Design of NLP technique fore-customer review
Anjali Dadhich1, Blessy Thankachan2
1Anjali Dadhich, master’s degree in Computer Science from Banasthali Vidhyapeeth, Rajasthan, India.
2Dr. Blessy Thankachan, Ph.D. field of Software engineering, University of Rajasthan, India.
Manuscript received on August 01, 2020. | Revised Manuscript received on August 05, 2020. | Manuscript published on September 30, 2020. | PP: 653-655 | Volume-9 Issue-3, September 2020. | Retrieval Number: 100.1/ijrte.C4614099320 | DOI: 10.35940/ijrte.C4614.099320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: With the passage of time and the growth of ecommercea new web world needs to be built their users can share their ideas and opinions differently domains.There are thousands of websites that sell these various products. The quick growth in the number of reviews and their availability and the arrival of rich reviews for rich products for sale online, the right choice for many products has been difficult for users. Consumers will soon be able to verify the authenticity and quality of the products. What better way is there to ask people who have already bought the product? That’s where customer reviews come from. What’s worse is the popular products with thousands of updates — we don’t have the time or the patience to read all of them thousands. Therefore, our app simplifies this task by analysing and summarizing all the reviews that will help the user determine what other consumers have experienced in purchasing this product. This function focuses on mining updates from websites like Amazon, allowing the user to write freely to view. Automatically removes updates from websites. It also uses algorithms such as the Naïve Bayes classifier, Logistic Regression and SentiWordNet algorithm to classify reviews as good and bad reviews. Finally, we used quality metric parameters to measure the performance of each algo.
Keywords: Sentiment Analysis, Naïve Bayes classifier, Logistic Regression, Senti Word Net, Opinion Mining.