RUS  ENG
Full version
JOURNALS // Proceedings of the Institute for System Programming of the RAS // Archive

Proceedings of ISP RAS, 2025 Volume 37, Issue 1, Pages 185–200 (Mi tisp959)

Using artificial intelligence technologies to conduct psychological testing

D. D. Grigoreva, D. V. Serov, D. S. Sorokin, A. I. Martyshkin

Penza State Technological University

Abstract: This research paper focuses on the use of artificial intelligence technologies for the Rorschach test. Machine learning techniques are considered. Both of these methods are used for multi-class classification of response categories. The paper describes the algorithms of machine learning and deep learning methods for interpreting the results, the algorithm for scoring one of the categories and the final test result in the web interface for the user. The application of artificial intelligence for projective testing techniques, on the example of the Rorschach test, opens new opportunities for self-diagnosis and therapy.

Keywords: neural network, multiclass classification, machine learning, psychological testing, Rorschach test, naive Bayes classifier, Python, TensorFlow, scikit-learn, Keras, LLM, langchain

DOI: 10.15514/ISPRAS-2025-37(1)-11



© Steklov Math. Inst. of RAS, 2025