Bamshad Mobasher - Academia.edu (original) (raw)

Uploads

Papers by Bamshad Mobasher

International Joint Conference on Artificial Intelligence, Jul 27, 2015

Recommender systems have been used in many domains to assist users' decision making by provid... more Recommender systems have been used in many domains to assist users' decision making by providing item recommendations and thereby reducing information overload. Context-aware recommender systems go further, incorporating the variability of users' preferences across contexts, and suggesting items that are appropriate in different contexts. In this paper, we present a novel recommendation task, "Context Suggestion", whereby the system recommends contexts in which items may be selected. We introduce the motivations behind the notion of context suggestion and discuss several potential solutions. In particular, we focus specifically on user-oriented context suggestion which involves recommending appropriate contexts based on a user's profile. We propose extensions of well-known context-aware recommendation algorithms such as tensor factorization and deviation-based contextual modeling and adapt them as methods to recommend contexts instead of items. In our empirical evaluation, we compare the proposed solutions to several baseline algorithms using four real-world data sets.

ACM Transactions on Information Systems, Nov 16, 2021

arXiv (Cornell University), May 21, 2020

arXiv (Cornell University), May 3, 2020

arXiv (Cornell University), Jul 10, 2019

It is well known that explicit user ratings in recommender systems are biased towards high rating... more It is well known that explicit user ratings in recommender systems are biased towards high ratings, and that users differ significantly in their usage of the rating scale. Implementers usually compensate for these issues through rating normalization or the inclusion of a user bias term in factorization models. However, these methods adjust only for the central tendency of users' distributions. In this work, we demonstrate that lack of \textit{flatness} in rating distributions is negatively correlated with recommendation performance. We propose a rating transformation model that compensates for skew in the rating distribution as well as its central tendency by converting ratings into percentile values as a pre-processing step before recommendation generation. This transformation flattens the rating distribution, better compensates for differences in rating distributions, and improves recommendation performance. We also show a smoothed version of this transformation designed to yield more intuitive results for users with very narrow rating distributions. A comprehensive set of experiments show improved ranking performance for these percentile transformations with state-of-the-art recommendation algorithms in four real-world data sets.

arXiv (Cornell University), Aug 7, 2021

Conference on Recommender Systems, 2013

ABSTRACT User and item splitting are well-known approaches to context-aware recommendation. To pe... more ABSTRACT User and item splitting are well-known approaches to context-aware recommendation. To perform item splitting, multiple copies of an item are created based on the contexts in which it has been rated. User splitting performs a similar treatment with respect to user-s. The combination of user and item splitting: UI splitting, splits both users and items in the data set to boost context-aware recom-mendations. In this paper, we perform an empirical comparison of these three context-aware splitting approaches (CASA) on multiple data sets, and we also compare them with other popular context-aware collaborative filtering (CACF) algorithms. To evaluate those algorithms, we propose new evaluation metrics specific to contex-tual recommendation. The experiments reveal that CASA typically outperform other popular CACF algorithms, but there is no clear winner among the three splitting approaches. However, we do find some underlying patterns or clues for the application of CASA.

The exponential growth of the Social Web both poses challenges, and presents opportunities for Re... more The exponential growth of the Social Web both poses challenges, and presents opportunities for Recommender System research. The Social Web has turned information consumers into active contributors who generate large volumes of rapidly changing online data. Recommender Systems strive to identify relevant content for users at the right time and in the right context but achieving this goal has become more difficult, in part due to the volume and nature of information contributed through the Social Web. The emergence of the Social Web marked a change in Web users' attitude to online privacy and sharing. Social media systems encourage users to implicitly and explicitly provide large volumes of information which previously they would have been reluctant to share. This information includes personal details such as location, age, and interests, friendship networks, bookmarks and tags, opinion and preferences which can be captured explicitly or more often by monitoring user interaction with the systems (e.g. commenting, friending, rating,tagging etc). These new sources of knowledge can be leveraged by Recommender Systems to improve existing techniques and develop new strategies which focus on social recommendation. In turn recommender technologies can play a huge part in fuelling the success of the Social Web phenomenon by reducing the information overload problem facing social media users. The goal of this one day workshop was to bring together researchers and practitioners to explore, discuss, and understand challenges and new opportunities for Recommender Systems and the Social Web. The workshop consisted both of technical sessions, in which selected participants presented their results or ongoing research, as well as informal breakout sessions on more focused topics. Papers discussing various aspects of recommender system in the Social Web were submitted and selected for presentation and discussion in the workshop in a formal reviewing process. The topics of the submitted papers included, among others, the following main areas: Case studies and novel fielded social recommender applications Economy of community-based systems: Using recommenders to encourage users to contribute and sustain participation Social network and folksonomy development: Recommending friends, tags, bookmarks, blogs, music, communities etc. Recommender system mash-ups, Web 2.0 user interfaces, rich media recommender systems Recommender applications involving users and groups directly in the recommendation process Exploiting folksonomies, social network information, interaction user context and communities or groups for recommendations Trust and reputation aware social recommendations Semantic Web recommender systems, use of ontologies and microformats Empirical evaluation of social recommender techniques, success and failure measures Social recommender systems in the enterprise The list of short papers, the workshop schedule and downloadable versions of the papers can be found at the workshop's homepage at: http://www.dcs.warwick.ac.uk/~ssanand/RSWEb.htm and are also published at: http://ftp.informatik.rwth-aachen.de/Publications/CEUR-WS/

