Skip to main content

Information Management


Information Management

Overview Research People Publications
Massive Analytic Solutions for Telecom using Hadoop

Telecommunication companies face the challenge of managing and exploiting massive amounts of customer and network operations data that they accumulate at an ever-increasing rate. Data drive business decisions. Telecommunication companies rely on organized data to offer continuous improvement in cell coverage, optimally route calls, and provide responsive and personal customer care services. Companies that succeed in turning data into information and products can gain important business advantage in an intensely competitive industry.

Hadoop is a promising infrastructure for data-intensive distributed analytics using inexpensive commodity hardware. However, its focus is on scalable processing of large amounts of input data with shallow analytics, and it ignores resource data requirements for deep analytics. Many large-scale data analytic tasks depend on massive amounts of analytic resource data (for example, statistical models), and it is difficult to co-locate or migrate them.

This project focuses on problems related to challenges in using Hadoop over data generated in the telecommunications industry. In particular, we look at how existing data models, storage techniques, access methods can be adapted to work with Hadoop and how to leverage these adaptations for performing analysis.

Researchers: Ullas Nambiar, Himanshu Gupta, Tanveer Faruquie and Mukesh Mohania

CRM Framework for Business Analytics

Predictive analytics (i.e., propensity to buy and churn) forms a cornerstone of the customer care and insight strategy. Delivering predictive analytics in a scalable fashion requires software tooling to minimize reliance on scarce highly-skilled resources. Tooling is also needed to reduce the time and effort required to access, extract, and transform data as input to predictive analyses. In this project, we aim to enhance the customer relationship management (CRM) framework for predictive analytics by leveraging the Cognos Adaptive Analytics Framework and the InfoSphere Design Studio to assist in mapping customer databases to IFW Industry Models.

Researchers: Prasad Deshpande, Sumit Negi, Ramakrishnan Kannan, Anup K Chalamala, Soujanya Soni, Bipen K Telkar

Efficient Processing of Large XML Data

The use of large XML documents, exceeding 1 GB, is on the rise. Existing software are not capable of handling such large XML documents in an efficient manner. The project aims to design techniques for efficient processing of large XML documents in parallel on multiple machines.

Researchers: Manish Bhide, Manoj Agarwal, Srinivas Kiran Mittapalli

System T Enhancements

System T is a framework for rule-based information extraction in which the rules are expressed declaratively using a SQL like language called AQL. In this project, we aim to enhance System T functionality by adding new features to AQL such as UDF support, recursion, among others.

Researchers: Prasad Deshpande


Page help