IJCAI-2007 Workshop on Analytics for Noisy Unstructured Text Data

Panel Discussion

Noisy Text Analytics: An Exercise in Futility?

Human language is, as we know, hard for machines to handle. Add to this the complexities introduced when text is degraded by transcription errors, spelling mistakes, extreme abbreviations, informal writing, missing punctuation, and ever-changing style conventions arising from new communications paradigms, and the situation seems almost hopeless. As researchers working in this arena, is our optimism warranted, or are we blinded by past successes in easier domains?

We are fortunate to be able to bring together a panel of leading experts to debate this provocative issue at the AND-07 workshop. The list of participants includes:

• Daniel Lopresti (moderator), Lehigh University, Bethelehem, PA, USA.
• Sreeram Balakrishnan, IBM Research, New Delhi, India.
• Hwee Tou Ng, National University of Singapore, Singapore.
• Rohini Srihari, Janya Inc., USA.

We look forward to much interesting discussion as a fitting way to conclude our one-day workshop on Analytics for Noisy Unstructured Text Data.