Iterative Development and Evaluation of a Social Conversational Agent

We show that an agent with fairly good social conversational abilities can be built based on a limited number of topics and dialogue strategies if it is tailored to its intended users through a high degree of user involvement during an iterative development process. The technology used is pattern matching of question-answer pairs, coupled with strategies to handle: followup questions, utterances not understood, abusive utterances, repetitive utterances, and initiation of new topics.