Collaborative Filtering for Recommender Systems

Zhang, Ruisheng; Liu, Qi-dong; Chun-Gui,; Wei, Jiaxuan; Huiyi-Ma, None

doi:10.1109/cbd.2014.47

Cited by 45 publications

(20 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this work we analyze different recommendation algorithms with respect to a number of measures and in particular also with respect to their characteristics in terms of "aggregate" diversity and the concentration on certain items in the sense of [3] and [11]. Our results show that while some algorithms are on a par or comparable with respect to their accuracy, they recommend items from quite different areas of the product spectrum.…”

Section: Introductionmentioning

confidence: 91%

“…The top 100 products for Funk-SVD, BPR, and CB-Filtering still accounted for 87.89%, 61.0% and 63.20% of the recommendations. Figure 1 also shows the corresponding Gini index values for the concentration of the recommendations, see also [11]. Higher values of this index -whose values can be between 0 and 1 -indicate a stronger concentration on a small set of items.…”

Section: Popularity-bias Coverage and Aggregate Diversitymentioning

confidence: 99%

“…In particular, we measured how many items of the catalog actually ever appear in top-10 recommendation lists. Note that we use the term "(aggregate) diversity" to denote this special form of catalog coverage as done in [3] or [11]. In other works, the diversity of items in recommendation lists with respect to their content features was also identified to be an important factor that can influence the perceived value of a recommender system, see e.g., in [21].…”

Section: Popularity-bias Coverage and Aggregate Diversitymentioning

confidence: 99%

“…As a first step of our analysis, we used the evaluation approach described in [11]: We grouped items in bins of 100 elements and sorted them in increasing order based on their actual frequency of appearing in top-10 recommendation lists. Figure 1 shows the first four bins containing the most frequently recommended items by a representative subset of the analyzed algorithms for illustration purposes 3 .…”

Section: Popularity-bias Coverage and Aggregate Diversitymentioning

confidence: 99%

See 3 more Smart Citations

What Recommenders Recommend – An Analysis of Accuracy, Popularity, and Sales Diversity Effects

Jannach

Lerche

Gedikli

et al. 2013

User Modeling, Adaptation, and Personalization

View full text Add to dashboard Cite

Abstract. In academic studies, the evaluation of recommender system (RS) algorithms is often limited to offline experimental designs based on historical data sets and metrics from the fields of Machine Learning or Information Retrieval. In real-world settings, however, other businessoriented metrics such as click-through-rates, customer retention or effects on the sales spectrum might be the true evaluation criteria for RS effectiveness. In this paper, we compare different RS algorithms with respect to their tendency of focusing on certain parts of the product spectrum. Our first analysis on different data sets shows that some algorithmswhile able to generate highly accurate predictions -concentrate their top 10 recommendations on a very small fraction of the product catalog or have a strong bias to recommending only relatively popular items than others. We see our work as a further step toward multiple-metric offline evaluation and to help service providers make better-informed decisions when looking for a recommendation strategy that is in line with the overall goals of the recommendation service.

show abstract

Section: Introductionmentioning

confidence: 91%

Section: Popularity-bias Coverage and Aggregate Diversitymentioning

confidence: 99%

Section: Popularity-bias Coverage and Aggregate Diversitymentioning

confidence: 99%

Section: Popularity-bias Coverage and Aggregate Diversitymentioning

confidence: 99%

See 2 more Smart Citations

What Recommenders Recommend – An Analysis of Accuracy, Popularity, and Sales Diversity Effects

Jannach

Lerche

Gedikli

et al. 2013

User Modeling, Adaptation, and Personalization

View full text Add to dashboard Cite

show abstract

“…It proposes the items based on the analysis the feedback provided by the users [3], [25]- [28]. Moreover, CF can be grouped into two main models: model-based and memory-based models [1], [29]. Where model-based need to build a model that will be used later to predict what the users will be preferred.…”

Section: Introductionmentioning

confidence: 99%

A Developed Collaborative Filtering Similarity Method to Improve the Accuracy of Recommendations under Data Sparsity

Al-bashiri¹,

Abdulgabber²,

Romli³

et al. 2018

ijacsa

View full text Add to dashboard Cite

This paper presented a new similarity method to improve the accuracy of traditional Collaborative Filtering (CF) method under sparse data issue. CF provides the user with items, that what they need, based on analyses the preferences of users who have a strong correlation to him/her preference. However, the accuracy is influencing by the method that use to find neighbors. Pearson correlation coefficient and Cosine measures, as the most widely used methods, depending on the rating of only co-rated items to find the correlations between users. Consequently, these methods have lack of ability in addressing the sparsity. This paper presented a new proposed similarity method based on the global user preference to address the sparsity issue and improve the accuracy of recommendation. Thus, the novelty of this method is the ability to solve the similarity issue with a capability of finding the relationship among non-correlated users. Furthermore, to determine the right neighbors during the process of computing the similarity between a pair of users, the developed method considered two main factors (fairness and proportion of co-rated). The MovieLens 100K benchmark dataset is used to evaluate the developed method accuracy. The experiments' result showed that the accuracy of the developed method is improved compared to the traditional CF similarity methods using a specific common CF evaluation metrics.

show abstract

A Novel Hybrid Recommendation Approach Based on Correlation and Co-occurrence Between Activities Within Social Learning Network

Souabi

Retbi

Idrissi

et al. 2021

Advanced Information Networking and Applications

View full text Add to dashboard Cite

Collaborative Filtering for Recommender Systems

Cited by 45 publications

References 7 publications

What Recommenders Recommend – An Analysis of Accuracy, Popularity, and Sales Diversity Effects

What Recommenders Recommend – An Analysis of Accuracy, Popularity, and Sales Diversity Effects

A Developed Collaborative Filtering Similarity Method to Improve the Accuracy of Recommendations under Data Sparsity

A Novel Hybrid Recommendation Approach Based on Correlation and Co-occurrence Between Activities Within Social Learning Network

Contact Info

Product

Resources

About