Application of Generalized Possibilistic Fuzzy C-Means Clustering for User Profiling in Mobile Networks (original) (raw)

2019

Abstract

User profiling is the process of constructing a normal profile by accumulating the past calling behavior of a user. The technique of clustering focusses on outcome of a structure or an intrinsic grouping in unlabeled data collection. In this paper, our main intention is on building appropriate user profile by applying generalized possibilistic fuzzy c-means (GPFCM) clustering technique. All the call features required to build a user profile is collected from the call detail record of the individual users. The behavioral profile modeling of users is prepared by implementing the clustering on two relevant calling features from the reality-mining dataset. The labels are not present in the dataset and thus we have applied clustering which is an unsupervised approach. Before applying the clustering algorithm, a proper cluster validity analysis has to be done for finding the best cluster value and then the cluster analysis is done using some performance parameters.

Suvasini Panigrahi hasn't uploaded this paper.

Let Suvasini know you want this paper to be uploaded.

Ask for this paper to be uploaded.