Filters








96 Hits in 2.6 sec

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.  ...  To evaluate our algorithm and compare it with others, we employ a serendipity metric that captures each component of serendipity, unlike the most common metric.  ...  Conclusions and future research We proposed a serendipity-oriented greedy (SOG) recommendation algorithm.  ... 
doi:10.5220/0006232800320040 dblp:conf/webist/KotkovVW17 fatcat:j2mau6f2ivcpjnnffavjd5jr5u

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

Denis Kotkov, Jari Veijalainen, Shuaiqiang Wang
2018 Computing  
A serendipity-oriented greedy algorithm To describe the proposed algorithm, we present the notation in Table 1 .  ...  The paper has the following contributions: -We propose a serendipity-oriented recommendation algorithm. -We evaluate existing serendipity-oriented recommendation algorithms.  ... 
doi:10.1007/s00607-018-0687-5 fatcat:6fiohjom3renncuxrucfzpgvw4

Serendipity Identification Using Distance-Based Approach

Widhi Hartanto, Noor Akhmad Setiawan, Teguh Bharata Adji
2021 IJITEE (International Journal of Information Technology and Electrical Engineering)  
The recommendation system is a method for helping consumers to find products that fit their preferences. However, recommendations that are merely based on user preference are no longer satisfactory.  ...  The serendipity data identification is based on a distance-based approach using collaborative filtering and k-means clustering algorithms.  ...  An offline method, such as conducted in [11] , produces a serendipity recommendation system algorithm called Serendipity Oriented Greedy (SOG) algorithm.  ... 
doi:10.22146/ijitee.62344 fatcat:4h55v233yrbqbeoji5fnky7qdu

Time Distribution Based Diversified Point of Interest Recommendation

Fan Mo, Huida Jiao, Hayato Yamana
2020 2020 IEEE 5th International Conference on Cloud Computing and Big Data Analytics (ICCCBDA)  
inappropriateness for the user to visit those POIs.  ...  To solve this problem, we propose a new concept-time diversity and a time distribution based recommendation method to improve time diversity of recommended POIs.  ...  The maximum value for each metric is noted bold. For both base algorithms, our proposed method achieves higher ILD@K and Serendipity@K.  ... 
doi:10.1109/icccbda49378.2020.9095741 fatcat:ywkmtxpotfcbdnqpkvo7sidra4

Influence of tweets and diversification on serendipitous research paper recommender systems

Chifumi Nishioka, Jörn Hauke, Ansgar Scherp
2020 PeerJ Computer Science  
of a list of recommended items further improves serendipity.  ...  As an evaluation metric, we use the serendipity score (SRDP), in which the unexpectedness of recommendations is inferred by using a primitive recommendation strategy.  ...  ADDITIONAL INFORMATION AND DECLARATIONS Funding This work was supported by the EU H2020 project MOVING (No. 693092), the JSPS Grant-in-Aidfor Scientific Research (S) (No. 16H06304), and the JSPS Grant-in-Aid for  ... 
doi:10.7717/peerj-cs.273 pmid:33816924 pmcid:PMC7924691 fatcat:ppkqtvifgnhhbllnoypej4yeju

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
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.  ...  As a result, many state-of-the-art algorithms are designed to solve supervised learning problems, and progress is judged only by offline metrics.  ...  Herlocker et al. [2004] proposed a variety of metrics for assessing a recommender's coverage, diversity, novelty, and serendipity, while Kaminskas and Bridge [2016] provide a more recent survey of the  ... 
arXiv:2011.07931v1 fatcat:fre2cuepjzcv5gtnk3ulnblywu

A Multinomial Probabilistic Model for Movie Genre Predictions

Eric Arnaud Makita Makita, Artem Lenskiy
2016 International Journal of Machine Learning and Computing  
We achieved 70% prediction rate using only 15% of the whole set for training.  ...  The prediction of a movie's genre has many practical applications including complementing the item's categories given by experts and providing a surprise effect in the recommendations given to a user.  ...  [8] enhanced the diversity of recommendations by adopting the re-ranking approach with a greedy selection. enhanced the diversity of recommendations by adopting the re-ranking approach with a greedy  ... 
doi:10.18178/ijmlc.2016.6.2.580 fatcat:4byecngy7fa45mlrc7g7zgk5hm

