Developing and Deploying a
Social Listening System

While Machine Learning (ML) is a powerful technology for building applications, the transition from prototype to production of an ML application proves challenging. The challenge lies in the ML development lifecycle, including but not limited to the following steps: defining business goals, acquiring data, developing, and deploying models. In this project, we build an ML management platform to streamline that cycle. We apply MLOps practices into developing our application, leveraging the automation pipeline to orchestrate the ML workflows. Also, we utilize open-source tools and Kubernetes technology to make the system platform-independent. For the usage demonstration, we adopt the proposed solution to build a social listening application. The app leverages the sentiment analysis technique to analyze product reviews on e-commerce sites. The main contribution of this project is a platform to manage the ML development lifecycle and social listening application.