Dartmouth Events

Computer Science Colloquium talk by Dr. Jamie Macbeth, MIT

Talking Users out of Harmful Actions

Wednesday, January 8, 2014
Steele 006
Intended Audience(s): Public
Categories: Lectures & Seminars

Talking Users out of Harmful Actions

An important body of research in human-computer interaction creates richer and more effective interactions between users and systems by interpreting natural language input, both in voice and text form--intelligent personal assistant applications such as Apple's Siri and Google Now are important successes in this domain.  In this talk I will discuss the development of systems which use formal or natural language as an essential element of the interaction in two challenging and complex domains; the common goal in both is to prevent users and operators from performing sequences of actions that lead to hazardous system states or harm other users.

First, I will discuss work to develop systems in safety- and mission-critical domains that require users to follow formal procedures composed by administrators.  Specifically, a procedural adherence metric is developed from natural language processing methods and used as a tool to design nuclear power plant control rooms that strike the right balance between software automation and manual human control.  Second, I will discuss the development of intelligent interfaces to social media to aid in cyberbullying prevention, using automated narrative analyses of text communications between users to trigger selective interventions and prevent negative outcomes.  In both lines of work, the proposed approach attempts to penetrate through patterns found at the syntactic layers of expressions to a deeper semantic-conceptual layer of the language by collecting and applying commonsense knowledge structures.

Jamie Macbeth is a research affiliate with the Humans and Automation Lab at MIT and holds a Ph.D. in Computer Science from UCLA. He has a wide range of experience in software engineering and UI design: work on listening and communication training for the hearing impaired at Neurotone Inc., on professional audio mixing interfaces at Euphonix (now a division of Avid), and on digital media distribution at Sony Music Entertainment and Liquid Digital Media.  Prior to that he earned an M.S. in Physics from Stanford University and a B.S. in Physics from Brown University.  His broad interests include human-computer interaction, human factors of software engineering, and natural language understanding.

For more information, contact:
Shannon Holly Stearne

Events are free and open to the public unless otherwise noted.