Big Five Prediction from Handwriting Images Using ResNet-50
DOI:
https://doi.org/10.30871/jaic.v10i2.12426Keywords:
Big Five, CNN, Handwriting, ResNet-50Abstract
Personality plays an important role in determining job performance, as mismatches between personality traits and job roles can lead to disengagement and reduced work performance. One of the most widely accepted models for understanding personality is the Five-Factor Model (FFM) or Big Five personality traits, which provides a comprehensive yet parsimonious representation of human personality. Although the Big Five is commonly measured using self-report questionnaires, this method is prone to response bias. Therefore, alternative approaches based on behavioral cues, such as graphology, have gained research attention. This study proposes a CNN-based approach using the ResNet architecture to predict Big Five personality traits from offline handwriting images. The objectives of this study are to evaluate the effectiveness of the ResNet-based model, analyze the impact of Otsu Thresholding preprocessing under normal and uneven lighting conditions, and determine optimal hyperparameter configurations. Experimental results show that the proposed model can learn personality-related patterns from handwriting images, with the best performance achieved using original images without preprocessing, resulting in an MAE of 7.94, RMSE of 10.71, and three-class classification accuracy of 66%. The findings indicate that Otsu Thresholding does not consistently improve performance and may remove important handwriting details. Overall, the results demonstrate that CNN-based ResNet models are effective for offline handwriting-based Big Five personality prediction.
Downloads
References
[1] H. Perera and L. Costa, “MBTI Personality Classification through Analyzing CVs/ Personal Statements for e-Recruitment,” International Journal of Machine Learning, vol. 14, no. 1, 2024, doi: 10.18178/ijml.2024.14.1.1153.
[2] H. W. Khan, M. U. Sattar, S. Noor, and M. I. Alyousef, “A Personality-Informed Candidate Recommendation Framework for Recruitment Using MBTI Typology,” Information, vol. 16, no. 10, p. 863, Oct. 2025, doi: 10.3390/info16100863.
[3] G. Angelini, “Big five model personality traits and job burnout: a systematic literature review,” BMC Psychol., vol. 11, no. 1, p. 49, Feb. 2023, doi: 10.1186/s40359-023-01056-y.
[4] N. A. Utami, W. Maharani, and I. Atastina, “Personality Classification of Facebook Users According to Big Five Personality Using SVM (Support Vector Machine) Method,” Procedia Comput. Sci., vol. 179, pp. 177–184, 2021, doi: 10.1016/j.procs.2020.12.023.
[5] S. Mustafa, Y. Qiao, X. Yan, A. Anwar, T. Hao, and S. Rana, “Digital Students’ Satisfaction With and Intention to Use Online Teaching Modes, Role of Big Five Personality Traits,” Front. Psychol., vol. 13, Jul. 2022, doi: 10.3389/fpsyg.2022.956281.
[6] A. Salecha, M. E. Ireland, S. Subrahmanya, J. Sedoc, L. H. Ungar, and J. C. Eichstaedt, “Large language models display human-like social desirability biases in Big Five personality surveys,” PNAS Nexus, vol. 3, no. 12, Nov. 2024, doi: 10.1093/pnasnexus/pgae533.
[7] J. Tiwari and R. Sadiwala, “Personality Prediction from Handwriting using Fine-tuned Transfer Learning Models,” SAMRIDDHI : A Journal of Physical Sciences, Engineering and Technology, vol. 15, no. 01, pp. 38–44, Jan. 2023, doi: 10.18090/samriddhi.v15i01.06.
[8] G. Marano et al., “The Neuroscience Behind Writing: Handwriting vs. Typing—Who Wins the Battle?,” ., vol. 15, no. 3, p. 345, Feb. 2025, doi: 10.3390/life15030345.
[9] C. Dazzi and L. Pedrabissi, “Graphology and Personality: An Empirical Study on Validity of Handwriting Analysis,” Psychol. Rep., vol. 105, no. 3_suppl, pp. 1255–1268, Dec. 2009, doi: 10.2466/PR0.105.F.1255-1268.
[10] F. Alaei and A. Alaei, “Handwriting analysis: Applications in person identification and forensic,” Breakthroughs in Digital Biometrics and Forensics, pp. 147–165, 2022, doi: 10.1007/978-3-031-10706-1_7.
[11] J. F. Vargas, M. A. Ferrer, C. M. Travieso, and J. B. Alonso, “Off-line signature verification based on grey level information using texture features,” Pattern Recognit., vol. 44, no. 2, pp. 375–385, Feb. 2011, doi: 10.1016/j.patcog.2010.07.028.
