Automatic Personality Prediction Based on Users’ Chinese Handwriting Change

In recent years, personality has been considered as a valuable personal factor being applied to many fields. Although lately some studies have endeavored to implicitly obtain user’s personality from her/his handwriting, they failed to achieve satisfactory prediction performance. Most of the related methods focus on constructing handwriting features, while the handwriting change information is ignored. In fact, user’s handwriting change could reflect her/his physical and mental state more finely, which is helpful for recognizing the user’s personality. Furthermore, the related studies may not fully use Chinese character features to analyze the change of Chinese handwriting. In this paper, we propose an effective Chinese Handwriting Change based Personality Prediction (CHCPP) model to identify users’ personalities. To be specific, we construct the handwritten character sequence based on the writing order. We then extract the Chinese character features and the visual signals of each handwritten character in the sequence to analyze the handwriting change. Meanwhile, we also construct the statistical Chinese character features based on the whole handwritten character set to assist in modeling the change of Chinese handwriting. Lastly, we utilize the handwriting change information and the statistical Chinese character features to acquire the prediction results. The experimental results show that our CHCPP model outperforms the related methods on a real-world dataset.