Research

Web Intelligence and Text Mining

The Text Mining team is dedicated to developing mechanisms for extracting knowledge hidden in unstructured text and integrating it to a causal and predictive analytics framework to derive Enterprise Intelligence. Our current projects include:

Enterprise Web Intelligence

The Web continues to play an important role in the smooth functioning of an enterprise be it due to the continuous flow of consumer comments on the social media, or the News or the expert opinions on the blogs and discussion forums. While gathering the right information at the right time continues to be important in this context, it is also important to judge the relevance and criticality of the content gathered. Though there are many tools today to gather and present all relevant information to the end-user, very little work is focused on integrating the information with Enterprise content and develop analytical techniques to measure the possible impact of the information. The focus of this group is to identify, categorize and measure the possible impact of different world “events” on an enterprise. The group uses machine learning techniques to identify and categorize relevant events from unstructured data streams obtained from the Web. Reasoning frameworks are being explored to model and measure impacts of different kinds of events.

eMail Analytics

Emails provide the back-bone of enterprise communications and are rich repositories of knowledge containing both explicit and implicit information about all ongoing activities. Though email clients have improved in email organization and spam detection, very little attention has gone into trapping the enterprise knowledge implicit in emails. This includes knowledge about expertise, communication patterns, project execution etc. and can yield rich benefits if extracted and woven into a causal analytics framework. This group focuses on analyzing noisy unstructured content in emails using statistical text processing, Natural Language Processing, machine learning and statistics and graphical techniques.

This group is headed by Dr. Gautam Shroff

 

Reach Us.

Share