Filters








64 Hits in 11.4 sec

How does serendipity affect diversity in recommender systems? A serendipity-oriented greedy algorithm

Denis Kotkov, Jari Veijalainen, Shuaiqiang Wang
2018 Computing  
Evaluation metrics The main objective of our algorithm is to improve serendipity of a recommender system. A change of serendipity might affect other properties of a recommender system.  ...  A serendipity-oriented greedy algorithm To describe the proposed algorithm, we present the notation in Table 1 .  ... 
doi:10.1007/s00607-018-0687-5 fatcat:6fiohjom3renncuxrucfzpgvw4

A Serendipity-Oriented Greedy Algorithm for Recommendations

Denis Kotkov, Jari Veijalainen, Shuaiqiang Wang
2017 Proceedings of the 13th International Conference on Web Information Systems and Technologies  
In this paper, we propose a serendipity-oriented algorithm, which improves serendipity through feature diversification and helps overcome the overspecialization problem.  ...  Most recommender systems suggest items to a user that are popular among all users and similar to items the user usually consumes.  ...  Evaluation metrics The main objective of our algorithm is to improve serendipity of a recommender system. A change of serendipity might affect other properties of a recommender system.  ... 
doi:10.5220/0006232800320040 dblp:conf/webist/KotkovVW17 fatcat:j2mau6f2ivcpjnnffavjd5jr5u

Juxtapoze

William Benjamin, Senthil Chandrasegaran, Devarajan Ramanujan, Niklas Elmqvist, SVN Vishwanathan, Karthik Ramani
2014 Proceedings of the 32nd annual ACM conference on Human factors in computing systems - CHI '14  
ABSTRACT Juxtapoze is a clipart composition workflow that supports creative expression and serendipitous discoveries in the shape domain.  ...  Results from a qualitative evaluation show that Juxtapoze makes the process of creating image compositions enjoyable and supports creative expression and serendipity.  ...  Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the sponsors.  ... 
doi:10.1145/2556288.2557327 dblp:conf/chi/BenjaminCREVR14 fatcat:72faey7r3zf2ti6oes2kg3pzne

Do Offline Metrics Predict Online Performance in Recommender Systems? [article]

Karl Krauth, Sarah Dean, Alex Zhao, Wenshuo Guo, Mihaela Curmei, Benjamin Recht, Michael I. Jordan
2020 arXiv   pre-print
Recommender systems operate in an inherently dynamical setting. Past recommendations influence future behavior, including which data points are observed and how user preferences change.  ...  We study the impact of adding exploration strategies, and observe that their effectiveness, when compared to greedy recommendation, is highly dependent on the recommendation algorithm.  ...  Since this oracle baseline is still greedy, it does not plan for environment dynamics.  ... 
arXiv:2011.07931v1 fatcat:fre2cuepjzcv5gtnk3ulnblywu

Optimized Recommender Systems with Deep Reinforcement Learning [article]

Lucas Farris
2021 arXiv   pre-print
Recommender Systems have been the cornerstone of online retailers.  ...  This work investigates and develops means to setup a reproducible testbed, and evaluate different state of the art algorithms in a realistic environment.  ...  Acknowledgements The proposal for this work is to leverage interaction data from large retailers, use them to generate a RL environment, and measure how different Deep Reinforcement Learning (DRL) algorithms  ... 
arXiv:2110.03039v1 fatcat:oqjh4aezxjdb7jkvrn6qw4254a

Active Learning in Recommender Systems [chapter]

Neil Rubens, Dain Kaplan, Masashi Sugiyama
2010 Recommender Systems Handbook  
Depending on the task at hand, such as specific goal oriented assistants, this may also be a nice fit for a Recommender System.  ...  In essence, the recommended items should be as representative and diverse as possible, which should be possible without appreciably affecting their similarity to the user query.  ...  Algorithm 1 Output estimates-based Active Learning (Section 6.1.1). # G estimates predictive error that rating an item x a would allow to achieve function G(x a ) # learn a preference approximation function  ... 
doi:10.1007/978-0-387-85820-3_23 fatcat:engjjwo3evcifg6ws456w7nu4u

Active Learning in Recommender Systems [chapter]

Neil Rubens, Mehdi Elahi, Masashi Sugiyama, Dain Kaplan
2015 Recommender Systems Handbook  
Depending on the task at hand, such as specific goal oriented assistants, this may also be a nice fit for a Recommender System.  ...  In essence, the recommended items should be as representative and diverse as possible, which should be possible without appreciably affecting their similarity to the user query.  ...  Algorithm 1 Output estimates-based Active Learning (Section 6.1.1). # G estimates predictive error that rating an item x a would allow to achieve function G(x a ) # learn a preference approximation function  ... 
doi:10.1007/978-1-4899-7637-6_24 fatcat:dz6whn427rhzzpdu4vavy2pqxq

