This session will present an introduction to the emerging and evolving topics of “Big Data” and predictive analytics particularly as they apply to higher education and the use of data to improve student persistence and outcomes. An overview of Big Data, an introduction to the Predictive Analytics Reporting (PAR) Framework, and an institution’s perspective on these issues along with their implementation of analytics will be presented.Predictive analytics techniques, such as neural networks and decision trees, help anticipate behavior and events.
Predictive Analytics to predict user intentions towards a certain product, or category on an e-commerce site, based on historical interactions with a website is very useful for advertising, recommendation engines and for demand forecasting. Clickstream data can be used to quantify search and purchase behaviors. The goal of the Predictive Analytics is to identify activity patterns of the users that lead to purchase decisions and develop a model to mimic user behavior. We will use logistic regression and deep learning/neural networks to predict user activity on the website based on clickstream data.