Building Predictive Analytics Module for E-Commerce Platform Share: InData Labs Score Awaiting client review n/a Date Published 11 March 2020 Reading Time 1-Minute Read InData Labs was challenged to create a custom predictive analytics & recommendation system to enhance the client’s existing platform and improve sales. The Client The client is an e-commerce provider who cooperates with more than 50 brands needed a solution for giving single-brand recommendations. The business goal was to use machine learning (ML) to increase sales, revamp customer experience, attract new clients, and retain loyal users of the online platform. Challenge InData Labs was challenged to create a custom predictive analytics & recommendation system to enhance the existing platform and also to improve sales. Solution We based our development approach on a collaborative filtering technique based on matrix factorization used in the recommender system. We used the confidence metric to train the model to emphasize items purchased several times over items purchased only once. We created a model that addresses all the client’s business needs. Now the model is used for completing multiple tasks to: Make the search process highly personalized Automating routine tasks of shop assistants Ensure excellent online shopping experience Boost customer loyalty Technologies and Tools Python, Scikit-learn, Implicit, Docker