Findability Day 2016 - Big data analytics and machine learning

This document discusses leveraging machine learning and big data analytics. It outlines an analytical pipeline that includes data acquisition, data munging, exploratory data analysis, model building, model improvement, validation, and real-time processing. A case study is presented on using these techniques to predict when to scrap parts in an assembly line to reduce costs. Key takeaways are that machine learning can find hidden insights in historical big data, models derived from this can be applied to real-time event processing without redevelopment, and this enables automated actions based on predictive analytics.