Beyond Artificial Intelligence Markup Language (AIML) Rapid prototyping of a Q&A system
The Virtual People Factory (VPF) is suggested as a good solution to rapidly built Q&A systems and a novel natural language processing (NLP) algorithm is suggested, as an additional layer to increase the VPF accuracy.
Abstract
Although Q&A systems have attracted a lot of research interest and efforts during the past few years (e.g. [1] and [2]), there are still a limited number of freely available tools for rapid application prototyping. This paper reviews and compares the available resources and suggests the Virtual People Factory (VPF) [3] as a good solution to rapidly built such systems. It then suggests a novel natural language processing (NLP) algorithm, as an additional layer to increase the VPF accuracy. Finally, the paper discusses a short study where the two approaches (script-based vs. language parsing) are compared and draws conclusions on their performance.