Using Data Science to Build an End-to-End Recommendation System

June 20, 2018
We get recommendations everyday: Facebook recommends people we should connect with; Amazon recommends products we should buy; and Google Maps recommends routes to take. What all these recommendation systems have in common are data science and modern software development. Recommendation systems are also valuable for companies in industries as diverse as retail, telecommunications, and energy. In a recent engagement, for example, Pivotal data scientists and developers worked with a large energy company to build a machine learning-based product recommendation system to deliver intelligent and targeted product recommendations to customers to increase revenue. In this webinar, Pivotal data scientist Ambarish Joshi will take you step-by-step through the engagement, explaining how he and his Pivotal colleagues worked with the customer to collect and analyze data, develop predictive models, and operationalize the resulting insights and surface them via APIs to customer-facing applications. In addition, you will learn how to: - Apply agile practices to data science and analytics. - Use test-driven development for feature engineering, model scoring, and validating scripts. - Automate data science pipelines using pyspark scripts to generate recommendations. - Apply a microservices-based architecture to integrate product recommendations into mobile applications and call center systems. Presenters: Ambarish Joshi and Jeff Kelly, Pivotal
Previous
Microservices Approaches for Continuous Data Integration
Microservices Approaches for Continuous Data Integration

How can businesses modernize their existing data integration flows? How can they connect a rapidly evolving...

Next Presentation
Next-Generation Spring Data and MongoDB
Next-Generation Spring Data and MongoDB

MongoDB 4.0, scheduled for release in Summer 2018, will add support for multi-document ACID transactions. T...

SpringOne Platform 2019

Learn More