This project describes a method to predict the Big-Five personality traits of an individual by analyzing his or her handwriting patterns with machine learning. This novel architecture has a feature extraction mechanism in which all the handwriting features(t-bar, pressure, baseline angle, word spacing and slant angle). For each of the personality traits, three classifiers are trained i.e. Logistic Regression, Random forest classifier and SVM. After sufficient training, we are able to predict personality traits on new handwriting image samples with great accuracy and efficiency.The proposed system computes the results in less than 90 seconds, which is faster than current methods of determining personality traits through lengthy self-report questionnaires, which are typically more cumbersome and time-consuming to fill out and require more effort from both the subject and the psychologist, who must post-process the questionnaire results and evaluate the five personality traits; this demonstrates that our current approach could be used as an attractive and more faster alternative to these commonly use personality evaluation techniques
