Classification-based Collaborative filtering: A Machine Learning Recommendation System

Social networking platforms like, Twitter, Face book etc., have now emerged as a major forum for the application of Recommended Systems (RSs). Of course these famous sites are taken as the main source of people-related knowledge and thus to be a great choice for exploiting modern and creative approaches to the recommendation, backing the old methods, in order to improve accuracy It was thought that helping users cope with the issue of data overload was the original role of information retrieval systems or search engines, but what separates recommended systems from  the existing search engines is the requirements of personalized useful and interesting. The "intelligence" aspect is what suggests more interesting and useful. Intelligence is one of the main routes of personalization to know the interests of the user, anticipate the unknown favorites of the user, and eventually provide suggestions by matching the question and the content beyond a basic search. This article provides simple approaches to Recommendation Systems, provides recommendation for similar items based on the correlation and classification methods of machine learning to build a collaborative filtering system by making use of Logistic Regression model.