Replicating and Improving Top-N Recommendations in Open Source Packages

Collaborative filtering techniques have been studied extensively during the last decade. Many open source packages (Apache Mahout, LensKit, MyMediaLite, rrecsys etc.) have them implemented, but typically the top-N recommendation lists are only based on a highest predicted ratings approach. However, exploiting frequencies in the user/item neighbourhood for the formation of the top-N recommendation lists has been shown to provide superior accuracy results in offline simulations. In this paper, we have therefore implemented extensions to the open source recommendation package for the R language - denoted rrecsys - and compare its performance across open source packages for reasons of replicability. Our experimental results clearly demonstrate that using the most frequent items in neighborhood approach significantly outperforms the highest predicted rating approach on two public datasets.

[1]  Alejandro Bellogín,et al.  Precision-oriented evaluation of recommender systems: an algorithmic comparison , 2011, RecSys '11.

[2]  Lars Schmidt-Thieme,et al.  MyMediaLite: a free recommender system library , 2011, RecSys '11.

[3]  John Riedl,et al.  LensKit: a modular recommender framework , 2011, RecSys '11.

[4]  Guy Shani,et al.  Evaluating Recommendation Systems , 2011, Recommender Systems Handbook.

[5]  John Riedl,et al.  An Empirical Analysis of Design Choices in Neighborhood-Based Collaborative Filtering Algorithms , 2002, Information Retrieval.

[6]  Jöran Beel,et al.  Towards reproducibility in recommender-systems research , 2016, User Modeling and User-Adapted Interaction.

[7]  Bamshad Mobasher,et al.  Improving the Effectiveness of Collaborative Filtering on Anonymous Web Usage Data , 2001 .

[8]  Panagiotis Symeonidis,et al.  Visual Analysis of Recommendation Performance , 2017, RecSys.

[9]  Saul Vargas,et al.  Improving sales diversity by recommending users to items , 2014, RecSys '14.

[10]  George Karypis,et al.  Evaluation of Item-Based Top-N Recommendation Algorithms , 2001, CIKM '01.

[11]  Sean Owen,et al.  Mahout in Action , 2011 .

[12]  Alan Said,et al.  Comparative recommender system evaluation: benchmarking recommendation frameworks , 2014, RecSys '14.

[13]  Markus Zanker,et al.  Replication and Reproduction in Recommender Systems Research - Evidence from a Case-Study with the rrecsys Library , 2017, IEA/AIE.

[14]  F. Maxwell Harper,et al.  User perception of differences in recommender algorithms , 2014, RecSys '14.

[15]  F. Maxwell Harper,et al.  The MovieLens Datasets: History and Context , 2016, TIIS.

[16]  Jonathan L. Herlocker,et al.  A collaborative filtering algorithm and evaluation metric that accurately model the user experience , 2004, SIGIR '04.

[17]  John Riedl,et al.  Rethinking the recommender research ecosystem: reproducibility, openness, and LensKit , 2011, RecSys '11.

[18]  Panagiotis Symeonidis,et al.  Collaborative recommender systems: Combining effectiveness and efficiency , 2008, Expert Syst. Appl..

[19]  John Riedl,et al.  Analysis of recommendation algorithms for e-commerce , 2000, EC '00.

[20]  Hsinchun Chen,et al.  Applying associative retrieval techniques to alleviate the sparsity problem in collaborative filtering , 2004, TOIS.

[21]  John Riedl,et al.  Item-based collaborative filtering recommendation algorithms , 2001, WWW '01.