Virtual Facilitation of Human Group Interactions Employing a State-based Learning Classifier System
Department
Computer Science
Major
Computer Science
Research Advisor
Tauritz, Daniel R.
Advisor's Department
Computer Science
Funding Source
Missouri S&T Opportunities for Undergraduate Research Experiences (OURE) Program
Abstract
Natural human group dynamics sometimes can lead a group down unproductive pathways. A human expert group facilitator may need to intervene to return the group to a productive workflow. Unfortunately, human expert group facilitators are scarce and prohibitively expensive. The circumstances that lead a group astray can be translated into a state-based decision graph with sets of matching rules along with an appropriate intervention for each situation. The goal of this project is to successfully develop a Virtual Facilitator software system that employs a state-based learning classifier system to evolve increasingly higher quality matching rules and conversation models based on crowd sourced feedback. By doing so, ubiquitous access to a low cost, high quality means of group facilitation will become a feasible reality.
Biography
Jeffery will be receiving a Bachelor of Computer Science in the spring 2011 after spending three years as an undergraduate. During the summer of 2010, he interned at Sandia National Laboratories in the Center for Cyber Defenders. His current research project was previously presented at the Undergraduate Research Day at the Capitol in Jefferson City. In the spring of 2011, Jeffery plans on pursuing a Masters of Computer Science degree with funding from the Sandia National Laboratories Critical Skills Master’s Program.
Research Category
Sciences
Presentation Type
Poster Presentation
Document Type
Poster
Location
Upper Atrium/Hallway
Presentation Date
06 Apr 2011, 9:00 am - 11:45 am
Virtual Facilitation of Human Group Interactions Employing a State-based Learning Classifier System
Upper Atrium/Hallway
Natural human group dynamics sometimes can lead a group down unproductive pathways. A human expert group facilitator may need to intervene to return the group to a productive workflow. Unfortunately, human expert group facilitators are scarce and prohibitively expensive. The circumstances that lead a group astray can be translated into a state-based decision graph with sets of matching rules along with an appropriate intervention for each situation. The goal of this project is to successfully develop a Virtual Facilitator software system that employs a state-based learning classifier system to evolve increasingly higher quality matching rules and conversation models based on crowd sourced feedback. By doing so, ubiquitous access to a low cost, high quality means of group facilitation will become a feasible reality.
