Joshua B. Gordon

Data mining to support human-machine dialogue for autonomous agents

Next-generation autonomous agents will be expected to converse with people to achieve their mutual goals. Human-machine dialogue, however, is challenged by noisy acoustic data, and by peoples preference for more natural interaction. This paper describes an ambitious project that embeds human subjects in a spoken dialogue system. It collects a rich and novel data set, including spoken dialogue, human behavior, and system features. During data collection, subjects were restricted to the same databases, action choices, and noisy automated speech recognition output as a spoken dialogue system. This paper mines that data to learn how people manage the problems that arise during dialogue under such restrictions. Two different approaches to successful, goal-directed dialogue are identified this way, from which supervised learning can predict appropriate dialogue choices. The resultant models can then be incorporated into an autonomous agent that seeks to assist its user.

Evaluation of the emerging scholars program at Columbia University (abstract only)

The Columbia Emerging Scholars Program (CESP) in Computer Science is a Peer Led Team Learning (PLTL) approach to bringing undergraduates new to the discipline together with peer mentors to work on computational problems, and to expose them to the broad array of disciplines within computer science. CESP demonstrates that computer science is necessarily a collaborative activity that focuses more on problem solving and algorithmic thinking than on programming. In spring 2012 the computer science department at Columbia University completed the 9th iteration of CESP, with 104 women and 36 men having completed the program to date. Female enrollment at Columbia during the past four years has increased from 9% to 23%, but did CESP play a part in this increase? This poster presents our evaluation data, which indicates that CESP increased enrollment in the computer science major, especially for women. Students who took CESP along with the introduction to computer programming course in 2009-10 were three times more likely to major in computer science the following year than the students who took introduction to programming without CESP. 47% of CESP students subsequently chose the computer science major. In addition, survey results indicated that a large majority of students intended to take another computer science course, were enthusiastic about the program, and found the workshop topics exciting and engaging. Participants reported that they learned more about computer science in CESP, and would recommend CESP to others.