The recommender system: operations research in OTT-platforms

Due to the pandemic, OTT platforms are facing a sudden burst in viewership and are meetingan all-time high viewership. Every OTT platform uses different types of recommender systems to suggest a user about a title the user can watch, and thus it provides a personalized environment for each user. Despite using such extensive and advanced techniques to recommend a movie/show, OTT platforms fail to recommend appropriate titles of various genres that the user can watch. Therefore this paper identifies this gap and proposes a recommender system based on user's past preferences of genres and viewership history. The recommender system is based on the operations research technique called the Hungarian Assignment method. This model is yet to be tested using real-time data because of the limited access to backend data. However, it is established that it contributes to a new category of recommender systems.