MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202621081465 A) filed by Symbiosis International Deemed University on July 01, 2026, for A Lexicon-Based Sentiment Analysis System Using Sentiwordnet For E-Commerce Product Reviews.
Inventors include Dr. Aniket K. Shahade; and Dr. Priyanka V. Deshmukh.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: ABSTRACT A LEXICON-BASED SENTIMENT ANALYSIS SYSTEM USING SENTIWORDNET FOR E-COMMERCE PRODUCT REVIEWS The present invention discloses a lexicon-based sentiment analysis system (100) for classifying e-commerce product reviews without labeled training data. The system (100) comprises a text preprocessing module (120) with tokenization (122), cleaning (124), lemmatization (126), POS tagging (128), and negation handling (129), a sentiment scoring module (130) with SentiWordNet lookup (132), word-level polarity computation (134) with negation reversal, and review-level aggregation (136), a threshold- based classification module (140), and an evaluation module (150). The system achieves 79 percent accuracy, correctly classifying 89 percent of negative and 78 percent of positive reviews, providing a lightweight interpretable solution for e-commerce applications. [
Disclaimer: Curated by HT Syndication.
