Abstract:
The Smart Feed News Aggregator is a web-based application designed to deliver personalized news recommendations by aggregating articles from various reliable sources. In an era where users are overwhelmed with information, this project aims to enhance user engagement by providing tailored content that aligns with individual preferences and historical interactions. The system utilizes advanced machine learning techniques, specifically Neural Collaborative Filtering (NCF) and content-based filtering, to create a hybrid recommendation model. NCF analyzes user-item interactions to predict click-through rates, while contentbased filtering employs BERT to generate embeddings from article titles and abstracts, capturing their semantic meaning. This allows the system to recommend similar articles based on users’ past behavior. With a user-friendly interface, the application enables users to create profiles, manage preferences, and access both general and personalized news feeds. By continuously adapting recommendations based on user interactions, the Smart Feed News Aggregator aims to improve the overall news consumption experience, ensuring that users receive relevant, timely, and engaging content. This project highlights the significance of personalized news delivery in fostering user satisfaction and retention in a digital landscape.