[12] S. Singh, N. K. Garg, and M. Kumar, “Feature extraction and classification techniques for handwritten Devanagari text recognition: a survey,” Multimed. Tools Appl., vol. 82, no. 1, pp. 747–775, Jan. 2023, doi: 10.1007/s11042-022-13318-9.
[13] A. Riadi and R. Sulaehani, “Analisis implementasi preprocessing dengan Otsu-Gaussian pada pengenalan wajah.,” ILKOM Jurnal Ilmiah, vol. 11, no. 3, pp. 200–205, Dec. 2019, doi: 10.33096/ilkom.v11i3.457.200-205.
[14] D. Gagiu and D. Sendrescu, “Detection of Personality Traits Using Handwriting and Deep Learning,” Applied Sciences, vol. 15, no. 4, p. 2154, Feb. 2025, doi: 10.3390/app15042154.
[15] A. A. Elngar et al., “Image classification based on CNN: A survey,” Journal of Cybersecurity and Information Management (JCIM), vol. 6, no. 1, pp. 18–50, 2021.
[16] G. Chaubey and S. K. Arjaria, “Personality prediction through handwriting analysis using convolutional neural networks,” in In Proceedings of International Conference on Computational Intelligence: ICCI 2020, Singapore: Springer Singapore, 2022, pp. 59–70. doi: 10.1007/978-981-16-3802-2_5.
[17] D. Alamsyah, Samsuryadi, W. Widhiarso, and S. Hasan, “Handwriting analysis for personality trait features identification using CNN,” in 2022 International Conference on Data Science and Its Applications (ICoDSA), IEEE, Jul. 2022, pp. 232–238. doi: 10.1109/ICoDSA55874.2022.9862910.
[18] M. Shafiq and Z. Gu, “Deep Residual Learning for Image Recognition: A Survey,” Applied Sciences, vol. 12, no. 18, p. 8972, Sep. 2022, doi: 10.3390/app12188972.
[19] D. Indra, T. Hasanuddin, R. Satra, and N. R. Wibowo, “Eggs Detection Using Otsu Thresholding Method,” in 2018 2nd East Indonesia Conference on Computer and Information Technology (EIConCIT), IEEE, Nov. 2018, pp. 10–13. doi: 10.1109/EIConCIT.2018.8878517.
[20] X. Zhang, X. Zhang, and W. Wang, “Convolutional Neural Network,” in Intelligent Information Processing with Matlab, Singapore: Springer Nature Singapore, 2023, pp. 39–71. doi: 10.1007/978-981-99-6449-9_2.
[21] Rikendry and A. Maharil, “Perbandingan arsitektur VGG16 dan ResNet50 untuk rekognisi tulisan tangan aksara lampung,” Jurnal Informatika dan Rekayasa Perangkat Lunak (JATIKA), vol. 3, no. 2, pp. 236–243, 2022, [Online]. Available: http://jim.teknokrat.ac.id/index.php/informatika
[22] Al Fathir Rizal Januar, J. Indra, D. S. Kusumaningrum, and S. Faisal, “Application of Convolutional Neural Network (CNN) Algorithm with ResNet-101 Architecture for Monkey Pox Detection in Human,” Journal of Applied Informatics and Computing, vol. 9, no. 3, pp. 1006–1012, Jun. 2025, doi: 10.30871/jaic.v9i3.9621.
[23] Q. A. Al-Haija and A. Adebanjo, “Breast Cancer Diagnosis in Histopathological Images Using ResNet-50 Convolutional Neural Network,” in 2020 IEEE International IOT, Electronics and Mechatronics Conference (IEMTRONICS), IEEE, Sep. 2020, pp. 1–7. doi: 10.1109/IEMTRONICS51293.2020.9216455.
[24] A. Shabbir et al., “Satellite and scene image classification based on transfer learning and fine tuning of ResNet50,” Math. Probl. Eng., vol. 2021, pp. 1–18, Jul. 2021, doi: 10.1155/2021/5843816.
[25] A. A. Suryanto, “Penerapan metode Mean Absolute Error (MAE) dalam algoritma regresi linear untuk prediksi produksi padi,” SAINTEKBU, vol. 11, no. 1, pp. 78–83, Feb. 2019, doi: 10.32764/saintekbu.v11i1.298.
[26] F. Rustam et al., “Wireless capsule endoscopy bleeding images classification using CNN based model,” IEEE Access, vol. 9, pp. 33675–33688, 2021, doi: 10.1109/ACCESS.2021.3061592.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Januar Firnando, Derry Alamsyah
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
