4037:8202591 3882:225–231 3696:June 1, 3284:"STAMP" 3193:1159769 2923:: 1–9. 2582:4411601 2517:8202591 2079:253–260 1797:. WIRED 1780:4460749 1504:June 1, 1312:YouTube 1304:Twitter 1245:(SVD), 1241:(LSA), 1044:Konstan 956:Privacy 703:Cascade 674:Netflix 353:History 312:Last.fm 214:, or a 5547: 5522: 5505: 5235: 5096: 5061: 5051: 5020: 5010: 4973:388764 4971: 4963: 4923:333956 4921: 4911: 4880: 4870: 4826: 4816: 4744: 4734: 4697: 4647: 4637: 4584: 4478: 4470: 4398: 4324: 4314: 4296: 4235: 4173: 4120: 4035: 4025: 4007: 3981: 3948: 3916: 3906: 3888: 3852: 3834: 3802: 3578: 3452: 3434: 3407: 3397: 3312: 3302: 3267: 3257: 3191: 3181: 3136: 3126: 2894: 2858: 2819:653908 2817: 2807: 2707: 2689: 2619: 2580: 2570: 2515: 2505: 2461: 2443: 2415: 2397: 2358: 2136: 2085: 1975: 1937: 1899: 1891: 1839: 1778: 1770: 1750:Nature 1714: 1682: 1631: 1623: 1580: 1531:(59). 1471: 1269:PubMed 1135:, and 1032:RecSys 650:) and 617:, and 602:tf–idf 228:engine 220:system 5496:Books 5198:arXiv 5157:arXiv 5094:S2CID 5059:S2CID 5018:S2CID 4969:S2CID 4919:S2CID 4878:S2CID 4824:S2CID 4775:(PDF) 4764:(PDF) 4742:S2CID 4675:arXiv 4645:S2CID 4609:arXiv 4582:S2CID 4556:arXiv 4517:(PDF) 4476:S2CID 4396:S2CID 4374:(PDF) 4344:(PDF) 4322:S2CID 4290:54–62 4255:(PDF) 4233:S2CID 4225:(PDF) 4195:(PDF) 4118:S2CID 4098:(PDF) 4072:(PDF) 4057:(PDF) 4033:S2CID 3914:S2CID 3830:–37. 3800:S2CID 3690:WIRED 3654:Index 3627:(PDF) 3620:(PDF) 3576:S2CID 3558:arXiv 3519:(PDF) 3479:(PDF) 3472:(PDF) 3450:S2CID 3405:S2CID 3377:arXiv 3329:arXiv 3310:S2CID 3265:S2CID 3237:arXiv 3208:arXiv 3189:S2CID 3161:arXiv 3134:S2CID 3086:arXiv 3060:arXiv 2815:S2CID 2793:(PDF) 2705:S2CID 2664:(PDF) 2640:arXiv 2617:S2CID 2578:S2CID 2550:(PDF) 2524:(PDF) 2513:S2CID 2487:(PDF) 2459:S2CID 2413:S2CID 2356:S2CID 2134:S2CID 2068:SIGIR 1935:S2CID 1897:S2CID 1821:(PDF) 1776:S2CID 1704:(PDF) 1664:(PDF) 1629:S2CID 1300:Polis 1004:Trust 759:DRARS 697:Mixed 548:Many 230:, or 5596:2018 5545:ISBN 5520:ISBN 5503:ISBN 5470:2024 5443:2024 5417:2024 5391:link 5377:2024 5351:2024 5318:2024 5292:2014 5266:2015 5233:ISBN 5049:ISBN 5008:ISBN 4961:ISSN 4909:ISBN 4868:ISBN 4814:ISBN 4783:2021 4732:ISBN 4695:ISBN 4656:2019 4635:ISBN 4529:2013 4468:ISSN 4356:2013 4312:ISBN 4207:2013 4171:ISBN 4080:2021 4023:ISBN 3979:ISBN 3946:ISBN 3904:ISBN 3850:ISBN 3724:2011 3698:2015 3635:2009 3535:2011 3395:ISBN 3300:ISBN 3255:ISBN 3179:ISBN 3124:ISBN 2957:2016 2892:ISBN 2856:ISBN 2805:ISBN 2568:ISBN 2532:2013 2503:ISBN 2279:2017 2096:2008 2083:ISBN 1973:ISBN 1889:ISSN 1837:ISBN 1803:2015 1768:PMID 1712:ISBN 1680:ISBN 1621:ISSN 1578:ISBN 1506:2015 1494:TIME 1469:ISBN 1050:and 1048:Said 889:and 797:Lyft 795:and 793:Uber 577:and 397:Beel 374:SICS 314:and 282:and 5208:doi 5194:109 5167:doi 5128:doi 5086:doi 5041:doi 5000:hdl 4992:doi 4953:doi 4901:doi 4860:doi 4806:doi 4724:doi 4685:doi 4627:hdl 4619:doi 4574:hdl 4566:doi 4458:hdl 4450:doi 4386:doi 4304:doi 4163:doi 4110:doi 4061:doi 4015:doi 3971:doi 3938:doi 3896:doi 3842:doi 3792:doi 3568:doi 3527:ACM 3442:doi 3387:doi 3292:doi 3247:doi 3171:doi 3114:doi 3016:doi 2973:", 2925:doi 2921:157 2888:325 2797:doi 2697:doi 2609:doi 2560:doi 2495:doi 2451:doi 2405:doi 2348:doi 2126:doi 2122:112 2075:ACM 2065:ACM 2036:doi 2002:doi 1965:doi 1927:doi 1881:doi 1829:doi 1758:doi 1754:513 1672:doi 1611:doi 1570:doi 1533:doi 1461:doi 1267:or 1195:An 1089:by 918:or 899:DCG 897:or 5623:: 5580:. 