Arabic conversational agent for modern Islamic education

Alobaidi, Omar (2015) Arabic conversational agent for modern Islamic education. Doctoral thesis (PhD), Manchester Metropolitan University.

Preview

Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (6MB) | Preview

Abstract

This thesis presents research that combines the benefits of intelligent tutoring systems (ITS), Arabic conversational agents (CA) and learning theories by constructing a novel Arabic conversational intelligent tutoring system (CITS) called Abdullah. Abdullah CITS is a software program intended to deliver a tutorial to students aged between 10 and 12 years old, that covers the essential topics in Islam using natural language. The CITS aims to mimic a human Arabic tutor by engaging the students in dialogue using Modern standard Arabic language (MSA), whilst also allowing conversation and discussion in classical Arabic language (CAL). Developing a CITS for the Arabic language faces many challenges due to the complexity of the morphological system, non-standardization of the written text, ambiguity, and lack of resources. However, the main challenge for the developed Arabic CITS is how the user utterances are recognized and responded to by the CA, as well as how the domain is scripted and maintained. This research presents a novel Arabic CA and accompanying a scripting language that use a form of pattern matching, to handle users’ conversations when the user converse in MSA. A short text similarity measure is used within Abdullah CITS to extract the responses from CAL resources such as the Quran, Hadith, and Tafsir if there are no matching patterns with the Arabic conversation agent’s scripts. Abdullah CITS is able to capture the user’s level of knowledge and adapt the tutoring session and tutoring style to suit that particular learner’s level of knowledge. This is achieved through the inclusion of several learning theories and methods such as Gagne’s learning theory, Piaget learning theory, and storytelling method. These learning theories and methods implemented within Abdullah’s CITS architecture, are applied to personalise a tutorial to an individual learner. This research presents the first Arabic CITS, which utilises established learning typically employed in a classroom environment. The system was evaluated through end user testing with the target age group in schools both in Jordan and in the UK. Empirical experimentation has produced some positive results, indicating that Abdullah CITS is gauging the individual learner’s knowledge level and adapting the tutoring session to ensure learning gain is achieved.