International Journal of Artificial Intelligence
Vol 11 No 2: December 2024

Development of an Arabic-Language Virtual Assistant for Public Services to Improve Accessibility of Government Services for Iraqis

Alkarkhi, Dabbagh Adawiya (Unknown)
Al-Majedy, Zainab (Unknown)
Abbas, Mohammad Raziq (Unknown)
Aziz, Salim Hama (Unknown)



Article Info

Publish Date
19 Dec 2024

Abstract

This research explores the creation of an Arabic-language virtual assistant designed to enhance the accessibility of public services in Iraq. The study centers on developing a chatbot driven by Artificial Intelligence (AI) technologies, especially Natural Language Processing (NLP) and Machine Learning (ML), to help citizens in obtaining government services in Arabic, with a focus on the Iraqi dialect. The chatbot was created to address frequent questions regarding government services like passport renewals, birth registrations, and general inquiries. The system development employed platforms such as Google Dialogflow and TensorFlow to build an intuitive interface that can effectively handle and reply to user inquiries. Data collection comprised Arabic conversational data from Iraqi individuals, obtained via surveys and feedback on government services. The system's effectiveness was assessed through language comprehension accuracy, relevance of responses, and satisfaction of users, with findings indicating high satisfaction levels (88%), accuracy (95%), and relevance (92%). The results indicate that the chatbot greatly improves access to government services, minimizing the necessity for physical visits and increasing service efficiency. Nonetheless, issues regarding the complexities of the Iraqi Arabic dialect and voice recognition capabilities persist. The study finds that the chatbot presents a hopeful approach for addressing language and tech obstacles in providing public services. Future studies should aim at perfecting the language model and boosting voice input functionalities to better the chatbot's performance in Iraq’s public sector.

Copyrights © 2024






Journal Info

Abbrev

ijai

Publisher

Subject

Computer Science & IT Decision Sciences, Operations Research & Management

Description

The aim is to publish high-quality articles dedicated to Artificial Intelligence. IJAI published in biannual, and in Indonesian, Malay and ...