5460:. 5434:. 5408:. 5387:}} 5383:{{ 5368:. 5338:. 5326:^ 5309:. 5256:. 5206:. 5192:. 5188:. 5165:. 5153:35 5151:. 5122:. 5118:. 5106:^ 5092:. 5082:25 5080:. 5057:. 5047:. 5016:. 5006:. 4998:. 4967:. 4959:. 4949:26 4947:. 4943:. 4931:^ 4917:. 4907:. 4876:. 4866:. 4822:. 4812:. 4800:. 4766:. 4740:. 4730:. 4718:. 4693:. 4683:. 4643:. 4633:. 4625:. 4617:. 4603:. 4580:. 4572:. 4564:. 4552:39 4550:. 4546:. 4497:. 4474:. 4466:. 4456:. 4446:21 4444:. 4440:. 4394:. 4382:22 4380:. 4376:. 4320:. 4310:. 4302:. 4292:. 4288:: 4280:. 4276:. 4257:. 4227:. 4169:. 4145:^ 4116:. 4106:23 4104:. 4100:. 4031:. 4021:. 4013:. 3977:. 3944:. 3912:. 3902:. 3894:. 3884:. 3876:. 3864:^ 3848:. 3840:. 3828:25 3798:. 3788:19 3786:. 3774:^ 3688:. 3652:. 3600:. 3588:^ 3574:. 3566:. 3554:45 3552:. 3525:: 3504:^ 3448:. 3440:. 3428:26 3426:. 3403:. 3393:. 3385:. 3371:. 3352:. 3308:. 3298:. 3286:. 3263:. 3253:. 3245:. 3231:. 3187:. 3177:. 3169:. 3155:. 3132:. 3122:. 3100:^ 3074:^ 3050:^ 3010:. 3006:. 2948:. 2919:. 2915:. 2890:. 2813:. 2803:. 2786:. 2703:. 2695:. 2683:42 2681:. 2615:. 2605:17 2603:. 2599:. 2576:. 2566:. 2552:. 2511:. 2501:. 2489:. 2472:^ 2457:. 2449:. 2437:22 2435:. 2411:. 2403:. 2391:17 2389:. 2385:. 2369:^ 2354:. 2344:19 2342:. 2132:. 2120:. 2116:. 2104:^ 2081:. 2073:: 2032:20 2030:. 2026:. 2014:^ 1998:24 1996:. 1971:. 1947:^ 1933:. 1923:13 1921:. 1909:^ 1895:. 1887:. 1877:12 1875:. 1835:. 1774:. 1766:. 1752:. 1748:. 1730:^ 1720:. 1678:. 1666:. 1627:. 1619:. 1607:77 1605:. 1601:. 1576:. 1547:^ 1527:. 1523:. 1492:. 1467:. 1439:^ 1131:, 1042:, 995:– 922:. 654:. 642:, 638:, 613:, 609:, 457:. 434:. 388:. 318:. 278:, 258:A 226:, 210:A 5598:. 5472:. 5445:. 5419:. 5393:) 5379:. 5353:. 5320:. 5294:. 5268:. 5241:. 5216:. 5210:: 5200:: 5173:. 5169:: 5159:: 5136:. 5130:: 5124:7 5100:. 5088:: 5065:. 5043:: 5024:. 5002:: 4994:: 4975:. 4955:: 4925:. 4903:: 4884:. 4862:: 4830:. 4808:: 4785:. 4748:. 4726:: 4703:. 4687:: 4677:: 4658:. 4629:: 4621:: 4611:: 4588:. 4576:: 4568:: 4558:: 4531:. 4482:. 4460:: 4452:: 4425:. 4421:: 4402:. 4388:: 4358:. 4328:. 4306:: 4282:5 4239:. 4209:. 4179:. 4165:: 4124:. 4112:: 4082:. 4063:: 4039:. 4017:: 3987:. 3973:: 3954:. 3940:: 3920:. 3898:: 3858:. 3844:: 3806:. 3794:: 3726:. 3700:. 3656:. 3637:. 3604:. 3582:. 3570:: 3560:: 3537:. 3456:. 3444:: 3411:. 3389:: 3379:: 3337:. 3331:: 3316:. 3294:: 3271:. 3249:: 3239:: 3216:. 3210:: 3195:. 3173:: 3163:: 3140:. 3116:: 3094:. 3088:: 3068:. 3062:: 3024:. 3018:: 3012:6 2959:. 2933:. 2927:: 2900:. 2864:. 2821:. 2799:: 2759:. 2728:, 2726:. 2711:. 2699:: 2648:. 2642:: 2623:. 2611:: 2584:. 2562:: 2534:. 2497:: 2467:. 2465:. 2453:: 2421:. 2419:. 2407:: 2364:. 2362:. 2350:: 2281:. 2140:. 2128:: 2098:. 2042:. 2038:: 2008:. 2004:: 1981:. 1967:: 1941:. 1929:: 1903:. 1883:: 1845:. 1831:: 1805:. 1782:. 1760:: 1724:. 1688:. 1674:: 1635:. 1613:: 1586:. 1572:: 1541:. 1535:: 1529:9 1508:. 1477:. 1463:: 1112:) 1106:( 1101:) 1097:( 1083:. 525:. 199:e 192:t 185:v

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.