Abstract

In today’s information-rich era it has becoming increasingly easily to collect, store and access large collections of unstructured text-based documents. The challenge for many companies now is to make sense of a wide variety of data sources such as unstructured customer feedback forms, social media posts, and media reports. Techniques for detecting events, tracking trends and discovering anomalies are important aides for understanding the information contained in a constant stream of text data. The integration of algorithms that perform automatic phrase discovery, topic modeling and sentiment analysis are all critical components for such tasks. This talk will present on-going work within Microsoft to provide these capabilities to our customers on Microsoft’s Azure Machine Learning platform.

Biography

Dr. Timothy J. Hazen is currently a Principal Data Science Manager within Microsoft where he leads a science team that develops natural language processing, image processing and genomic analysis capabilities for the Azure Machine Learning platform. Dr. Hazen has also developed natural language understanding capabilities that are currently deployed within Microsoft’s Bing and Cortana products. Prior to joining Microsoft, Dr. Hazen was a member of the Human Language Technology Group at MIT Lincoln Laboratory (2007-2013) and a Research Scientist at the MIT Computer Science and Artificial Intelligence Laboratory (1998-2007). Dr. Hazen holds S.B., S.M., and Ph.D. degrees in Electrical Engineering and Computer Science from the Massachusetts Institute of Technology.