Scholarlink

This study examines the integration of graphology and machine learning, including deep learning, to analyze personality traits from handwriting. The research builds on the premise that neuromuscular movements in handwriting reflect personality characteristics. A dataset of 1,108 handwriting images, sourced from the Centre for Pattern Recognition and Machine Intelligence (CENPARMI) and a graphology expert, was analyzed using machine learning algorithms such as k-Nearest Neighbor (k-NN), Random Forest, Logistic Regression, and deep learning techniques via transfer learning. Data balancing was addressed using the Synthetic Minority Over-Sampling Technique (SMOTE), and ensemble methods like Majority Voting and Stacking improved classification performance. Results showed over 90% accuracy for traits such as “Agreeableness” and “Openness to Experience” using the Stacking method. Key contributions include integrating graphology with machine learning and deep learning, innovating methods for imbalanced data, and advancing handwriting analysis for personality assessment. This interdisciplinary approach demonstrates significant potential for improving personality prediction accuracy and sets a foundation for future research in computational psychology and related fields.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Multi-model Approach for Personality Detection from Handwriting: Deep Features, Machine Learning, and Ensemble Techniques

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Maedeh Safar,
  • Ching Y. Suen

摘要

This study examines the integration of graphology and machine learning, including deep learning, to analyze personality traits from handwriting. The research builds on the premise that neuromuscular movements in handwriting reflect personality characteristics. A dataset of 1,108 handwriting images, sourced from the Centre for Pattern Recognition and Machine Intelligence (CENPARMI) and a graphology expert, was analyzed using machine learning algorithms such as k-Nearest Neighbor (k-NN), Random Forest, Logistic Regression, and deep learning techniques via transfer learning. Data balancing was addressed using the Synthetic Minority Over-Sampling Technique (SMOTE), and ensemble methods like Majority Voting and Stacking improved classification performance. Results showed over 90% accuracy for traits such as “Agreeableness” and “Openness to Experience” using the Stacking method. Key contributions include integrating graphology with machine learning and deep learning, innovating methods for imbalanced data, and advancing handwriting analysis for personality assessment. This interdisciplinary approach demonstrates significant potential for improving personality prediction accuracy and sets a foundation for future research in computational psychology and related fields.