Strategizing Warfighter Competency: Leveraging NLP Based Recommenders for Targeted Learning Outcomes

Abstract

This executive summary synthetizes the research and findings of the NPS-25-N096-A project titled Strategizing Warfighter Competency: Leveraging NLP-Based Recommenders for Targeted Learning Outcomes. The research investigates how natural language processing (NLP) and artificial intelligence (AI) techniques can support competency-driven curriculum analysis and strategic alignment in modern warfighting education. Grounded in a comprehensive literature review of recommender systems, text classification methods, intelligent curriculum design, and competency frameworks, the study developed a proof-of-concept Curriculum Analyzer that processes course-level learning outcomes to extract structural and semantic insights. The system integrates transformer-based semantic embeddings, Bloom’s Taxonomy (Ajayi, 2024; Krathwohl, 2002) cognitive level identification, clustering algorithms, topic modeling, graph construction, and large language models (LLMs). Learning outcomes are clustered using multiple AI techniques, comparatively evaluated for robustness and interpretability, and labeled into coherent topic domains. Semantic relationships among clusters are computed using centroid-based cosine similarity and LLM-assisted relationship classification, forming a graph representation of curricular architecture. Competencies are then generated for each cluster and associated with external requirements, including accreditation standards and institutional learning outcomes. Experimental results, including a case study of a Space Systems Operations program, demonstrate the feasibility and practical value of the approach. Embedding-based clustering methods, particularly centroid-based techniques, produced stable and interpretable groupings, while the layered macro- and micro-level alignment framework enabled transparent mapping between curricula and competency requirements. The findings confirm that NLP-driven recommender architectures can provide scalable, interpretable, and strategically actionable insights for curriculum evaluation, competency mapping, and accreditation alignment. The proposed analyzer offers educational institutions a data-informed mechanism to enhance responsiveness, support operational readiness, and systematically and strategically align academic offerings with evolving warfighter competency needs. Future research may extend validation across additional domains, refine semantic relationship classification with the use of LLMs, and enrich explainability mechanisms to strengthen institutional decision-making support.