| Resumen |
"In this paper, we approach the problem of automated recognition and analysis of handwriting, applied in the context of personality trait estimation. Handwriting presents a singular and distinct expression for every individual,
carrying valuable insights into the writer’s personality and psychological traits. Despite several proposals already made to tackle this issue, there are still obstacles to overcome, such as the selection of algorithmic techniques for
image quality enhancement. Our proposed methodology is based on analyzing handwritten text images, where we seek to identify patterns and features that allow us to infer specific personality traits. To achieve this, we have relied
on the theoretical framework of the Big Five model, one of the most widely accepted models. The proposed methodology involved preprocessing images using a U-Net neural network and a convolutional layer-based architecture to
classify personality traits." |
