Filters








16,890 Hits in 5.1 sec

Adaptive matrix completion for the users and the items in tail

Mohit Sharma, George Karypis
2019 The World Wide Web Conference on - WWW '19  
with Dropout (TMF + Dropout) and Inverse Frequency Weighted Matrix Factorization (IFWMF), that outperforms traditional matrix-completion-based approaches for the users and the items with few ratings in  ...  In this work, we show that the skewed distribution of ratings in the user-item rating matrix of real-world datasets affects the accuracy of matrix-completion-based approaches.  ...  The matrix completion approach [3] assumes that the user-item rating matrix is low rank and estimates the missing ratings based on the observed ratings in the matrix.  ... 
doi:10.1145/3308558.3313736 dblp:conf/www/SharmaK19 fatcat:67eqbrvvcfhfnk73cqk2pw6yfi

Toward a New Protocol to Evaluate Recommender Systems [article]

Frank Meyer, Françoise Fessant, Fabrice Clérot, Eric Gaussier
2012 arXiv   pre-print
A segmentation of both users and items is proposed to finely analyze where the algorithms perform well or badly.  ...  In this paper, we propose an approach to analyze the performance and the added value of automatic recommender systems in an industrial context.  ...  In particular, customers who watched items in the long tail are in fact heavy users, light users tend to focus only on popular items.  ... 
arXiv:1209.1983v1 fatcat:rliawbdenzdhvmyp6oebl644jy

A Survey of Long-Tail Item Recommendation Methods

Jing Qin, Danfeng Hong
2021 Wireless Communications and Mobile Computing  
The long-tail item recommendation method not only considers the recommendation of short-head items but also considers recommending more long-tail items to users, thus improving the coverage and diversity  ...  of the research on long-tail item recommendation methods (from clustering-based, which began in 2008, to deep learning-based methods, which began in 2020) and the future directions associated with this  ...  Acknowledgments The author would like to thank the authors of all the references.  ... 
doi:10.1155/2021/7536316 fatcat:3in4tt3ntng6lew6gkvysz3rsi

Balancing Accuracy and Diversity in Recommendations using Matrix Completion Framework [article]

Anupriya Gogna, Angshul Majumdar
2019 arXiv   pre-print
In this work, we attempt to achieve accuracy vs diversity balance, by exploiting available ratings and item metadata, through a single, joint optimization model built over the matrix completion framework  ...  However, several studies have suggested the need for diversified recommendations, with acceptable level of accuracy, to avoid monotony and improve customers experience.  ...  as, X UV = (2) where U is the latent factor matrix of all users and V for all items.  ... 
arXiv:2001.04349v1 fatcat:krhuvj7cnvfmthxp7pnof7aw5i

Unifying Task-oriented Knowledge Graph Learning and Recommendation

Qianyu Li, Xiaoli Tang, Tengyun Wang, Haizhi Yang, Hengjie Song
2019 IEEE Access  
from recommendations for KG completion, and unifies the two tasks in a joint model for mutual enhancements.  ...  Second, the RCoLM provides a general task-oriented negative sampling strategy on KG completion task, which further improves the adaptive ability of the proposed algorithm and plays an essential role for  ...  ACKNOWLEDGMENT (Qianyu Li and Xiaoli Tang are co-first authors.)  ... 
doi:10.1109/access.2019.2932466 fatcat:wvlchwdaynblbfymecyt5p3m7y

Recommender systems in industrial contexts [article]

Frank Meyer
2012 arXiv   pre-print
This thesis consists of four parts: - An analysis of the core functions and the prerequisites for recommender systems in an industrial context: we identify four core functions for recommendation systems  ...  Then we study the performance of the KNN method in critical uses cases in cold start situation. - A methodology for analyzing the performance of recommender systems in industrial context: this methodology  ...  Acknowledgments First of all, I would like to thank all members of the Jury for having kindly agreed to evaluate my work.  ... 
arXiv:1203.4487v2 fatcat:ozrsxdvi7ndahhqwsl3yvhnqsy

Use of Deep Learning in Modern Recommendation System: A Summary of Recent Works

Ayush Singhal, Pradeep Sinha, Rakesh Pant
2017 International Journal of Computer Applications  
for recommendation.  ...  With the exponential increase in the amount of digital information over the internet, online shops, online music, video and image libraries, search engines and recommendation system have become the most  ...  The proposed approach completely replaces the matrix factorization based approach or matrix factorization is expressed as a special case of the proposed generic model for generating user and item latent  ... 
doi:10.5120/ijca2017916055 fatcat:m6icpquumbgczhrdnya7x35of4

Intent Disentanglement and Feature Self-supervision for Novel Recommendation [article]

Tieyun Qian, Yile Liang, Qing Li, Xuan Ma, Ke Sun, Zhiyong Peng
2021 arXiv   pre-print
One key property in recommender systems is the long-tail distribution in user-item interactions where most items only have few user feedback.  ...  Improving the recommendation of tail items can promote novelty and bring positive effects to both users and providers, and thus is a desirable property of recommender systems.  ...  [8] transfer knowledge from head items to tail items for leveraging the rich user feedback in head items and the semantic connections between head and tail items.  ... 
arXiv:2106.14388v1 fatcat:m4i73l45fzgzfhp7k3ud4aqc5y

