Recommendation systems help people find the right products and services in tremendous digital landscape. However, many users still struggle to find relevant product because of wide range of products availability. This project aims to develop a recommendation system that uses deep learning and content-based filtering to analyze product reviews. The system will match user queries with product descriptions and retrieve the most relevant product according to the user’s query. Additionally, analyze the reviews of retrieved products and rank according to customer’s feedback. This hybrid approach by analyzing the content and sentiment of reviews, aims to provide accurate and personalized suggestions. This system is not only beneficial in e-commerce but also provides personalized recommendations in diverse fields such as entertainment and education, to further improve decision-making and user satisfaction.
Tools: Google colab, Numpy, Pandas, Gensim, Scikit-learn, Tensor flow.
Department: Department of Mathematics
Project Poster
