19,564 Hits in 8.7 sec

Algorithms are not neutral

Catherine Stinson
2022 AI and Ethics  
Here we illustrate the point that algorithms themselves can be the source of bias with the example of collaborative filtering algorithms for recommendation and search.  ...  Biased algorithms for applications such as media recommendations can have significant impact on individuals' and communities' access to information and culturally-relevant resources.  ...  Collaborative filtering, as shown here, is standard in its industry, does not use proxies for protected categories, and its objective function, prediction accuracy, is a valid objective; however, the algorithm  ... 
doi:10.1007/s43681-022-00136-w pmid:35128540 pmcid:PMC8802245 fatcat:lfhaw2upffcvhnaylmeaokjbly

Beyond Parity: Fairness Objectives for Collaborative Filtering [article]

Sirui Yao, Bert Huang
2017 arXiv   pre-print
We study fairness in collaborative-filtering recommender systems, which are sensitive to discrimination that exists in historical data.  ...  Biased data can lead collaborative-filtering methods to make unfair predictions for users from minority groups.  ...  Fairness Objectives for Collaborative Filtering This section introduces fairness objectives for collaborative filtering. We begin by reviewing the matrix factorization method.  ... 
arXiv:1705.08804v2 fatcat:fpnudfzjbvcsjaz5z6bq7zsbn4

Causal Inference in Recommender Systems: A Survey and Future Directions [article]

Chen Gao, Yu Zheng, Wenjie Wang, Fuli Feng, Xiangnan He, Yong Li
2022 arXiv   pre-print
Existing recommender systems extract the user preference based on learning the correlation in data, such as behavioral correlation in collaborative filtering, feature-feature, or feature-behavior correlation  ...  For example, the recommender systems can recommend a battery charger to a user after buying a phone, in which the latter can serve as the cause of the former, and such a causal relation cannot be reversed  ...  Improving these beyond-accuracy objectives may hurt the recommendation accuracy, resulting in a dilemma.  ... 
arXiv:2208.12397v1 fatcat:jpjp5sjunvczhmikx5gfjvc3qq

Recommendation Systems: Classification, Open Issues and Recent Developments

Texts here trace out diverse characteristics concerned with recommendation system and highlight possible recommendation-methodologies capacity in this arena  ...  Everyone is confronted with information flood phenomena and the recommendation engine alleviate Information flood on internet.  ...  where person interacting to system is not interested in organization of objects beyond dual classes such as good or bad, relevant or Non-Relevant.  ... 
doi:10.35940/ijitee.j8831.078919 fatcat:yb2l37hrbbesjkuxx4oebqihmy

A Brief History of Recommender Systems [article]

Zhenhua Dong, Zhe Wang, Jun Xu, Ruiming Tang, Jirong Wen
2022 arXiv   pre-print
We hope the brief review can help us to know the dots about the progress of web recommender systems, and the dots will somehow connect in the future, which inspires us to build more advanced recommendation  ...  Soon after the invention of the Internet, the recommender system emerged and related technologies have been extensively studied and applied by both academia and industry.  ...  Thanks to the colleagues of Huawei Noah's Ark Lab, there is no this work without their fabulous work in both recommender system applications and academic studies in the past 10 years.  ... 
arXiv:2209.01860v1 fatcat:dvmrna4orjhhxdvrdkon57n54y

Fairness in Music Recommender Systems: A Stakeholder-Centered Mini Review

Karlijn Dinnissen, Christine Bauer
2022 Frontiers in Big Data  
However, many factors make recommender systems prone to biases, resulting in unfair outcomes.  ...  of artist gender in the recommendations.  ...  Ferraro et al. (2020) and Shakespeare et al. (2020) found that collaborative filtering algorithms could propagate or even amplify those biases in a MRS, thereby negatively impacting minority genders  ... 
doi:10.3389/fdata.2022.913608 pmid:35937551 pmcid:PMC9353048 fatcat:adjh25phcvh7nfomgquhbr2vei

An Explainable Autoencoder For Collaborative Filtering Recommendation [article]

Pegah Sagheb Haghighi, Olurotimi Seton, Olfa Nasraoui
2019 arXiv   pre-print
They have also been successfully used in Collaborative Filtering (CF) recommender systems to predict missing ratings.  ...  In this work, we design an explainable recommendation system using an Autoencoder model whose predictions can be explained using the neighborhood based explanation style.  ...  Latent factor models have been the state of the art in Collaborative Filtering recommender systems.  ... 
arXiv:2001.04344v1 fatcat:cwtrzis2v5c77g2p6c2wk2gv4y

Research on Collaborative Filtering Recommendation Algorithm Based on Mahout

2020 DEStech Transactions on Environment Energy and Earth Science  
The similarity distance and other parameters in the general recommendation algorithm are used to compare and analyze the recommended results.  ...  The principle analysis of the current mainstream recommendation algorithm is based on project-based collaborative filtering recommendation.  ...  Recommendation Based on Collaborative Filtering The recommendation algorithm based on collaborative filtering is one of the most mature algorithms in the recommendation system.  ... 
doi:10.12783/dteees/peems2019/34001 fatcat:m266cn427vdhlowjwfw3ejz6oa

