Evaluating Recommendation Systems

Shani, Guy; Gunawardana, Asela

doi:10.1007/978-0-387-85820-3_8

Cited by 953 publications

(691 citation statements)

References 46 publications

Supporting

Mentioning

634

Contrasting

Unclassified

Order By: Relevance

“…Unlike for information retrieval systems, a consistent notion of relevance has not been established. Shani and Gunawardana [25] distinguish three evaluation methodologies: offline experiments, user studies, and online evaluation. In News-REEL, our focus is on the static environments of offline experimentation and dynamic environments of online evaluation.…”

Section: Evaluation Of News Recommendation Systemsmentioning

confidence: 99%

CLEF 2017 NewsREEL Overview: A Stream-Based Recommender Task for Evaluation and Education

Lommatzsch

Kille

Hopfgartner

et al. 2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. News recommender systems provide users with access to news stories that they find interesting and relevant. As other online, stream-based recommender systems, they face particular challenges, including limited information on users' preferences and also rapidly fluctuating item collections. In addition, technical aspects, such as response time and scalability, must be considered. Both algorithmic and technical considerations shape working requirements for realworld recommender systems in businesses. NewsREEL represents a unique opportunity to evaluate recommendation algorithms and for students to experience realistic conditions and to enlarge their skill sets. The NewsREEL Challenge requires participants to conduct data-driven experiments in NewsREEL Replay as well as deploy their best models into NewsREEL Live's 'living lab'. This paper presents NewsREEL 2017 and also provides insights into the effectiveness of NewsREEL to support the goals of instructors teaching recommender systems to students. We discuss the experiences of NewsREEL participants as well as those of instructors teaching recommender systems to students, and in this way, we showcase NewsREEL's ability to support the education of future data scientists.

show abstract

Section: Evaluation Of News Recommendation Systemsmentioning

confidence: 99%

CLEF 2017 NewsREEL Overview: A Stream-Based Recommender Task for Evaluation and Education

Lommatzsch

Kille

Hopfgartner

et al. 2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Recommender frameworks can now be found in numerous present day applications that open the client to an enormous accumulations of things [246]. Such frameworks normally give the client a rundown of prescribed things they may incline toward, or foresee the amount they may lean toward everything.…”

Section: About Recommendation Systemsmentioning

confidence: 99%

Semantically-enhanced advertisement recommender systems in social networks

Pazahr

Zapater

García‐Sánchez

et al. 2016

Proceedings of the 18th International Conference on Information Integration and Web-Based Applications and Services

View full text Add to dashboard Cite

“…For this purpose, we used Precision, Recall and F1 measures [10]. We can now define recall, precision and F1 for Top n recommender systems in the following way:…”

Section: B Evaluation Measuresmentioning

confidence: 99%

A Categorized Item Recommender System Coping with User Interest Changes

Ko¹,

Jiamthapthaksin²

2014

IJMLC

View full text Add to dashboard Cite

Abstract-One significant characteristic of data in specific domain like movie challenges research in recommender systems that user preferences naturally changes over time. Traditional collaborative filtering (CF) method does not take in consideration sequences of customer's rating, which reflects changes of customer's preference over a period of time. This paper proposes a novel recommender system that overcomes the limitation of CF by combining collaborative filtering and sequential pattern mining with time interval which reflects user's preference changes over a period of time. Sequential patterns of categories of items are generated which represents and summarizes interest changes of users varied over time, and are used for revising recommended items produced by traditional CF. Experimental results show that the proposed system show improvements over the traditional collaborative filtering method.Index Terms-Collaborative filtering, interest drift, sequential pattern mining, recommender system.

show abstract

Evaluating Recommendation Systems

Cited by 953 publications

References 46 publications

CLEF 2017 NewsREEL Overview: A Stream-Based Recommender Task for Evaluation and Education

CLEF 2017 NewsREEL Overview: A Stream-Based Recommender Task for Evaluation and Education

Semantically-enhanced advertisement recommender systems in social networks

A Categorized Item Recommender System Coping with User Interest Changes

Contact Info

Product

Resources

About