Interpretable Recognition of Cognitive Distortions in Natural Language Texts

TL;DR

This paper introduces an interpretable AI method for detecting cognitive distortions in texts, improving accuracy over existing approaches and providing transparent, robust models for psychological applications.

Contribution

It presents a novel multi-factor classification approach using weighted structured patterns that considers heterarchical relationships, enhancing detection of cognitive distortions.

Findings

Significant F1 score improvements on two datasets

Models and code made publicly available

Enhanced interpretability and robustness of detection algorithms

Abstract

We propose a new approach to multi-factor classification of natural language texts based on weighted structured patterns such as N-grams, taking into account the heterarchical relationships between them, applied to solve such a socially impactful problem as the automation of detection of specific cognitive distortions in psychological care, relying on an interpretable, robust and transparent artificial intelligence model. The proposed recognition and learning algorithms improve the current state of the art in this field. The improvement is tested on two publicly available datasets, with significant improvements over literature-known F1 scores for the task, with optimal hyper-parameters determined, having code and models available for future use by the community.