Paper Contents
Abstract
ABSTRACTShopping online has become increasingly common, but choosing the right product among countless options can be overwhelming. Most shoppers turn to reviews on platforms like Reddit and YouTube to make informed decisions, yet these reviews are scattered, unstructured, and often buried within lengthy discussions and comments. This makes it impractical for users to manually gather and synthesize information from multiple sources. RecomAI solves this problem by automatically collecting product reviews from Reddit and YouTube through their public APIs and using the Google Gemini Large Language Model (LLM) to transform raw reviews into clear, organized insights. Our system delivers structured pros and cons lists, feature-speci c ratings (such as battery life, camera quality, performance, and pricing), estimated price ranges, and an overall product verdict. The backend is built with FastAPI and asynchronous Python libraries for smooth external API integration, while the frontend uses HTML5, Bootstrap 5, and Tailwind CSS to create a responsive, user-friendly interface that supports both single-product analysis and side-by- side product comparisons. We validated our approach on real consumer electronics and found that RecomAI can produce meaningful, consolidated product insights in just 37 seconds, enabling users to make con dent purchase decisions quickly without spending hours reading through scattered online discussions.
Copyright
Copyright © 2025 Dhanush Reddy B. H, Dhruthilesh n murthy , Akash s. This is an open access article distributed under the Creative Commons Attribution License.
