Aspect-based sentiment analysis on current state of the market by classifying customer product reviews using support vector machine with several kernel learning

Abstract

Aspect-based Sentiment Analysis identifies a customer's feelings about a given product feature. Aspect-based analysis is often used in product analytics to keep track of how a product is regarded by customers and what its strengths and weaknesses are. Product analytics is a subset of the classification problem. The Support Vector Machine (SVM) is a technology that can effectively tackle classification problems. In fact, SVM is a linear classifier, which implies it can only be used to classify data that is linearly separable. This approach should be enhanced with kernel learning to categorize non-linear separable data. This suggested system would create a classification model based on the Support Vector Machine, which will be updated with multiple kernel learning techniques and applied to a market data set for analytics. Sentiment Analysis examines a product's perception and market comprehension via the prism of sentiment data. To study customers, the proposed aspect-based sentiment analysis includes three levels of content analysis and four categories of opinions on a market data set.