Similarity Indices of Academic and Professional Trajectories Through Data Vectorization

It is well-known that academic background significantly influences career paths. However, identifying the most prominent factors that shape specific professional trajectories remains a complex challenge. This study proposes a data-driven approach leveraging automated similarity indices based on vector distance measures to identify analogous academic and professional patterns. The vectorization process involves defining key variables---including educational duration, degrees attained, work experience, job mobility, turnover, and tenure---extracted from professional platforms such as LinkedIn. Following a validation process guided by domain experts, results indicate that cosine similarity computed on robustly scaled representations outperforms Euclidean distance in capturing the nuances of trajectory similarities and divergences. Experimental results, applied to a dataset of \num{8614} LinkedIn profiles, demonstrate the effectiveness of the proposed framework. Furthermore, since the selection of variables dictates the similarity criteria, the method is highly adaptable to alternative variables or custom weighting schemes.