Products And Movie Recommendation System For Social Networking Sites

Recommendation systems are an integral part of information filtering system in data science, that are widely used in order to identify the pattern a user would likely choose on the basis of the previous choices of the user as well as from studying the pattern in which others have chosen. For a fact, the recommendation can never be a cent percent correct at providing recommendations to the user but can be close enough to please them to a certain extent. Thus, the same is widely used in the industries these days to get higher profit and have a good hold in the market. The data scientists of every company design some algorithm that studies the information from the social network and clusters the data. There can be a single algorithm for classification like k-Means clustering or Hidden Markov model or can be done by bagging and boosting techniques. With this technique of displaying the movies or products into the profile of a particular customer, they not only increase their business but also enhances the customer experiences but there are several issues related to the standard techniques like the cold start problem, shrill attack, etc. thereby increasing the scope of research in this field. This work deals with both Collaborative Filtering and Content-Based Filtering to form a product and movie recommendation system for the social networking sites that shows the effectiveness of collaborative filtering and portrays the challenges faced by content-based filtering.