Developing a Career Guidance Expert System for Selection of University Programme by Secondary School Students

Selecting an appropriate university programme is a critical decision in the life of a secondary school student. However, many students make this decision without adequate professional guidance, leading to poor career alignment and underutilization of their potential. The lack of structured and accessible career counseling services in many schools further exacerbates this issue. This study presents the design and implementation of an Intelligent Career Guidance Expert System using Decision Tree Algorithm, Rule-Based Inference Engine, and Fuzzy Logic to assist students in making informed university programme choices. The system leverages artificial intelligence techniques to provide personalized recommendations based on students’ academic performance, vocational interests, personality traits, and career aspirations. A diverse dataset was compiled, including academic records, interest inventories, and personality indicators. The architecture of the expert system integrates a decision tree algorithm optimized via hyperparameter tuning (e.g., adjusting max_depth) alongside rule-based inference and fuzzy logic components to simulate expert reasoning. The system was trained and validated using real-world data from secondary schools and historical university admission records. Results showed that the decision tree model achieved high accuracy in mapping student profiles to appropriate programmes. The integration of rule-based reasoning and fuzzy logic further enhanced the system’s ability to handle uncertainty and expert-level nuances in decision-making. This expert system demonstrated the potential to offer reliable, cost-effective, and time-saving guidance, outperforming traditional counseling approaches. It provided students with data-driven, objective recommendations aligned with their long-term academic and career goals. The study underscores the value of integrating expert systems into school counseling services. Future improvements may include incorporating psychometric assessments, expanding datasets across various regions, and implementing feedback mechanisms involving human counselors to refine system performance continuously.