Revolutionizing Candidate Registration Support: Leveraging AI for an Optimized Public Examination Administration Process
Louisa Muparuri, Zimbabwe School Examinations Council (ZIMSEC) (Zimbabwe)
Abstract
The Zimbabwe Schools’ Examination Council (ZIMSEC) administers public examinations to between 700,000 and 1 million candidates annually, spanning from primary to high school candidates who are spread over the country’s 10 provinces through an estimated 8000 examination centers, of which 80% are in the rural areas. Whilst ZIMSEC has invested in a computerized candidate registration system, there is always a support challenge in handling both technical and non-technical stakeholder queries regarding the process of candidate registration across this huge number of stakeholders. Efforts of rural modernization, which are currently underway by the Zimbabwean Government, present opportunities of converting public examination activities to mimic private businesses, which are known to provide maximum support to stakeholders by leveraging emerging technologies such as social media platforms, websites, data analytics, and AI bots offering 24/7 support. Candidate registration is a key process leading to assessment, collection of certification data, and revenue. This paper presents a pioneering approach to providing real-time support for candidate registration-related queries, leveraging an Agentic AI chatbot developed using Microsoft Copilot Studio. The chatbot is designed to assist examination centers, parents, and candidates, offering 24/7 support and significantly reducing support and travel costs across the 10 provinces of Zimbabwe. By harnessing the power of Generative AI, Natural Language Processing (NLP), and machine learning (ML), the chatbot provides instant and accurate responses to a wide range of queries from technical, fees-related, due dates, financial support, and subjects offered, amongst others, ensuring a seamless registration experience. Results from a 21-day observation window show a stakeholder engagement of 61.8% satisfied customers with a satisfaction index of 3.6/5.0, a Total Answer Rate of 85% from a 100% knowledge source usage. The paper concludes by highlighting the benefits of integrating data analytics into the chatbot framework, enabling data-driven insights to inform decision-making at the ZIMSEC level up to the ministry level, optimizing support processes, and enhancing overall stakeholder satisfaction. The proposed solution has far-reaching implications for educational institutions, public exam boards seeking to streamline processes and provide exceptional support services.
Keywords: AI, Emerging Technologies, Data- driven insights, stakeholder engagement