Nowadays, the use of social information is extending to more and more application domains. In the... more Nowadays, the use of social information is extending to more and more application domains. In the field of recommender systems, this information has been exploited in different ways to address some problems, especially associated with collaborative filtering methods, and thus achieve more reliable recommendations. Specifically, social tagging is used in this area mainly to characterize the items that are the subject of the recommendations. In this work, a user-based collaborative filtering approach is presented, where tags processed by word embedding techniques are used to characterize users. User similarities based on both tag embedding and ratings are combined to generate the recommendations. In the study conducted on two popular datasets, the reliability of this approach for rating prediction and top-N recommendations was tested, showing the best performance against the most widely used collaborative filtering methods.

ACM Transactions on The Web, Jun 23, 2017

arXiv (Cornell University), Aug 2, 2019

Lecture Notes in Computer Science, 2018

Suppliers of music streaming services are showing an increasing interest for providing users with... more Suppliers of music streaming services are showing an increasing interest for providing users with reliable personalized recommendations since their practically unlimited offerings make it difficult for users to find the music they like. In this work, we take advantage of social tags that users give to music through streaming platforms for improving recommendations. Most of the works in the literature use the tags in the context of content based methods for finding similarities between songs and artists, but we use them for characterizing users, instead of characterizing music, aiming at improving user-based collaborative filtering algorithms. The expertise level of users is inferred from the frequency analysis of their tags by using TF-IDF (Term Frequency-Inverse Document Frequency), which is an indicator of the quantity and relevance of the tags that users provide to items. User expertise has been studied in the context of recommender systems and other domains, but, as far as we know, it has not been studied in the context of music recommendations.

International Joint Conference on Artificial Intelligence, Jul 27, 2015

Recommender systems have been used in many domains to assist users' decision making by provid... more Recommender systems have been used in many domains to assist users' decision making by providing item recommendations and thereby reducing information overload. Context-aware recommender systems go further, incorporating the variability of users' preferences across contexts, and suggesting items that are appropriate in different contexts. In this paper, we present a novel recommendation task, "Context Suggestion", whereby the system recommends contexts in which items may be selected. We introduce the motivations behind the notion of context suggestion and discuss several potential solutions. In particular, we focus specifically on user-oriented context suggestion which involves recommending appropriate contexts based on a user's profile. We propose extensions of well-known context-aware recommendation algorithms such as tensor factorization and deviation-based contextual modeling and adapt them as methods to recommend contexts instead of items. In our empirical evaluation, we compare the proposed solutions to several baseline algorithms using four real-world data sets.

ACM Transactions on Information Systems, Nov 16, 2021

arXiv (Cornell University), May 21, 2020

arXiv (Cornell University), May 3, 2020

arXiv (Cornell University), Jul 10, 2019

It is well known that explicit user ratings in recommender systems are biased towards high rating... more It is well known that explicit user ratings in recommender systems are biased towards high ratings, and that users differ significantly in their usage of the rating scale. Implementers usually compensate for these issues through rating normalization or the inclusion of a user bias term in factorization models. However, these methods adjust only for the central tendency of users' distributions. In this work, we demonstrate that lack of \textit{flatness} in rating distributions is negatively correlated with recommendation performance. We propose a rating transformation model that compensates for skew in the rating distribution as well as its central tendency by converting ratings into percentile values as a pre-processing step before recommendation generation. This transformation flattens the rating distribution, better compensates for differences in rating distributions, and improves recommendation performance. We also show a smoothed version of this transformation designed to yield more intuitive results for users with very narrow rating distributions. A comprehensive set of experiments show improved ranking performance for these percentile transformations with state-of-the-art recommendation algorithms in four real-world data sets.

arXiv (Cornell University), Aug 7, 2021

Conference on Recommender Systems, 2013

ABSTRACT User and item splitting are well-known approaches to context-aware recommendation. To pe... more ABSTRACT User and item splitting are well-known approaches to context-aware recommendation. To perform item splitting, multiple copies of an item are created based on the contexts in which it has been rated. User splitting performs a similar treatment with respect to user-s. The combination of user and item splitting: UI splitting, splits both users and items in the data set to boost context-aware recom-mendations. In this paper, we perform an empirical comparison of these three context-aware splitting approaches (CASA) on multiple data sets, and we also compare them with other popular context-aware collaborative filtering (CACF) algorithms. To evaluate those algorithms, we propose new evaluation metrics specific to contex-tual recommendation. The experiments reveal that CASA typically outperform other popular CACF algorithms, but there is no clear winner among the three splitting approaches. However, we do find some underlying patterns or clues for the application of CASA.

