Abstract

Signature is the representation of a personal identity that written on a selected medium. Signature can describe people's personality, which includes character, potential, motivation, emotional stability, mental state, intellectual tendencies, and one's work habits. Understanding people's personality is useful to show personality traits in criminology, medical science, and counseling. Previous research has investigated the relationship between online signature and personality by using tablet digitizer. In this paper, we are improving the method by increasing the amount of respondents, using a classification algorithm, and using psychological tests for validation named Big Five Inventory Test (BFI Test). The two best features pressure and speed have been analyzed and classified using k-Nearest Neighbor (kNN). The highest classification accuracy from kNN is 87.5%. From 40 respondents who were involved, 90% were confirmed well with BFI test for the first or second dominant personality. As a conclusion, this research shows that online signature analysis could predict personality with high accuracy.

Original languageEnglish
Title of host publicationProceedings - 2019 International Seminar on Intelligent Technology and Its Application, ISITIA 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages404-409
Number of pages6
ISBN (Electronic)9781728137490
DOIs
Publication statusPublished - Aug 2019
Event2019 International Seminar on Intelligent Technology and Its Application, ISITIA 2019 - Surabaya, Indonesia
Duration: 28 Aug 201929 Aug 2019

Publication series

NameProceedings - 2019 International Seminar on Intelligent Technology and Its Application, ISITIA 2019

Conference

Conference2019 International Seminar on Intelligent Technology and Its Application, ISITIA 2019
Country/TerritoryIndonesia
CitySurabaya
Period28/08/1929/08/19

Keywords

  • BFI Test
  • Classification Algorithm
  • Online Signature
  • kNN

Fingerprint

Dive into the research topics of 'Personality Classification from Online Handwritten Signature using k-Nearest Neighbor'. Together they form a unique fingerprint.

Cite this