A New Approach to Improve Online Customer Review Analysis by a Sentence Level Using Vector Similarity Related Text Extraction

Dr.A. Suriya

Sentiment analysis is one of the Natural Language Processing (NLP) technique that consists in extracting emotions related to some raw texts. It’s usually used on social media posts and customer reviews to automatically understand if some users are positive or negative and why. The text mining is the process of converting unstructured text data into an extensive collection of analytics and visualization and model building. Online reviews of e-commerce giants are a paradigm that can be used to arrive at profitable results. Online reviews allow you to purchase information about the product, including its quality, performance, and recommendations, as it gives the buyer a clear image of the product. Favorable reviews by analyzing customer needs, manufacturers have realized the benefits of invisible energy online reviews. Reviewing proposed method sentences using text extraction from customers. They give positive and negative feedback from a variety of penalties, from which the syntax is extracted and analyzed.This work improves the text mining method using Vector Similarity RelatedText Extraction (VSRTE).This methodology not only classified the text into positive and negative sentiment but have also included emotions of customer’s reactions like usual, angry, and sentiments. This text extraction mining can help assess product quality to enable better decision-making for consumers and also predict their customer behavior.

Volume 11 | 12-Special Issue

Pages: 1151-1162

DOI: 10.5373/JARDCS/V11SP12/20193322