Today, there exists different types of chatbots and the two most common ones are either rule-based or data-driven. Rule based chatbots processes the user input through a large collection of rules. An example of a rule-based chatbot is A.L.I.C.E. which uses AIML, an XML language created specifically for chatbots to match the input pattern to predefined response templates.
This project explores a data driven approach for a chatbot where it utilises a database of transcripts from a TV-series in an attempt to hide its lack of linguistic knowledge and familiarity of the human language. A built upon version of the naive implementation with rules prioritizing responses from the same scene to increase coherency is implemented and is compared to the original. Both versions of the chatbot shows some coherency for the individual instance but this diminishes in longer conversations. No significant difference can be found between the two versions.
This paper will explore a data-driven approach where the chatbot uses a database of transcripts from a TV-series. Transcripts from TV-series contain human interaction and behaviour and with the use of this the chatbot will try to hide its lack of knowledge in the human language. A built-upon version of the chatbot will be implemented using a few rules in an attempt to increase coherency.
Author: Blomqvist, Niklas | Hygerth, Henrik