An Efficient Recommender System Using Collaborative Correlation Methodology

Recommendation engines are the insights of every frontline application. They are working as the ideologists in determining the data to be shown for each specific user with respective of their interested categories. Current recommendation systems provide insights determining user interests but with a drawback of more amount of time taken for a recommendation, involving large user data to determine the user categories, also at some point of time giving the recommendations out of boundaries and also considering the domain of the user interest which is very vast. The proposed Collaborative Correlative Filtering (CCF) methodology involves the powerful integration of collaborative filtering with correlation factors. The combinatory method improves recommendation and improvises the existing system with a recommendation system involving the least amount of data, less space, and least domain specificity. The efficiency of the system is measured by metrics such as precision, recall, and F1-score to determine how relevant the data has been given by the proposed recommendation system.