Prediction of the Level of Alexithymia through Machine Learning Methods Applied to Automatic Thoughts

Abstract

This study aims to investigate the relationship among alexithymia levels and automatic thoughts from cognitive behavioral therapy concepts. For this aim, Fisher Score analysis was applied to determine the most effective attributes of the automatic thoughts scale. In addition, the level of alexithymia was predicted by the introduction of the data set into the machine learning methods of the Artificial Neural Network (ANN) and Support Vector Machine (SVM). It is aimed to develop a roadmap of what automatic thoughts should be given priorities in studies. The research, from 10 different provinces of Turkey, was performed with a total of 714 participants, of which 386 (54%) male and 328 (46%) female. Personal information form, Automatic Thoughts Scale and Toronto Alexithymia scale were applied to the participants. The data set obtained from the scale of automatic thoughts was applied to the feature selection by using the Fisher Score method and a data set containing 5 attributes was obtained. As a result of the implementation of the SVM method to this data set, the alexithymia level was predicted with 4.01 RMSE error. The results show that the features of the automatic thoughts are related to the alexithymia level.

Keywords

• alexithymia
• automatic thoughts
• machine learning