A location based novel recommender framework of user interest through data categorization

Abstract

An efficient location-based recommender system in the digital era enables new dimensions of business and service opportunities to all its stake holders. The data from a particular geographical location will be collected according to their lifestyle and categorized. Increasing the use of social media, overwhelming digital information flow in every user movement will create a new array of dimensions for an online or offline business. The location-based recommender system effectively serves its users in the best possible way as per their interests. The user requirements of their interests will be served effectively as per their interests and location. The location-based recommender system will build new avenues in the business domain. This paper presents the introduction to geographical location-based recommender systems—a detailed review of various recommender systems based on geographical regions. Identify the parameters to categorize the users and a proposed model for location-based recommender systems.One of the most widely used approach for providing personalized services to customers is the Collaborative filtering technique. Similarity between users based on user-item rating matrix is the key essence of this approach for providing recommendations to users. Similarity algorithms like cosine, manhatten, euclidean etc. play a prominent role in determining similarity between users. Our proposed paper presents an innovative approach for enhancing prediction using machine learning for unrated items. Along with local context information global preference of user behavior is also taken into account. Experiments for our proposed system are carried on MovieLens dataset using machine learning. Performance enhancement can be seen in our proposed model.