Sentiment Analysis of Malaysian Insurance Companies (SAMIC): A Visualization using Support Vector Machine Algorithm

Abstract

This paper proposed implementing a web application to visualize Malaysia’s best insurance companies to assist policyholders in understanding the public views of the policy offered by the top insurance companies in Malaysia, which are AIA, Prudential, and Great Eastern. The complexities of insurance purchasing issues include assessing financial needs and choosing an insurance policy from which the established companies became confusing and challenging for potential policyholders as they will be entering into a long-term investment. They need to allocate adequate time to review each insurance company’s offer to make a wise decision. Hence, this research aimed to design and develop a visualization application through Twitter sentiment analysis using the Support Vector Machine algorithm (SVM). The application acts as a medium to visualize tweets’ sentiment analysis results that mentioned these insurance companies. Twitter was used as a source of data in this study. The tweets extracted using dates and keywords were analyzed as it is one of the metrics that will advance insurance companies’ online presence. Testing phases have shown that the classifier successfully classified tweets’ sentiment with 90% accuracy, with every feature in the application functions correctly and visualizing details of sentiment analysis.