MOBILE APP RECOMMENDATION SYSTEM USING K-MEANS AND ITEM-BASED COLLABORATIVE FILTERING

Abstract

Recommender Systems are being widely used in many application settings to suggest products, services, and items to potential users. They are the software techniques providing suggestions for items to be of use to a user. The main purpose of Recommender Systems is to generate meaningful recommendations about the items to a collection of users for their interested items. A variety of approaches in recommendation are user- based collaborative filtering, item-based collaborative filtering, model- based collaborative filtering, content-based recommendation, context- aware recommendations and so on. However, there are two main approaches in recommendation: user-based and item-based collaborative filtering and the difference between them is that user-based takes the users’behavior and item-based takes items’ rating values for similarity measurement. Since the computational complexity of user-based recommendation grows linearly with the number of users, item-based recommendation techniques have been developed. The goal of this system is to provide meaningful recommended applications to the mobile phone users that are relative to their needs or targets. This system uses K- means clustering algorithm to cluster the users based on their age and rating values and item-based collaborative filtering method based on rating values of the items. By using this system, the mobile phone users can get very effective recommendations about applications without waste of time and effort.