Personality Analysis Based on The Handwriting Shape using Integration of Image Processing and Machine Learning Algorithm

In Indonesia, mental health awareness is not as prominent as physical health, and many adolescents have experienced mental health problems. Some people hide their mental and emotional problems by appearing cheerful and peaceful, but research shows that body language and voice tone can reveal a person's inner psychological state. Similarly, handwriting can be analyzed through graphology to reveal personality traits such as honesty, emotional stability, and fears. Graphology is a time-consuming and expensive process, using image processing and machine learning, graphology can be done quickly without human bias. This research aims to predict the personality traits by implement image processing combined with machine learning to analyze handwriting features. Using lazy learner algorithm of K-Nearest Neighbor algorithm, SVM and Naïve Bayes, the application can predict the MBTI personality from handwriting image with an accuracy of 94%, 78%, and 93% respectively. From the experiment, it is shown that the application able to do the analysis of one's handwriting and then predict the personality traits, fully automated using a computer