Looking for "Good" Recommendations: A Comparative Evaluation of Recommender Systems [chapter]

Paolo Cremonesi, Franca Garzotto, Sara Negro, Alessandro Vittorio Papadopoulos, Roberto Turrin
2011 Lecture Notes in Computer Science  
A number of researches in the Recommender Systems (RSs) domain suggest that the recommendations that are "best" according to objective metrics are sometimes not the ones that are most satisfactory or useful  ...  We measured the user's perceived quality of each of them, focusing on accuracy and novelty of recommended items, and on overall users' satisfaction.  ...  Collaborative algorithms are trained (e.g., tuned) to achieve the best performance in terms of objective accuracy.  ... 
doi:10.1007/978-3-642-23765-2_11 fatcat:6h2blumuxrgjbdqabxz4cowshi

StakeRare: Using Social Networks and Collaborative Filtering for Large-Scale Requirements Elicitation

Soo Ling Lim, A. Finkelstein
2012 IEEE Transactions on Software Engineering  
This paper proposes StakeRare, a novel method that uses social networks and collaborative filtering to identify and prioritise requirements in large software projects.  ...  It then asks the stakeholders to rate an initial list of requirements, recommends other relevant requirements to them using collaborative filtering, and prioritises their requirements using their ratings  ...  accuracy, and using other collaborative filtering algorithms can also improve prediction accuracy.  ... 
doi:10.1109/tse.2011.36 fatcat:ruuycvz3yjc55pztjgyoepoduq

News Recommender System: A review of recent progress, challenges, and opportunities [article]

Shaina Raza, Chen Ding
2021 arXiv   pre-print
In the first part, we present an overview of the conventional recommendation solutions, datasets, evaluation criteria beyond accuracy and recommendation platforms being used in NRS.  ...  By providing the state-of-the-art knowledge, this survey can help researchers and practical professionals in their understanding of developments in news recommendation algorithms.  ...  Acknowledgment: This work is partially sponsored by Natural Science and Engineering Research Council of Canada (grant 2020-04760).  ... 
arXiv:2009.04964v4 fatcat:s7jl63nwm5e55myezsxpzquuje

Thresholding for Top-k Recommendation with Temporal Dynamics [article]

Lei Tang
2015 arXiv   pre-print
of items and users in practice.  ...  Such a bias learning process alleviates data sparsity in constructing the engine, and at the same time captures recent trend shift observed in data.  ...  in most collaborative filtering that training and test data share the same distribution.  ... 
arXiv:1506.02190v2 fatcat:5fcbe4c3zvef5jesgjzus7shfa

Collaborative Metric Learning

Cheng-Kang Hsieh, Longqi Yang, Yin Cui, Tsung-Yi Lin, Serge Belongie, Deborah Estrin
2017 Proceedings of the 26th International Conference on World Wide Web - WWW '17  
In this work we study the connection between metric learning and collaborative filtering.  ...  The proposed algorithm outperforms stateof-the-art collaborative filtering algorithms on a wide range of recommendation tasks and uncovers the underlying spectrum of users' fine-grained preferences.  ...  Group, Google, Pfizer, RWJF, NIH and NSF.  ... 
doi:10.1145/3038912.3052639 dblp:conf/www/HsiehYCLBE17 fatcat:xco5r7gptjdq3lqugyzfopzvzm

Exploring Gender Distribution in Music Recommender Systems

Dougal Shakespeare, Lorenzo Porcaro, Emilia Gómez
2020 Zenodo  
Whilst accuracy metrics have been widely applied to evaluate recommendations in mRS literature, evaluating a user's item utility from other impact-oriented perspec-tives, including their potential for  ...  Music Recommender Systems (mRS) are designed to give personalised and meaning-ful recommendations of items (i.e. songs, playlists or artists) to a user base, thereby reflecting and further complementing  ...  Accuracy and beyond-accuracy metrics.  ... 
doi:10.5281/zenodo.4091510 fatcat:5ug2dprnr5haxgfukjtpofnlxe

Latent Unexpected Recommendations [article]

Pan Li, Alexander Tuzhilin
2020 arXiv   pre-print
Unexpected recommender system constitutes an important tool to tackle the problem of filter bubbles and user boredom, which aims at providing unexpected and satisfying recommendations to target users at  ...  in the loss of accuracy measures in order to improve unexpectedness performance.  ...  Beyond-Accuracy Metrics As researchers have pointed out, accuracy is not the only important objective of recommendations [41] , while other beyond-accuracy metrics should also be taken into account, including  ... 
arXiv:2007.13280v1 fatcat:oxnronmcp5bm7h5sllm3axkryu
« Previous Showing results 1 — 15 out of 19,564 results