<![CDATA[The recommendation system assists users in finding items that match their preferences from the large number of items that exist. Recommendation systems have two types of approaches: a content-based approach and a Collaborative Filtering (CF) approach. CF approaches can be categorized into model-based and memory-based CF. The problem faced in the CF method is the complexity or long computation time due to the large data dimensions, data sparsity, and accuracy. In overcoming the problems mentioned, several data mining and machine learning techniques are used in collaboration with traditional CF methods. Many studies are using numerical data clustering techniques on CF-based recommendation systems. However, to date, there is still no literature review regarding the implementation of clustering techniques to numerical data to develop recommendation system methods based on the CF approach. Therefore, a literature study was carried out regarding the implementation of clustering techniques to numerical data to develop recommendation system methods based on the CF approach using 20 related literature. As a result, the various clustering techniques used can be grouped into K-Means, Subspace Clustering, Bi-Clustering, Canopy Clustering, K-Medoids, Evolutionary Heterogeneous Clustering, Fuzzy, Self-Constructing Clustering (SCC), and Agglomerative Hierarchical Clustering (AHC). K-Means and Fuzzy clustering techniques are the most commonly found in the literature.]]>
Copyrights © 2024
