Abstract: Many advanced predictive analysis problems require non-trivial features from large-scale enterprise relational tables with billions of records. How can data scientists discover features on large, disparate data sets and develop insightful features, faster and more efficiently? This talk discusses and demonstrates recent advancements in feature engineering automation (AutoFE) integrated with Apache Spark using dotdata’s Automated feature factory.
Bio: Sharada Narayanan is a Senior Data Scientist at dotData. Currently working in ML and Feature Engineering automation technologies, Sharada has worked on multiple projects with large volume and variety of data which generate value for organizations in the Healthcare, Supply chain and Customer Analytics space. She has a passion for educating organizations new to data science to facilitate strategic and quick success in driving data driven decisions.