Cluster Analysis for Personalised Mobile Entertainment Content

There is much attention given to emerging technologies like mobile internet because of its increasing popularity. Much research has concentrated on hardware and some have focused on personalisation in terms of content visualisation. The focus of this paper is on mobile content personalisation, seeking to understand the user groups through clustering users based on their profile. This paper focuses on the implementation of a technique known as ‘Zoning-Centroid’, which is the evaluation technique used to determine the appropriate number of clusters required to best cluster the given users profile. The user profile used in this paper includes mobile content usage and their demographic factors. The clustering algorithm used in this paper is k-means clustering. The results show that the proposed technique could suggest appropriate number of clusters to be used with the k-values, in order to implement for mobile entertainment content personalisation.