FEROM (Feature extraction and refinement) using genetic algorithm

It is an area of text classification which continues gives contribution in research field and it analyse and classify the user generated data like reviews, blogs and comments etc. In opinion mining customer reviews generally contain the product opinions of many customers expressed in various forms consisting natural sentences. A common phenomenon in natural sentence-based customer reviews is that people generally do not express their opinions in a simple way such as "This Smartphone is good", but present them using features of the product such as "the battery life of this Smartphone is too short". The main purpose of this paper is to search for opinions about features of a target product from a collection of customer review data. Genetic algorithm analyses the opinion sentences and determines the orientations of the opinions also provide an exact summary to the user. This algorithm is used to categorize the user opinions into positive, negative and neutral classes and the categorization of text is called polarity of text.