Generic Recommendation Engine using Hybrid Filtering Model

Recommendation system provides the facility to understand a person's taste and find new, desirable content for them automatically based on the pattern between their likes and rating of different items. Recommendation systems are mainly employed in applications such as online market, which works with big data. Performing data mining on big data is a tedious task due to its distributed nature and enormity. There are humanely overwhelming number of items for us to inspect, evaluate and choose from. This poses a huge challenge, since overwhelming the customers with huge catalog of items out of which the major portion of items are unrelated to user preferences.There is an imminent need for a recommendation system that eases the process of choosing products by the user and thereby enriching the user experience. To overcome this problem, a recommendation system that uses multiple ML algorithms, a hybrid version of content based filtering and collaborative item-item filtering algorithm is implemented so as to achieve better accuracy in recommendations. The project is aimed to result in a generic recommendation engine suitable for using with any type of items irrespective of domain and datasets.