Sentimental Analysis On Voice Based Reviews Using Fuzzy Logic
D. Jayashree, O. Pandithurai, Smisha Prasad, Anupama Suresh, S. Vigneshwaran · 2021 International Conference on Advancements in Electrical, Electronics, Communication, Computing and Automation (ICAECA) · 2021
E-commerce websites becoming the future of the market, manufacturers have to keep their products updated to compete among different brands. Improvements according to their customer’s needs help them to stay active in the competition. To improve their product, reviews are an important medium in achieving it. Customers also prefer to buy products online depending on the ratings and reviews of a product. Reviews that are written may not deliver all the emotions a customer needs to convey but in the case of a speech review, it almost delivers the information that the customer needs to convey. In this process, the customer submits the review as an audio file which is then converted into a text file using Natural language Processing. The user-driven reviews of a product are processed using sentimental analysis. The proposed system involves fuzzy logic for classifying the reviews into positive, negative, and neutral reviews and for assigning weightage to the words.