Multi - Category & Multi - Criteria Recommendation System using Collaborative Based Filtering

Recommendations Systems have become one of the most popular application of data science today. It predicts or offers products to customers based on their past browsing history or purchases. Although a lot of effort, research and time has been spent on recommendation engines, we are yet to truly unlock their potential. At the core, a recommender system employs a machine learning algorithm whose job is to predict user's ratings for a particular entity. Through this project, we are employing a multi category recommendation system which will give the user recommendations across different categories based on the user data of multiple categories consisting of different attributes. The concept of K - Nearest Neighbor Algorithm is implemented to derive the similarity of unknown entities or users based on past ratings of a particular entity. The implementation is carried out using JavaScript in Node, thereby extending the capabilities of Collaborative based filtering Algorithm to multiple categories.