Natural language scripting within conversational agent design

This paper presents a novel Semantic-Based Conversational Agent (SCA). Traditional conversational agents (CA) interpret scripts consisting of structural patterns of sentences, which take no consideration of semantic content. The script writer must therefore anticipate the many variations of input the user may respond with during dialogue. This is evidently a high maintenance task. Furthermore, different script writers possess differing levels of skill and as such this can prove to be an exasperating task. The proposed SCA interprets scripts consisting of natural language sentences by means of a semantic sentence similarity measure. User input is measured semantically against the natural language sentences of the current context in order to respond with an appropriate output string. Such scripting is effortless and alleviates the burden of the traditional pattern-scripted languages. Experiments have involved the use of script writers to demonstrate the use of the language. Results have highlighted the potential of the language and shown improvements on traditional pattern-scripted languages.