The Analytics' Edge-M-

In the big data landscape, it becomes imperative to understand what tools are available to gather data, to subsequently aggregate data into information, and how to use this information to make better decisions. To this end, we discuss approaches to descriptive and predictive analytics.

Descriptive analytics refers to the preparation, exploration, and visualization of historical data in order to identify patterns or meaning. It is a prerequisite for predictive analytics, which refers to the ability of discerning patterns from past and current data to prediction of future events. In this course, we focus on algorithms to tackle classification and regression problems as well as unsupervised learning techniques such as clustering and dimensionality reduction. Examples include healthcare applications such as heart attack pattern detection and financial applications such as fraud detection in regards to credit card usage. The analyses will be carried out in R. R is an open source programming language and software environment for statistical computing and graphics.

Keywords:

Descriptive Analytics: data wrangling, visualization of data patterns;

Predictive Analytics: regression techniques, tree‑based regression, extreme gradient boosting, neural networks, clustering methods, and dimensionality reduction