Abstract
The paper formulates the problem of generating psychological tests using artificial intelligence and proposes a general principle of its functioning. This method of generating psychological tests (obtaining them automatically) by processing an array of experimental data is primarily intended for completing works of fiction (in particular, science fiction novels). The combination of psychological testing with works of art, in turn, is aimed at analyzing the psychological state of the population on a mass scale. The principle of operation of the proposed system is based on the use of permutation tests, in which the respondent ranks certain images based on his personal preferences, as well as on identifying a set of reference sequences (substitutions in terms of group theory). The latter are selected as the basis for the generated test. The selection of reference sequences is carried out by minimizing errors (deviations) between them and the sequences obtained during experiments. The work also proposes algorithms that make it possible to compact the information obtained through the use of permutation tests, based on the use of functions that take values in Galois fields.