Enhancing English-Arabic Chatbots for IT Support in Higher Education

Abstract

Nowadays, technical support teams in universities face challenges due to increasing student and staff demand for quick solutions to their technical issues. The growing demand for IT support in higher education has led to a growing interest in online support. Universities must always provide excellent service to ensure the satisfaction of students and staff. This research endeavours to address this challenge by developing two innovative English-Arabic IT service chatbots: chatbot one utilised TensorFlow, SpaCy, and Seq2Seq, and chatbot two used OpenAI. This research will also present a detailed comparison between the approaches. The chatbot of the OpenAI model performs exceptionally well and is human-like compared to the chatbot one. Both chatbots show competence in addressing IT service queries at universities in English and Arabic contexts. The evaluation process involved the use of both the BLEU metric and developer judgement to assess the quality of the results obtained. These results encourage universities to consider incorporating IT support chatbot solutions into their systems, enabling them to respond promptly and effectively to students and staff and reduce waiting times