• DocumentCode
    2650182
  • Title

    Does There Exist Relationship between Personality and Handwriting of Chinese Characters? A View from Image Mining

  • Author

    Chen, Zhanghui ; Zhou, Baoyao ; Fong, Alvis C M

  • Author_Institution
    Inst. of Semicond., Beijing, China
  • fYear
    2011
  • fDate
    7-9 Nov. 2011
  • Firstpage
    229
  • Lastpage
    232
  • Abstract
    This paper presents a study on the relationship between personality and handwriting of Chinese characters through image mining technologies. A questionnaire of personality test is used to quantify the 5 global personality factors of participants. The handwriting samples of participants are acquired and scanned into computer images. 23 handwriting features can be extracted from these sample images through image processing methods. Considering the imbalanced distribution of the sample data, a cost-sensitive neural network with modified training algorithms and correlation analysis are employed to examine the association between the handwriting features and global personality factors. The results hint that there indeed exist some weak linear and strong non-linear relationships between most of personality factors and specific handwriting features. These relationships provide the possibility for computerized analyzing people´s personality by their Chinese handwriting.
  • Keywords
    data mining; feature extraction; handwriting recognition; handwritten character recognition; image scanners; neural nets; Chinese character handwriting; computer image scanning; correlation analysis; cost-sensitive neural network; handwriting feature extraction; image mining; image processing method; modified training algorithm; people personality computerized analysis; personality test; Artificial intelligence; Conferences; correlation analysis; cost-sensitive neural network; handwriting analysis; image mining;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Tools with Artificial Intelligence (ICTAI), 2011 23rd IEEE International Conference on
  • Conference_Location
    Boca Raton, FL
  • ISSN
    1082-3409
  • Print_ISBN
    978-1-4577-2068-0
  • Electronic_ISBN
    1082-3409
  • Type

    conf

  • DOI
    10.1109/ICTAI.2011.42
  • Filename
    6103332