Automatic Ratings Generation System for Behavior Analysis

Abstract

User Behavior Analysis plays a pivotal role in finding the behavior of an individual regarding any certain business or social issue. It is helping a business to take an appropriate decision for improvementby understanding customers’ opinion about their products or services in a positive or negative way. Although there are many systems have been developed and implemented till date for performing behavior analysis in different ways; still better advancement is needed due to the nature of the data. In this paper, we have developed a system to generate automatic ratings based on the commentsgiven by the banking customers on social networking sites. These ratings then further analyzed to find out positive and negative behavior. PIG ETL tool on the top of MapReduce is used to develop and implement a proposed system and performing the analysis. AFINNsentiment lexicon is used to generate automatic ratings whereasvisualization is done using D3.js. Performance evaluation of the proposed system is done by comparing it with the existing system.