A design science approach to semantic alignment: quantifying the curriculum–market gap using NLP-driven vectorization

Abstract

The demands of modern society have increased expectations about the quality of university graduates, especially in technology-driven disciplines. Nigerian graduates appear to be struggling to meet labour-market expectations as emerging technologies continue to evolve. Graduates of Information and Communications Technology (ICT) programmes are expected to acquire employable digital skills, yet curriculum reform has not always kept pace with industry requirements. The National Universities Commission (NUC) introduced the Core Curriculum and Minimum Academic Standards (CCMAS) 70:30 policy, which permits universities to use a 30% institutional window to introduce innovative and market-responsive content. Since the policy was unveiled in 2023, however, limited use has been made of this provision. This study found that market-validated innovation was approximately 1%, far below the 30% allowance. The researchers developed a diagnostic artifact that processes curriculum learning outcomes and industry job postings through a four-stage natural language processing (NLP) pipeline comprising compound normalisation, whitelist filtering, rule-based semantic inference, and set-intersection scoring. The artifact computes a Programme Alignment Score (PAS) for each university. The study analysed learning outcomes from 150 university handbooks and 1,500 real-time ICT job postings. The results reveal statistically measurable misalignment, with a national average PAS of 28% and a legacy burden of 69%, indicating that most technical skills required by the modern workforce are absent from Nigeria's formal ICT curricula.