Design and Implementation of an AI-Driven Chatbot for Federal Polytechnic Nekede Admissions and Information Services
TL;DR
The study concludes that AI-powered chatbot systems offer a substantive remedy to institutional communication inefficiencies and recommends full deployment alongside continuous model retraining using current institutional data.
Abstract
The rising demand for real-time, automated information services within Nigerian tertiary institutions has brought the shortcomings of conventional inquiry management systems into sharp relief. Federal Polytechnic Nekede, Owerri, contends with a persistent backlog of prospective student queries, admission-related inquiries, and general information requests, particularly during peak application periods. This study reports the design and implementation of an Artificial Intelligence (AI)-driven chatbot system, designated NEKBOT, developed specifically for the institution's admissions and information services. The system employs Natural Language Processing (NLP), machine learning intent classification, and a structured institutional knowledge base to generate context-sensitive, accurate, and near-instantaneous responses to user queries. Development proceeded through the Agile Scrum methodology across six iterative sprints. The implementation stack comprised Python (Flask framework), Google Dialogflow NLP engine, JavaScript, and a MySQL relational database. Evaluation outcomes show an intent recognition accuracy of 91.4%, a mean response latency of 1.83 seconds, and a user satisfaction rating of 87% on standardised usability measures. The study concludes that AI-powered chatbot systems offer a substantive remedy to institutional communication inefficiencies and recommends full deployment alongside continuous model retraining using current institutional data.