# A method and system for generating a combination of psychological testing scales

## TL;DR

This paper introduces a new algorithm to optimize psychological testing scales by combining comprehensive and individual assessments more efficiently.

## Contribution

The novel contribution is framing psychological scale optimization as a multi-label classification problem and introducing a hierarchical algorithm for it.

## Key findings

- The algorithm improves efficiency and accuracy in diagnosing moderate to severe psychological symptoms.
- It dynamically adapts test content and sequences items based on historical diagnosis data and probability indices.
- The method shows lower accuracy for mild symptoms due to their lower positive rate.

## Abstract

Given the inherent limitations of traditional psychological testing scales in terms of breadth and specificity, both comprehensive and individual assessment scales offer distinct advantages. However, challenges persist because of their complementary deficiencies in practical applications. This article argues that intricate content optimization of scales is not necessary; instead, it conceptualizes comprehensive scales and individual assessment items as a multi-label classification problem. By employing a hierarchical framework, it becomes possible to achieve overall optimization of psychological testing scales. To this end, the paper introduces an algorithm designed to generate combinations of psychological scales, optimizing both comprehensive and individual assessments. This optimization is realized through a series of methodological steps, including the analysis of historical positive diagnosis data, the calculation of item probability indices, the dynamic adaptation of test content, the sequencing of items and the construction of a hierarchical scale system. Simulation experiments demonstrate that this approach enhances the efficiency and accuracy of psychological testing, particularly in diagnosing moderate to severe symptoms. However, the algorithm exhibits relatively lower accuracy for mild symptoms owing to their lower positive rate. The proposed algorithm significantly improves the optimization of psychological testing scales, particularly excelling in the assessment of moderate symptoms.