Improving Cluster Quality for Prediction of User Behaviors Using CTMSP Mining

Provided by: International forum of researchers Students and Academician
Topic: Data Management
Format: PDF
Increasing popularity of wireless technology leads to research in mining and prediction of mobile movements and associated transactions. Discovery of mobile patterns from logs may not be precise enough for predictions since the varying mobile behaviors among users and temporal periods are not considered. In this paper, Hubert's gamma statistics is proposed to calculate serial point correlation between clusters to improve its quality in CTMSP-mine algorithm to discover the CTMSPs. A prediction strategy is proposed to predict the mobile behaviors.

Find By Topic