Sentiresearch: Lexicon-Based Web Application For Indonesian Sentiment Analysis

Abstract

Today we are living in the “age of customer”, where customer or user knowledge and perception is a key for running successful business. Thus, many companies in the world have the role of researcher to dive deep and do research specifically for the customers or users. Analyzing qualitative data is one of the main tasks that every researcher has to tackle and it can even be laborious for advanced market research firms. Therefore, the researchers find the research results are difficult to analyze because the responses can be very haphazard understanding and there might be bias over feedbacks. To overcome this issue, Sentiment Analysis as part of Natural Language Processing is applied in the development of web-based applications that allow for visualization of current sentiments associated with a keyword of open question results from in-depth interview narrative. This research used lexicon dictionary-based sentiment analysis with research respondents’ answers or feedbacks as the dataset. The previous research also has proven to have better Sentiment Analysis result with using InSet Lexicon with the range of -5 to +5. The result of this research showed that this system can benefit the research team with a complete end-to-end system for Sentiment Analysis of customer feedbacks in Bahasa Indonesia from interview session along with the data visualization.