| Summary: | Human help is inherently personalized, customized, and delivered exactly when needed. Current help technology being attempted in the software industry and under the aegis of Intelligent Tutoring System (ITS) is insufficient to duplicate the sophistication and depth of human help. An ideal system could attempt to store or generate vast amounts of situated and individualized help information and to provide fast and structured access to it. Yet, such an approach introduces umnanageable computational complexity, inadequate failure handling, deficient self-improvement, and inflexible generalization. This thesis explores techniques to support the human helper who has been embedded in a human-computer help environment. It investigates the design of a help system that is capable of acquiring context information, making useful knowledge-based help responses, and ensuring delivery of help within acceptable time limits. In its full generality, the help system must be intelligent, informed of the user, informed of the tasks involved, informed of the information used, informed of the collaborative interactions, and informed of the help resources. Such an intensely demanding requirements of an ideal help system can be managed with the 'human-in-the-loop' approach in the domain of 'peer help networks' using 'partially-automated ' techniques. This thesis identifies a comprehensive design for a framework that adequately supports peer helpers in a peer help network. The contribution of the thesis is fourfold. First, the thesis investigates research methodologies in help systems to design, develop, and deploy a generic, prototypical help system that can deliver context-specific, personalized, and just-in-time help to a human helper (rather than to the helpee directly). Second, the thesis employs the human-in-the-loop approach to counter shortages of context information, occasional inability to match a help request to an appropriate help response, and challenges in meeting time limitations. Third, the thesis proposes partially-automated mixed-initiative tools, as opposed to fully-automated tools. Fourth, the thesis integrates generic tutorial strategies and modern educational technology in a help system framework. (Abstract shortened by UMI.)
|
|---|
