Paper Key : IRJ************845
Author: Ankur Tambe,Aniket Pate,Atharva Sankhe,Prof. Ushma Shah
Date Published: 04 Jul 2024
Abstract
Recommendation systems are vital in domains like entertainment, social networks, health, education, travel, food, and tourism for delivering personalized content that enhances user engagement. This study examines the effectiveness of content-based filtering and cosine similarity in identifying content similarities and recommending movies. By using metrics such as cosine similarity and Euclidean distance, the system measures genre and movie similarities to provide tailored recommendations. Additionally, the paper presents a content-based recommender system for online stores that adapts to user preferences based on viewed content, addressing the cold start problem and improving relevance. Hybrid recommender systems, which combine multiple techniques, are also explored to boost individual approach performance through various hybridization models. The paper categorizes existing work based on these models and machine learning algorithms, offering a comprehensive overview of state-of-the-art systems. Furthermore, a housing recommendation model integrating cosine similarity with deep learning in a grid environment is proposed. This model utilizes extensive housing data and user feedback to create and refine a recommendation model, enhancing accuracy and user satisfaction.
DOI Requested