The exponential growth of the Social Web both poses challenges, and presents opportunities for Re... more The exponential growth of the Social Web both poses challenges, and presents opportunities for Recommender System research. The Social Web has turned information consumers into active contributors who generate large volumes of rapidly changing online data. Recommender Systems strive to identify relevant content for users at the right time and in the right context but achieving this goal has become more difficult, in part due to the volume and nature of information contributed through the Social Web. The emergence of the Social Web marked a change in Web users' attitude to online privacy and sharing. Social media systems encourage users to implicitly and explicitly provide large volumes of information which previously they would have been reluctant to share. This information includes personal details such as location, age, and interests, friendship networks, bookmarks and tags, opinion and preferences which can be captured explicitly or more often by monitoring user interaction with the systems (e.g. commenting, friending, rating,tagging etc). These new sources of knowledge can be leveraged by Recommender Systems to improve existing techniques and develop new strategies which focus on social recommendation. In turn recommender technologies can play a huge part in fuelling the success of the Social Web phenomenon by reducing the information overload problem facing social media users. The goal of this one day workshop was to bring together researchers and practitioners to explore, discuss, and understand challenges and new opportunities for Recommender Systems and the Social Web. The workshop consisted both of technical sessions, in which selected participants presented their results or ongoing research, as well as informal breakout sessions on more focused topics. Papers discussing various aspects of recommender system in the Social Web were submitted and selected for presentation and discussion in the workshop in a formal reviewing process. The topics of the submitted papers included, among others, the following main areas: Case studies and novel fielded social recommender applications Economy of community-based systems: Using recommenders to encourage users to contribute and sustain participation Social network and folksonomy development: Recommending friends, tags, bookmarks, blogs, music, communities etc. Recommender system mash-ups, Web 2.0 user interfaces, rich media recommender systems Recommender applications involving users and groups directly in the recommendation process Exploiting folksonomies, social network information, interaction user context and communities or groups for recommendations Trust and reputation aware social recommendations Semantic Web recommender systems, use of ontologies and microformats Empirical evaluation of social recommender techniques, success and failure measures Social recommender systems in the enterprise The list of short papers, the workshop schedule and downloadable versions of the papers can be found at the workshop's homepage at: http://www.dcs.warwick.ac.uk/~ssanand/RSWEb.htm and are also published at: http://ftp.informatik.rwth-aachen.de/Publications/CEUR-WS/

Nowadays, the use of social information is extending to more and more application domains. In the... more Nowadays, the use of social information is extending to more and more application domains. In the field of recommender systems, this information has been exploited in different ways to address some problems, especially associated with collaborative filtering methods, and thus achieve more reliable recommendations. Specifically, social tagging is used in this area mainly to characterize the items that are the subject of the recommendations. In this work, a user-based collaborative filtering approach is presented, where tags processed by word embedding techniques are used to characterize users. User similarities based on both tag embedding and ratings are combined to generate the recommendations. In the study conducted on two popular datasets, the reliability of this approach for rating prediction and top-N recommendations was tested, showing the best performance against the most widely used collaborative filtering methods.

ACM Transactions on The Web, Jun 23, 2017

arXiv (Cornell University), Aug 2, 2019

Lecture Notes in Computer Science, 2018

Suppliers of music streaming services are showing an increasing interest for providing users with... more Suppliers of music streaming services are showing an increasing interest for providing users with reliable personalized recommendations since their practically unlimited offerings make it difficult for users to find the music they like. In this work, we take advantage of social tags that users give to music through streaming platforms for improving recommendations. Most of the works in the literature use the tags in the context of content based methods for finding similarities between songs and artists, but we use them for characterizing users, instead of characterizing music, aiming at improving user-based collaborative filtering algorithms. The expertise level of users is inferred from the frequency analysis of their tags by using TF-IDF (Term Frequency-Inverse Document Frequency), which is an indicator of the quantity and relevance of the tags that users provide to items. User expertise has been studied in the context of recommender systems and other domains, but, as far as we know, it has not been studied in the context of music recommendations.