IAES International Journal of Artificial Intelligence (IJ-AI) Vol. 9, No. 4, December 2020, pp. 569~575 ISSN: 2252-8938, DOI: 10.11591/ijai.v9.i4.pp569-575  569 Journal homepage: http://ijai.iaescore.com The role of chatterbots in enhancing tourism: a case study of Penang tourism spots Vinothini Kasinathan 1 , Aida Mustapha 2 , Mohamad Firdaus Che Abdul Rani 3 , Salama A. Mostafa 4 1,3 School of Computing, Asia Pacific University of Technology and Innovation, Malaysia 2,4 Faculty of Computer Science and Information Technology, Universiti Tun Hussein Onn Malaysia, Malaysia Article Info ABSTRACT Article history: Received Apr 18, 2020 Revised Jul 20, 2020 Accepted Aug 30, 2020 Chatterbots have been widely used as a tool for conversational booking assistance mainly for hotels such as the Expedia. This paper extends the use of chatterbot beyond booking by presenting the proof of concept of a chatterbot expert system called the VIZARD. The proposed VIZARD is developed using an expert system shell called verbot. The core of Vertbot 5 is the natural language processing (NLP) engine based on pattern matching. The core Verbot 5 engine is responsible for finding matches to a given user input string and firing the appropriate rule. The findings from the user acceptance test concluded that majority of the respondents agreed that the VIZARD expert system stands at an unbiased state while being more aligned on supporting the usefulness of the system. Keywords: Chatterbots Conversational agents Expert system Scheduling Tourism Verbot This is an open access article under the CC BY-SA license. Corresponding Author: Vinothini Kasinathan School of Computing Asia Pacific University of Technology and Innovation Technology Park Malaysia, 57000 Kuala Lumpur, Malaysia Email: vinothini@uthm.edu.my 1. INTRODUCTION Chatterbots are changing the travel industry. Chatterbot or chatbots (short of robots) have been widely used as a tool for conversational booking assistance mainly for hotels such as the Expedia at booking.com. According to the IBM Watson’s blog, the use of virtual agents and chatbots is able to reduce the costs of customer service by up to 30% by implementing conversation-friendly solutions [1]. Chatterbot works like a decision tree, where users can select variation of options hence the potential is very promising in the area of customer services. Chatterbot is a computer program designed to interact with people by simulating human conversation [2-4]. The bot works best when maintaining a conversation within a specific domain and is sophisticated enough to control the current age concepts such as semantic parsing and sentiment analysis. However, one shortfall for the contemporary chatterbots is that they are unable to observe and analyze the context of conversations. Three main concerns in creating a good chatterbot conversations includes the content, semantics, and evaluation. Content is basically knowing what to say, semantics is knowing how to express it through a conversation, and lastly the evaluation which is mainly having a standard benchmark to grade the conversation produced. There is a method to validate whether a chatterbot is truly an artificial intelligence that is named Grice’s Maxims [5]. Grice’s Maxims is the cooperative principle that describes how people interact with one another in making contribution to a series of exchanged talk, by the accepted purpose or direction of the talk exchange in which the user is engaged. It has been demonstrated that evaluating chatterbots using this method is one of the most effective ways to compare to other chatterbots competing for