CHARACTER ANALYSIS BASED ON HANDWRITING USING MACHINE LEARNING: CLASSIFICATION OF MANAGERIAL TRAITS OF TOP EXECUTIVES

This study aims to classify the managerial characteristics of top executives through character traits inferred from handwriting samples using a machine learning approach. A dataset of handwriting samples was analyzed using decision tree algorithms to identify patterns linked to leadership competencies. The methodology includes preprocessing of handwriting features, selection of relevant attributes, and application of supervised learning techniques. The results revealed a classification accuracy of 57%, suggesting that while the model can detect some managerial patterns, further improvement is needed. Limitations such as small sample size, limited feature diversity, and data quality may have influenced the results. Future studies are encouraged to use larger datasets and integrate advanced models such as deep learning techniques and multidimensional handwriting features. This study contributes to the growing literature on biometric indicators in organizational research by demonstrating a novel intersection between graphology and machine learning.