A Survey of Various Technique used in Aspect Level Sentiment Analysis

Abstract

In recent years, the vast majority of e-commerce companies have permitted users to write product review. Customers struggle to select the proper product among a plethora of brands. In this situation, sentiment analysis may help extract reviews from e-commerce websites and determine whether product brands are excellent or poor. Sentiment analysis is quickly becoming an indispensable technique for tracking and evaluating people's moods and emotions such as happiness, rage, and sadness. Sometimes merely providing negative and positive evaluations for a product is sufficient. A user may be interested in understanding the polarity of only a subset of a product's features at times. In this instance, aspect-based sentiment analysis allows the user to pick the product attributes of interest in order to acquire summary information about the product feature. The primary goal is to examine various techniques utilized in aspect level sentiment analysis from given text data, such as lexicon, machine learning, and deep learning techniques, in order to uncover aspect based sentiment analysis of provided text data.