Robust Collaborative Recommendation

Burke, Robin; O’Mahony, Michael P.; Hurley, Neil

doi:10.1007/978-0-387-85820-3_25

Cited by 52 publications

(50 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To simulate such an environment, we created 1,000 virtual spam users to conduct the attack, and we respectively selected 100 items from the head (0%∼20%) and the tail (20%∼100%) as the attack targets. In this experiment, we conducted nuke attack in the case of the average attack model [Burke et al 2015]. More specifically, we first randomly selected the 50 most popular items from the head of distribution to serve as the filler item set [Burke et al 2015].…”

Section: Evaluation Metricmentioning

confidence: 99%

Improving the Quality of Recommendations for Users and Items in the Tail of Distribution

Cao

et al. 2017

ACM Trans. Inf. Syst.

View full text Add to dashboard Cite

Short-head and long-tail distributed data are widely observed in the real world. The same is true of recommender systems (RSs), where a small number of popular items dominate the choices and feedback data while the rest only account for a small amount of feedback. As a result, most RS methods tend to learn user preferences from popular items since they account for most data. However, recent research in e-commerce and marketing has shown that future businesses will obtain greater profit from long-tail selling. Yet, although the number of long-tail items and users is much larger than that of short-head items and users, in reality, the amount of data associated with long-tail items and users is much less. As a result, user preferences tend to be popularity-biased. Furthermore, insufficient data makes long-tail items and users more vulnerable to shilling attack. To improve the quality of recommendations for items and users in the tail of distribution, we propose a coupled regularization approach that consists of two latent factor models: C-HMF, for enhancing credibility, and S-HMF, for emphasizing specialty on user choices. Specifically, the estimates learned from C-HMF and S-HMF recurrently serve as the empirical priors to regularize one another. Such coupled regularization leads to the comprehensive effects of final estimates, which produce more qualitative predictions for both tail users and tail items. To assess the effectiveness of our model, we conduct empirical evaluations on large real-world datasets with various metrics. The results prove that our approach significantly outperforms the compared methods.

show abstract

Section: Evaluation Metricmentioning

confidence: 99%

Improving the Quality of Recommendations for Users and Items in the Tail of Distribution

Cao

et al. 2017

ACM Trans. Inf. Syst.

View full text Add to dashboard Cite

show abstract

“…Moreover, disclosed user profiles may reveal sensitive private information. Profile injection is another possible attack that uses fake profiles to promote or demote recommendations of certain items [3], e.g. by artificially influencing their ratings.…”

Section: Towards Ethical Recommender Systemsmentioning

confidence: 99%

Recommender Systems from an Industrial and Ethical Perspective

Paraschakis

2016

Proceedings of the 10th ACM Conference on Recommender Systems

View full text Add to dashboard Cite

Over the recent years, a plethora of recommender systems (RS) have been proposed by academics. The degree of adoptability of these algorithms by industrial e-commerce platforms remains unclear. To get an insight into real-world recommendation engines, we survey more than 30 existing shopping cart solutions and compare the performance of popular recommendation algorithms on proprietary e-commerce datasets. Our results show that deployed systems rarely go beyond trivial "best seller" lists or very basic personalized recommendation algorithms, which nevertheless exhibit superior performance to more elaborate techniques both in our experiments and other related studies. We also perform chronological dataset splits to demonstrate the importance of preserving the sequence of events during evaluation, and the recency of events during training. The second part of our research is still ongoing and focuses on various ethical challenges that complicate the design of recommender systems. We believe that this direction of research remains mostly neglected despite its increasing impact on RS' quality and safety.

show abstract

“…Similarly, online shopping websites capture behavioral data such as the list of products that the user is browsing through and recommend them with the product(s) that they are most likely to buy. On the other hand, collaborative filtering recommend a customer from the history of the targeted customer and customers who have similar tastes with that of selected customer (Burke et al, 2015).…”

mentioning

confidence: 99%

An integrated recommender system for improved accuracy and aggregate diversity

Bag

Ghadge

Tiwari

2019

Computers & Industrial Engineering

View full text Add to dashboard Cite

Information explosion creates dilemma in finding preferred products from the digital marketplaces. Thus, it is challenging for online companies to develop an efficient recommender system for large portfolio of products. The aim of this research is to develop an integrated recommender system model for online companies, with the ability of providing personalized services to their customers. The K-nearest neighbors (KNN) algorithm uses similarity matrices for performing the recommendation system; however, multiple drawbacks associated with the conventional KNN algorithm have been identified. Thus, an algorithm considering weight metric is used to select only significant nearest neighbors (SNN). Using secondary dataset on MovieLens and combining four types of prediction models, the study develops an integrated recommender system model to identify SNN and predict accurate personalized recommendations at lower computation cost. A timestamp used in the integrated model improves the performance of the personalized recommender system. The research contributes to behavioral analytics and recommender system literature by providing an integrated decision-making model for improved accuracy and aggregate diversity. The proposed prediction model helps to improve the profitability of online companies by selling diverse and preferred portfolio of products to their customers.

show abstract

Robust Collaborative Recommendation

Cited by 52 publications

References 26 publications

Improving the Quality of Recommendations for Users and Items in the Tail of Distribution

Improving the Quality of Recommendations for Users and Items in the Tail of Distribution

Recommender Systems from an Industrial and Ethical Perspective

An integrated recommender system for improved accuracy and aggregate diversity

Contact Info

Product

Resources

About