Clustering and Predicting Driving Violations Using Web-Enabled Big Data Techniques

When quintillion bytes of data on a multitude of topics is being generated each day creating Big data in size and in scope, the need for analyzing such voluminous data, extract meaning from it and providing a visualization is also increasing. Driving violations is one of the topics that have been recorded over multiple years. Several studies have been conducted to predict driver behavior using simulations and other tools such as built-in sensors in the vehicles. This research activity focuses on the design of an interactive Big data web application to analyze a given dataset using techniques such as cluster analysis and predict driving violations based on available demographics. The rest of the paper describes the suite of technologies for Big data analytics that facilitated this development and the implications of this study.