A Bayesian methodology for semi-automated task analysis

This research proposes a new task analysis methodology that combines the fuzzy Bayesian model with classic task analysis methods to develop a semi-automated task analysis tool to better help traditional task analysts identify subtasks. We hypothesize that this approach could help task analysts identify activity units performed by the call center agent. The term activity units, in our study, represent the subtasks the agents perform during a remote troubleshooting process. We also investigate whether this tool could help predict the activity units as well. An effort-intensive field-based data collection for the call center's naturalistic decision making's environment was accomplished. A human expert and an additional 18 Purdue students participated in the validation of the assigned subtasks. The machine learning tool's performance was then examined. The preliminary results support our hypotheses that the fuzzy Bayesian based tool is able to learn and predict subtask categories from the agent/customer narrative telephone conversations.