Abstract
The Collaborative Filtering (CF) widely used in Recommendation System commonly suffers the sparsity issue since the unobserved rating entries usually over dominance the observed ones. A clustering technique is an alternative solution that can solve the problem. However, no in-depth work has investigated how the missing entries should be mitigated and how the cluster-based approach can be implemented. In this study, we show how the imputed cluster-based approach deals with the missing entries, improving the recommendation quality. The framework of our method consists of four main stages: rating imputation to replace the missing entries, K-means clustering to group users or items based on the imputed rating data, CF-based prediction model, and generating the list of top-N recommendation. This paper uses three variations of imputation techniques, i.e., null, mean, and mode. The cluster-based approach is employedby using the K-Means as the clustering technique, and either the user-based or the items-based model as the CF approach. Experiment results show that the null imputation technique gives the best results when dealing with the missing entries. This finding indicates that the implementation of the clustering techniqueis sufficient for solving the sparsity issue such that imputing the missing entries is not necessary. We also show that our imputed cluster-based CF methods always outperform the traditional CF methods in terms of the F1-Score metric.
Highlights
Recommendation Systems (RS) help users to tackle the problem of having to find items that suit their preference from the overwhelming amount of available items
The results confirm that the implementation of a cluster-based approach can improve the recommendation quality of traditional Collaborative Filtering (CF) methods
Our imputed cluster-based CF method implements a combination of an imputation technique and the cluster-based CF approach to deal with the missing rating entries, for generating the list of recommendations
Summary
Recommendation Systems (RS) help users to tackle the problem of having to find items that suit their preference from the overwhelming amount of available items. The summary of our contributions is as follows: (1) the implementation of three rating imputation techniques to deal with the missing rating entries, and (2) the imputed clusterbased CF methods that improve the recommendation quality of the traditional CF approach by implementing the K-Means clustering technique on two CF-based models. The framework of our method consists of four main stages (see Fig. 2), i.e., implementation of an imputation technique to replace the missing rating entries, implementation of K-means clustering to group users or items based on the imputed rating data, implementation of CF-based prediction model, and generating the list of top-N recommendation. UCCF is our cluster-based CF method that implements a combination of the user clustering algorithm and the user-based model to calculate the rating prediction and generate the list of recommendations. Based on the three imputation techniques implemented to deal with the missing rating entries, we vary the methods as UCCF-Null, UCCF-Mean, UCCF-Mode, ICCF-Null, ICCF-Mean, and ICCF-Mode
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
