Welcome to Suthaharan Home Page

CSC510 Course Website

Big Data and Machine Learning. Fall-2015.

Tuesday/Thursday, 1400-1515 in Petty Building 303

Dr Shanmugathasan (Shan) Suthaharan FIET

Professor of Computer Science

Department of Computer Science

University of North Carolina at Greensboro

161 Petty Building, Greensboro, NC 27412

TP: 336 256 1122, Fax: 336 256 0439

Email: s_suthah@uncg.edu, LinkedIn Profile

My Book - Springer US

Machine learning models and algorithms for big data classification - thinking with examples for effective learning.

கலாநிதி சன்முகதாசன் சுதாகரன்

Dr Suthaharan's research is focused on the characterization and detection of environmental events, the exploration of machine learning techniques, and the development of advanced statistical and computational techniques to discover relevant signatures and detect emerging events from structured and unstructured big data and the environment. Five important definitions that play significant role in his research are presented below for improved readability:

Environment defines the space that deploys modern and emerging technologies and monitors the activities occurring in the space to collect information about the activities.

Event defines an action, a behavior, or an object that subsists in the environment, monitored by a system. Events must be typified and detected to build a secure and reliable environment.

Signature defines the distinguishable measurements or patterns that can uniquely describe events. Signatures must be discovered and analyzed to characterize events for detection.

Descriptive Model defines the measure that helps determine the relationships between the events so they can be classified or grouped for detection.

Predictive Model defines the measure that helps predict the behavior of a single event by using the descriptive knowledge that was learned from the past experience.

The goal of his research is to build a secure and reliable environment using modern and emerging techniques and technologies. To achieve this goal, the meaning of environmental events that help characterize and detect regular and abnormal events (classification) must be understood. In an environment, events may be captured using multiple technologies such as wireless sensor networks, distributed file systems, digital cameras, and mobile devices. These technologies may produce core data units such as measurements, images, sound, and videos that may be used to characterize and detect environmental events. Many applications require signature discovery and event detection techniques to extract information from big data, and build a secure and reliable environment.

Research Activities