Architectural design and implementation of intelligent embodied conversational agents using fuzzy knowledge base

Abstract

In this paper, a general architecture is proposed for developing embodied conversational agents with fuzzy ontology knowledge base. The proposed architecture enables agents to interact with the user via multimodal channels in a virtual reality environment for the purpose of language learning. The agents play the role of emotional, rational, and friendly partners who provide a specific domain of knowledge based on user's queries in natural language. These queries are performed by an optimized fuzzy search engine. Two scenarios including two virtual airports and a virtual electronic gadget shop are implemented in this architecture to improve users' oral skills. The results show the users' average oral skills improved 11%. Moreover, 80% of the users ranked agents' logical sequence of actions and the total speed of responses as very good, and 90% of them evaluated agents' appropriateness of responses as very good based on Likert scale. ECAs have been implemented on a vast variety of platforms varying from mobile services [6, 32] to the World Wide Web [29, 33]. These sociable agents have been exploited in many service domains such as e-commerce [7, 25, 40], tourism query systems [2, 3], entertainment [31], and pedagogical ECAs [4, 21] for Computer-Assisted Language Learning (CALL) [13, 35]. The principal categories of ECA research include believability, social interface, application domains, computational issues, and production [24]. Many case studies show advantages of ECAs in enhancing problem solving skills, showing higher levels of motivation and interest, improving learning [30], having positive effect on memory performance [36], decreasing the intensity of user's negative emotions [19], etc. It is shown that ECAs have the role of a conversational partner and this effect is permanent [30].

Visualizing Natural Language Descriptions

“…Intelligent Virtual Environment for Language Learning (IVELL) [37,38] is a domainspecific multi-modal virtual reality system that consists of a few Embodied Conversational Agents (ECAs). It is designed to improve the speaking and listening skills of non-native users in English.…”