• DocumentCode
    2858105
  • Title

    Relationship between Personality and Handwriting of Chinese Characters Using Artificial Neural Network

  • Author

    Wang, Tingting ; Chen, Zhanghui ; Li, Wenlong ; Huang, Xiaohui ; Chen, Pengfei ; Zhu, Siyao

  • Author_Institution
    Dept. of Electron. Sci. & Technol., Huazhong Univ. of Sci. & Technol., Wuhan, China
  • fYear
    2009
  • fDate
    19-20 Dec. 2009
  • Firstpage
    1
  • Lastpage
    5
  • Abstract
    This paper adopts artificial neural network to identify the relationship between personality and handwriting of Chinese characters. The handwriting feature studied here is character spacing, which is extracted by samples scanning and image processing. And the personality factors are obtained by Cattel¿s 16PF. Results of 16-8-1 neural network show that the association is significant between 16 personality factors and character spacing. And from the training and non-linear fitting of 16-1-1 network, which is trained alternately by genetic algorithm and Levenberg-Marquardt algorithm, it is found that character spacing has strong correlations with reasoning and sensitivity.
  • Keywords
    behavioural sciences computing; correlation methods; genetic algorithms; handwriting recognition; image processing; neural nets; 16-1-1 network nonlinear fitting; 16-8-1 neural network; Cattells 16PF; Chinese character; Levenberg-Marquardt algorithm; artificial neural network; character spacing; genetic algorithm; handwriting feature; image processing; personality-handwriting relationship; samples scanning; Algorithm design and analysis; Artificial neural networks; Feature extraction; Genetic algorithms; Image processing; Monitoring; Paper technology; Personnel; Psychology; Regression analysis;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Information Engineering and Computer Science, 2009. ICIECS 2009. International Conference on
  • Conference_Location
    Wuhan
  • Print_ISBN
    978-1-4244-4994-1
  • Type

    conf

  • DOI
    10.1109/ICIECS.2009.5365840
  • Filename
    5365840