I study the problem of how to incorporate meaningful context into computer understanding of human language. A key aspect of that is the style that is expressed in a text, as a function of the text’s author, purpose, feeling, medium of expression, and more. I’ve done a fair bit of research on the analysis of language style, for applications such as forensic analysis, counterterrorism, sociolinguistics, and humanities computing, and have recently cofounded Subtext3, which is commercializing some of this work for large-scale text analytics applications.
This work brings together a wide range of disciplines, including machine learning, natural language processing, sociolinguistics, psychology, and literary theory. This breadth helps to keep things interesting.
Over the last 20 years, I have worked in various areas of AI, including mobile robotics, computer vision, machine learning, and natural language processing, and have done consultation work on algorithm and software development for linguistic and other applications.