Neuroticism and Handwriting – Analysis Using Machine Learning Techniques
Dr. Zuzanna Górska-Kanabus
Cardinal Stefan Wyszynski University, Warsaw, Poland
Dr. Artur Janicki
Warsaw University of Technology, Warsaw, Poland
DOI:
In Chernov, Y., & Nauer, M. A. (Eds.). (2018). Handwriting Research: Validation & Quality. P. 134-144.
Abstract
Neuroticism is one of the traits in the five-factor model of personality. It can be described as a continuum of emotional stability. In other words, it is susceptibility to experiencing negative emotions (fear, abashment, anger, or guilt) and sensitivity to psychological stress. For a person with a high neuroticism level, adaptation to the requirements of the environment can be difficult. He or she has low self-esteem. Such a person has a strong reaction to fear or feels tension, they also easily get discouraged and break down in difficult situations. A person with high neuroticism often experiences hostility and anger. Among other people, he or she can feel confused or embarrassed.
The level of the neuroticism determines behaviour of the individual in his or her personal and professional life. Several studies have been conducted on the possibility of assessing the neuroticism level based on handwriting analysis. In the present study, the authors wanted to examine if there exist any patterns in the handwriting features that are characteristic for low and medium levels of neuroticism. Similar to a previous study, machine learning approaches (including support vector machines, SVMs) were used. This was chosen because it gives the opportunity to combine a statistical and matching method. It was assumed that people characterized by a low level of neuroticism will share the same pattern of features in their handwriting. Based on this pattern, machine learning algorithms (classifiers) would be able to recognize, with acceptable accuracy, the level of neuroticism.