Abstract

Identification of neighbourhood based on multi-clusters has been successfully applied to recommender systems, increasing recommendation accuracy and eliminating divergence related to a difference in clustering schemes. The algorithm M-CCF was developed for this purpose that was described in author's previous papers. However, the solution do not equally take advantage on all the partitionings. Selection of clusters to forward to recommender system's input, without deterioration in recommendation accuracy, can simplify its structure. The article describes a solution of a cluster selection based on entropy measure between clustering schemes, eliminating ones, which are redundant. The results reported in this paper confirmed its positive impact on the M-CCF system's overall recommendation performance (measured by RMSE and Coverage).

Recommended Citation

Kużelewska, U. (2022). Performance of Multi-Clustering Recommender System after Selection of Clusters based on V-Measures. In R. A. Buchmann, G. C. Silaghi, D. Bufnea, V. Niculescu, G. Czibula, C. Barry, M. Lang, H. Linger, & C. Schneider (Eds.), Information Systems Development: Artificial Intelligence for Information Systems Development and Operations (ISD2022 Proceedings). Cluj-Napoca, Romania: Risoprint. ISBN: 978-973-53-2917-4. https://doi.org/10.62036/ISD.2022.33

Paper Type

Short Paper

DOI

10.62036/ISD.2022.33

Share

COinS
 

Performance of Multi-Clustering Recommender System after Selection of Clusters based on V-Measures

Identification of neighbourhood based on multi-clusters has been successfully applied to recommender systems, increasing recommendation accuracy and eliminating divergence related to a difference in clustering schemes. The algorithm M-CCF was developed for this purpose that was described in author's previous papers. However, the solution do not equally take advantage on all the partitionings. Selection of clusters to forward to recommender system's input, without deterioration in recommendation accuracy, can simplify its structure. The article describes a solution of a cluster selection based on entropy measure between clustering schemes, eliminating ones, which are redundant. The results reported in this paper confirmed its positive impact on the M-CCF system's overall recommendation performance (measured by RMSE and Coverage).