IBM Rolls Out Data Mining Software

IBM's India Research Laboratory (IBM IRL) have developed data mining software ProAct, which extracts and delivers business insights from the information gathered by companies during customer service calls and other communications.

The launch forms part of IBM's cross-company Information on Demand strategy and is being tested at third-party call centers - it is already running in IBM's own outsourcing centres. The product is said to mine both structured and unstructured data, giving users feedback not only from sources such as agent and product databases, but also from phone call transcripts and emails to call centres.

ProAct technology combines advanced text analytics, machine learning and natural language processing techniques, and is based on the Unstructured Information Management Analysis (UIMA) framework that IBM contributed to the open source Apache Software Foundation in 2006. The firm uses UIMA for text analysis, extraction and concept search capabilities in other search software products, including OmniFind Enterprise Edition, OmniFind Analytics Edition, and OmniFind Yahoo! Edition.

The lab is also working on a variety of other projects including pervasive user interface technologies for developing countries; context-oriented information integration; and software allowing the automated evaluation of a person's spoken language skills - such as pronunciation, grammar and comprehension - for use in 'hiring or training agents in BPO industries'.

The Laboratory home page is at www.research.ibm.com/irl.

All articles 2006-23 written and edited by Mel Crowther and/or Nick Thomas, 2024- by Nick Thomas, unless otherwise stated.