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  
canvas; (4) compose it with other shapes; and (5) repeat for a full drawing.  ...  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

Diversity in Big Data: A Review

Marina Drosou, H.V. Jagadish, Evaggelia Pitoura, Julia Stoyanovich
2017 Big Data  
work that will position diversity as an important component of a data-responsible society.  ...  Big data technology offers unprecedented opportunities to society as a whole and also to its individual members. At the same time, this technology poses significant risks to those it overlooks.  ...  In Ref. 7 a greedy algorithm is used for the coverage-based model that bases div(S, i) on diversity and relevance, whereas in Ref. 41 a greedy algorithm is used for the coverage-based model that bases  ... 
doi:10.1089/big.2016.0054 pmid:28632443 fatcat:tjyodpr3ujdvxakzjb6sbhoux4

A method for evaluating discoverability and navigability of recommendation algorithms

Daniel Lamprecht, Markus Strohmaier, Denis Helic
2017 Computational Social Networks  
Our work expands the arsenal of evaluation techniques for recommendation algorithms, extends from a one-click-based evaluation towards multi-click analysis, and presents a general, comprehensive method  ...  For (2) and (3), however, a search function is generally not sufficient.  ...  Serendipity Serendipity, or pleasant surprise, measures the fraction of recommendations that are both novel (surprising) and relevant (interesting) [2, 16, 23] .  ... 
doi:10.1186/s40649-017-0045-3 pmid:29266112 pmcid:PMC5732611 fatcat:x3nbulcksnggbjencyr2b6zv2e

A multinomial probabilistic model for movie genre predictions [article]

Eric Makita, Artem Lenskiy
2016 arXiv   pre-print
We achieved 70% prediction rate using only 15% of the whole set for training.  ...  The prediction of a movie genre has many practical applications including complementing the items categories given by experts and providing a surprise effect in the recommendations given to a user.  ...  Since then many recommender system algorithms and their variants have been proposed in literature, however, most of them were mainly accuracy-oriented algorithms that predict the rating of an item.  ... 
arXiv:1603.07849v1 fatcat:m6jc42s6vffyjhwesod3v6yanm

Modeling and broadening temporal user interest in personalized news recommendation

Lei Li, Li Zheng, Fan Yang, Tao Li
2014 Expert systems with applications  
Given the processed news articles (often represented as a topic vector), we can adopt a greedy algorithm to sequentially pick up the news article with the largest similarity.  ...  For example, News Dude (Billsus & Pazzani, 1999b) , is a personal news recommender agent that utilizes TF-IDF combined with the K-Nearest Neighbor algorithm to recommend news items to individual users  ... 
doi:10.1016/j.eswa.2013.11.020 fatcat:e4myzr2aijgjdaf6zatsv7aiyy

Reinforcement learning based recommender systems: A survey [article]

M. Mehdi Afsar, Trafford Crump, Behrouz Far
2021 arXiv   pre-print
We first recognize the fact that algorithms developed for RLRSs can be generally classified into RL- and DRL-based methods.  ...  Then, we present these RL- and DRL-based methods in a classified manner based on the specific RL algorithm, e.g., Q-learning, SARSA, and REINFORCE, that is used to optimize the recommendation policy.  ...  Two types of algorithms are developed for ad recommendation. The first one targets the greedy optimization (CTR) and uses random forest to learn a mapping from features to actions.  ... 
arXiv:2101.06286v1 fatcat:e234kqjtujazpdvt2wjek4et4i

Optimized Recommender Systems with Deep Reinforcement Learning [article]

Lucas Farris
2021 arXiv   pre-print
recommendations.  ...  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

Graph-based collaborative ranking

Bita Shams, Saman Haratizadeh
2017 Expert systems with applications  
The experimental results show a significant improvement in recommendation quality compared to the state of the art graph-based recommendation algorithms and other collaborative ranking techniques.  ...  In this paper, we propose a novel graph-based approach, called GRank, that is designed for collaborative ranking domain.  ...  The goal of rating-oriented algorithms is to accurately predict a user's ratings and then, recommend the items with the highest predicted rating for him.  ... 
doi:10.1016/j.eswa.2016.09.013 fatcat:c5pxcfm2inadnkwl6nyclyf6me
« Previous Showing results 1 — 15 out of 96 results