Diversity in Big Data: A Review

Marina Drosou, H.V. Jagadish, Evaggelia Pitoura, Julia Stoyanovich
2017 Big Data  
In this article, we give an overview of recent technical work on diversity, particularly in selection tasks, discuss connections between diversity and fairness, and identify promising directions for future  ...  work that will position diversity as an important component of a data-responsible society.  ...  Acknowledgments This work was supported in part by the National Science Foundation Grants No. 1464327, 1539856, and 1250880, and by the US-Israel Binational Science Foundation Grant No. 2014391.  ... 
doi:10.1089/big.2016.0054 pmid:28632443 fatcat:tjyodpr3ujdvxakzjb6sbhoux4

A Survey on Mobile Crowdsensing Systems: Challenges, Solutions and Opportunities

Andrea Capponi, Claudio Fiandrino, Burak Kantarci, Luca Foschini, Dzmitry Kliazovich, Pascal Bouvry
2019 IEEE Communications Surveys and Tutorials  
For data collection, MCS systems rely on contribution from mobile devices of a large number of participants or a crowd.  ...  Despite the growing interest in the research community, MCS solutions need a deeper investigation and categorization on many aspects that span from sensing and communication to system management and data  ...  Then, focusing on the online allocation model, they design a greedy algorithm which achieves a ratio of at most m.  ... 
doi:10.1109/comst.2019.2914030 fatcat:psvt24nrjbcldpixw6b7stzm3a

A method for evaluating discoverability and navigability of recommendation algorithms

Daniel Lamprecht, Markus Strohmaier, Denis Helic
2017 Computational Social Networks  
The proposed method tackles this by means of first evaluating the discoverability of recommendation algorithms by investigating structural properties of the resulting recommender systems in terms of bow  ...  In this paper, we propose a method to expand the repertoire of existing recommendation evaluation techniques with a method to evaluate the discoverability and navigability of recommendation algorithms.  ...  Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.  ... 
doi:10.1186/s40649-017-0045-3 pmid:29266112 pmcid:PMC5732611 fatcat:x3nbulcksnggbjencyr2b6zv2e

Addressing the Multistakeholder Impact of Popularity Bias in Recommendation Through Calibration [article]

Himan Abdollahpouri, Masoud Mansoury, Robin Burke, Bamshad Mobasher
2020 arXiv   pre-print
Popularity bias is a well-known phenomenon in recommender systems: popular items are recommended even more frequently than their popularity would warrant, amplifying long-tail effects already present in  ...  We also show that our proposed algorithm has a secondary effect of improving supplier fairness.  ...  We have seen that there is a limit to how many Tail items can be re-ranked, because the algorithm simply does not return them.  ... 
arXiv:2007.12230v1 fatcat:sgzqa6om4ranxirpm4xf3fvgjq

Bias and Debias in Recommender System: A Survey and Future Directions [article]

Jiawei Chen, Hande Dong, Xiang Wang, Fuli Feng, Meng Wang, Xiangnan He
2021 arXiv   pre-print
While recent years have witnessed a rapid growth of research papers on recommender system (RS), most of the papers focus on inventing machine learning models to better fit user behavior data.  ...  We then provide a taxonomy to position and organize the existing work on recommendation debiasing.  ...  All in all, factors or biases often evolve with the time going by. It will be interesting and valuable to explore how bias evolves and analyze how the dynamic bias affects a RS.  ... 
arXiv:2010.03240v2 fatcat:6fticc3otndsra2whs5e4nrdpi

Towards Effective Research-Paper Recommender Systems and User Modeling based on Mind Maps [article]

Joeran Beel
2017 arXiv   pre-print
To achieve this objective, we integrate a recommender system in our mind-mapping and reference-management software Docear.  ...  While user-modeling and recommender systems successfully utilize items like emails, news, and movies, they widely neglect mind-maps as a source for user modeling.  ...  Other factors that affect user satisfaction are confidence in a recommender system [336], data security [229] , diversity [438] , user tasks [275] , item's lifespan [72] and novelty [433] , risk  ... 
arXiv:1703.09109v1 fatcat:egcsnop34jbi7p2urz34pxr3vi

Popularity Bias in Recommendation: A Multi-stakeholder Perspective [article]

Himan Abdollahpouri
2020 arXiv   pre-print
stakeholders in a recommender system.  ...  In this dissertation, I study the impact of popularity bias in recommender systems from a multi-stakeholder perspective.  ...  successful recommendation such as diversity, serendipity and novelty [46, 66, 164 ] and impacts across system stakeholders [1] .  ... 
arXiv:2008.08551v1 fatcat:yiuamp6lcnc2bmhzkklakq6ui4

Real World Evaluation of Approaches to Research Paper Recommendation [article]

Siddharth Dinesh
2018 arXiv   pre-print
In this work, we have identified the need for choosing baseline approaches for research-paper recommendation systems.  ...  DLib a literature recommendation platform. User click data was collected as part of an ongoing experiment in collaboration with our partner Gesis.  ...  Their experiments deal with finding out how the length of the user profile (in years) affects the performance of the recommendations.  ... 
arXiv:1802.06892v1 fatcat:5e2jgjicxrg65ihagbxwjzvmai
« Previous Showing results 1 — 15 out of 64 results