Library Open Repository
Expert Software Support for Ad-hoc Teams - Enabling Online Interaction Rules
Kildare, RA and Williams, R and Hartnett, J (2004) Expert Software Support for Ad-hoc Teams - Enabling Online Interaction Rules. In: Artificial Intelligence in Science and Technology, AISAT-2004, 21-25 November 2004, Hobart, Tasmania, Australia.
AISAT04Final.pdf | Download (171kB)
Available under University of Tasmania Standard License.
An expert set of interaction rules may help to guide leaderless, transient teams of individuals that work online and asynchronously. Medical and engineering professionals as well as students fall into this category. Sufficient evidence exists to treat such teams as complex systems, to consider team processes as emergent and thus unpredictable. For software to support such teams in a variety of scenarios, the components must cater for the emergence of interaction rules. A software "Moderator" is proposed which gathers, monitors and executes these rules. The standard AI methodology of expert systems will be used. Symbolic rules link the attribute value pairs that software can monitor with language that is meaningful to the team members. The reuse of rules to form new rules allows for a process of evolution. Multiple Classification Ripple Down Rules (MCRDR) provides a mechanism for team-member control over the evolution of rules. One end result proposed is the collection of the rules developed by similar teams as an expert set of rules for the context in which the teams work. This may fulfil the traditional role of a human expert where one is unlikely to exist.
|Item Type:||Conference or Workshop Item (Paper)|
|Keywords:||Software Moderation of On-line Teams.|
|Date Deposited:||12 Dec 2004|
|Last Modified:||18 Nov 2014 03:10|
|Item Statistics:||View statistics for this item|
Actions (login required)
|Item Control Page|