Critical Analysis of several Feature-based Opinion Mining Methodologies

Feature-based opinion mining is the process of extracting user opinions from the characteristics of an object and is used to do granular analysis by isolating particular properties of an object about which a user has expressed an opinion. In the sentiment analysis method, emotions, feelings, attitudes, and views are all regarded as significant components. With the development of technology built on the World Wide Web, a person or a group can frequently express their opinions or sentiments about internet technology with the use of reviews, blogs, and ratings. The categorization and prediction of the various types of emotions, both good and negative, is the main goal of the machine learning algorithms. When classifying opinions at the sentence level, the polarity of a document is ascribed to a particular group of words, however when classifying sentiments at the aspect level, the emphasis is on identifying the numerous components that make up a text. Finding out if the document as a whole or the topic being covered is typically positive or negative is one of the main objectives. This contribution analyses numerous feature-based opinion classification methodologies.