Efficient Behavior Prediction Based on User Events

In 2020 we have witnessed the dawn of machine learning enabled user experience. Now can predict how users will use an application. Research progressed beyond recommendations, and are ready to events. Whenever a human interacts with system, events dispatched. They be as simple mouse click on menu item or more complex, such buying product from eCommerce site. Collaborative filtering (CF) has proven excellent approach Because each generate many events, this inevitably leads vast number in dataset. Unfortunately, operation time CF increases exponentially increase data-points. This paper presents generalized reduce dataset’s size without compromising prediction accuracy. Our solution transformed dataset containing over 20 million (20,692,840 rows) into sparse matrix about 7 minutes (434.08 s). We used train neural network accurately