• Rashmi Deshpande
• Shivam Ghate
• Anisha Bharuka
• Swapnil Sonawane
• Harshada Patil

Recommendation systems have emerged rapidly in the past decade. Recommendation of products to attract customers that meet their requirements is very important for the vendors to survive in the global market. The approach proposed in this paper is novel and serves as a better alternative to rate a product based on its technical specification by analyzing large number of user reviews which are extracted dynamically from several top e-commerce websites. This avoids the need, for the user, to search for opinions and comments online before making a purchase. The proposed approach in this study extracts specification list like battery, processor, camera etc. and customer reviews for a user specified product from different websites and identifies crucial terms corresponding to the technical features of the product in the review to determine polarity of the feature and classifying it under the specification list. Each specification is assigned a score based on polarity i.e. positive/negative feedback. Overall product rate is calculated by aggregating the score specific to individual features. This approach is very useful for those customer swhotar get at specific features in a product.

Preprocessing, Feature Extraction, Classification using Support Vector Machine, Machine Learning

"Real time product price monitoring and analysis for E-commerce website.", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 5, page no.c964-c967, May-2021, Available :http://www.jetir.org/papers/JETIR2105377.pdf

"Real time product price monitoring and analysis for E-commerce website.", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.8, Issue 5, page no. ppc964-c967, May-2021, Available at : http://www.jetir.org/papers/JETIR2105377.pdf