A Hierarchical and Multi-Tiered Personalized Career Recommender System Tailored to Individual Aptitudes

Objectives: To design and implement a hierarchical, multi-tiered career recommender system with real-world applications, that addresses data sparsity and cold-start limitations, while tackling the challenges posed by the low student-to-career counselor ratio in a context like India. The system aims to deliver automated, personalized career recommendations tailored to individual aptitudes, effectively guiding students toward appropriate educational pathways and contributing to optimizing the nation's human resource potential. Methods: This study has developed a novel hierarchical, multi-tiered recommendation system employing machine-learning models based on a content-based filtering approach. The model is built on a first-of-its-kind dataset comprising 612 unique real-world careers, curated by extracting ability profile information from the Occupational Information Network database (O*NET) and utilized a multi-tiered approach to deliver diverse yet personalized career recommendations, with visualizations of the recommended career clusters against the candidate’s user profile, providing clear insights into the rationale behind the system's recommendations. Findings: The system surpasses other career recommendation systems across multiple dimensions, demonstrating robust performance with accuracy, precision, recall, and f1-scores all exceeding 99%. It provides diverse career options aligned with the candidate’s profile through its unique multi-tiered outputs, while effectively addressing data sparsity and cold-start limitations. Novelty: The study curated a unique, first-of-its-kind dataset comprising 612 distinct careers, each mapped with their respective ability profile parameters. A novel hierarchical, multi-tiered recommendation engine was developed, introducing an innovative approach that enables the generation of diverse recommendations, overcoming the limited output capabilities characteristic of existing recommender systems. Keywords: Artificial Intelligence, Career Counselling, Psychometric Tests, Career Recommender Systems, Multi­Tiered Architecture, Data Sparsity and Cold-Start Limitations