Adaptive Alleviation for Popularity Bias in Recommender Systems with Knowledge Graph

Feng Wei, Shuyu Chen, Jie Jin, Shuai Zhang, Hongwei Zhou, Yingbo Wu
2022 Security and Communication Networks  
In this work, we propose a novel debias framework with knowledge graph (AWING), which adaptively alleviates popularity bias from the users' perspective.  ...  However, these methods ignore the users' personal popularity preferences and increase the exposure rate of the nonpopular items indiscriminately, which may hurt the user experience because different users  ...  (iii) KTUP [24] employs TransH on user-item interactions and KG triplets simultaneously to learn user preference and perform KG completion.  ... 
doi:10.1155/2022/4264489 doaj:d10ee085cd804d81ba21297d19716770 fatcat:jrsmvuomhndqjaj6365emppco4

Unifying Explicit and Implicit Feedback for Rating Prediction and Ranking Recommendation Tasks

Amir H. Jadidinejad, Craig Macdonald, Iadh Ounis
2019 Proceedings of the 2019 ACM SIGIR International Conference on Theory of Information Retrieval - ICTIR '19  
In contrast, ranking models leverage implicit feedback (e.g. clicks) in order to provide the user with a personalized ranked list of recommended items.  ...  explicit and implicit feedback and improve the effectiveness of the baseline explicit model on the ranking task by covering a broader range of long-tail items.  ...  ACKNOWLEDGMENTS The authors acknowledge support from EPSRC grant EP/ R018634/1 entitled Closed-Loop Data Science for Complex, Computationally-and Data-Intensive Analytics.  ... 
doi:10.1145/3341981.3344225 dblp:conf/ictir/JadidinejadMO19 fatcat:mh3whfspjnhe3aabuuu3xelj2a

Failing to plan – planning to fail

Knut Boge, AlenkaTemeljotov Salaj, Svein Bjørberg, Anne Kathrine Larssen
2018 Facilities  
In this way, buildings can contribute to good value creation during its lifetimeboth for owners and users.  ...  To get good, adaptable and usable buildings over time, there is a need for competent players with good decision and communication tools for projects and processes.  ...  Acknowledgements: Thanks to the respondents who took their time to answer the survey. Also, thanks to the reviewers for comments and suggestions that clearly improved the paper.  ... 
doi:10.1108/f-03-2017-0039 fatcat:43nhekvi4nct7ehykjb7rcip34

Billion-scale Pre-trained E-commerce Product Knowledge Graph Model [article]

Wen Zhang, Chi-Man Wong, Ganqiang Ye, Bo Wen, Wei Zhang, Huajun Chen
2021 arXiv   pre-print
Notably, PKGM could also complete knowledge graphs during servicing, thereby overcoming the common incompleteness issue in knowledge graphs.  ...  We test PKGM in three knowledge-related tasks including item classification, same item identification, and recommendation.  ...  showing what is the tail entity for a given entity and relation, and (3) completing the missing tail entity for a given entity and relation if it should exist.  ... 
arXiv:2105.00388v1 fatcat:ilgvwfc35bdnbp6hs36bii2q7i

Improving sales diversity by recommending users to items

Saúl Vargas, Pablo Castells
2014 Proceedings of the 8th ACM Conference on Recommender systems - RecSys '14  
Sales diversity is also linked with the long-tail novelty of recommendations, a quality dimension from the user perspective.  ...  We explore the inversion of the recommendation task as a means to enhance sales diversity -and indirectly novelty -by selecting which users an item should be recommended to instead of the other way around  ...  Promoting the recommendation of items in this long tail may offer benefits for both users and the business behind the recommender system.  ... 
doi:10.1145/2645710.2645744 dblp:conf/recsys/VargasC14 fatcat:pd5snrotnvehtg7jehmjkcvkxi

Recommender Systems for Large-Scale Social Networks: A review of challenges and solutions

Magdalini Eirinaki, Jerry Gao, Iraklis Varlamis, Konstantinos Tserpes
2018 Future generations computer systems  
In this work we review the various facets of large-scale social recommender systems, summarizing the challenges and interesting problems and discussing some of the solutions.  ...  Social networks have become very important for networking, communications, and content sharing.  ...  This in turns affects item popularity, which is constantly changing, bringing old items to the long tail of user preferences and moving new items to the head [6] .  ... 
doi:10.1016/j.future.2017.09.015 fatcat:jdllbp6snfckfj7xfpu4xycfii

Efficient Model-Based Collaborative Filtering with Fast Adaptive PCA [article]

Xiangyun Ding, Wenjian Yu, Yuyang Xie, Shenghua Liu
2020 arXiv   pre-print
A model-based collaborative filtering (CF) approach utilizing fast adaptive randomized singular value decomposition (SVD) is proposed for the matrix completion problem in recommender system.  ...  Firstly, a fast adaptive PCA frameworkis presented which combines the fixed-precision randomized matrix factorization algorithm [1] and accelerating skills for handling large sparse data.  ...  If none, delete this. relationships between users and items to identify new associations in user-item matrices, where latent factors for users and items are employed to represent the original data matrix  ... 
arXiv:2009.02251v1 fatcat:pfgfbiazrnanvixobfr4cikuxi
« Previous Showing results 1 — 15 out of 16,890 results