Extracting Graphological Features for Identifying Personality Traits using Agglomerative Hierarchical Clustering Algorithm

Abstract

Handwriting/graphology is a unique and exclusive tool that describes one's non-verbal expression, which indirectly portrays the mental state and psychological state of a writer in a subconscious manner. The graphology analysis has been proven to identify and predict the signs of mental health disorders. This study explores the distinctive graphological features in Malaysian handwritings towards the identification of early sign of mental health disorders. The Agglomerative Hierarchical Clustering algorithm was proposed to build up clusters over the handwriting data. The promising finding suggests that the distinctive features could be useful in the personality traits analysis. The results from this study could be extended and further explored for identifying the early signs of depression through one's handwriting.