Developing An Approach For Opinion Mining On Online Reviews

Abstract

People all over the world use social networking sites such as Facebook, Twitter, Instagram, etc. especially during the COVID-19 pandemic. In order to share their experiences, opinions and thoughts on specific top-ics with others. These opinions and thoughts are an important reference to customers in evaluating the products and taking decisions on which product to purchase. Moreover, the vendors use these opinions to know the positive and negative aspects of their products. Therefore, opin-ions collection and analysis play a vital role in the product design and marketing. Opinion mining is used to monitor the peoples’ feelings and emotions toward specific products or services Arunkarthi and Gandhi [2]. Opinion mining is useful for filtering the users’ opinions and gen-erating a quick summary from a large amount data in an acceptable time. However, the large number of opinions is considered as beneficial as challenging to analyze. There are different kinds of opinion analysis which are manual analysis, keyword analysis, and natural language anal-ysis Usharani [59]. Manual analysis is time consuming and may fail or cause incorrect decisions as it depends on human observation and inter-pretation. Keyword analysis classifies the individual terms into positive or negative and measures the overall score of the comment. Natural language analysis analyzes the language meaning and is also called text analytic and computational linguistics. Opinion mining is used in dif-