Advanced Recommendation System Using Sentiment Analysis

EasyChair Preprint 6503

Abstract

Recommendation Systems are the heart of ECommerce. They help consumers with the right contrast of products or items thereby increasing the revenue of the company. The current recommendation systems generally use click-based recommendation models wherein the products, movies, songs or any item, are registered and they are recommended respectively. Some platforms even use the customer star ratings and reviews to recommend the products. But the limitation of that is professional reviewers and general public commonly post their reviews on the product on E-commerce websites. The purpose of Advanced Recommendation System using Sentiment Analysis is to elevate the recommendations provided to the users by gathering the users’ sentiments. We are planning to use the sentiments of the people to further provide an advanced understanding of the people’s view on the product and suggest that to the future customers accordingly. Our paper will use click-based recommendation system as well as sentiment analysis on product reviews to look for the general sentiment of the item or the product and finally recommend the product to a consumer

Keyphrases: Natural Language Processing, Principal Component Analysis, Recommendation, Recommendation System, Sentiment, advanced, advanced recommendation system, analysis, collaborative filtering, data analysis, data pre-processing, eCommerce, feature engineering, machine learning, recommendation algorithm

@booklet{EasyChair:6503,
  author    = {Jaineel Mamtora and Suraj Chatterjee and Shawn Almeida and Vandana Patil},
  title     = {Advanced Recommendation System Using Sentiment Analysis},
  howpublished = {EasyChair Preprint 6503},
  year      = {EasyChair, 2021}}