Modeling of Compulsive Behavior Types of Obsessive-Compulsive Disorder Patients by Using the Data Mining Method.

Area of Science:

• Bioinformatics
• Psychiatry
• Data Science

Background:

• Obsessive-compulsive disorder (OCD) is a mental health condition characterized by obsessions and compulsions.
• Understanding the relationship between specific obsession types and compulsion types is crucial for effective treatment.
• Data mining offers advanced analytical methods to explore complex datasets in mental health research.

Purpose of the Study:

• To apply data mining techniques to classify algorithm accuracy for OCD patient data.
• To predict the types of compulsive behaviors in patients with obsessive-compulsive disorder.
• To investigate the influence of obsession types on compulsion types within an OCD cohort.

Main Methods:

• Data mining classification algorithms from WEKA software were utilized.
• Regression models were employed to determine the effect of obsession types on compulsion types.
• Data from 164 participants (70 males, 94 females, age 7-73) including sociodemographic, clinical, and familial information were analyzed.

Main Results:

• A moderate positive correlation (r=.35) was found between obsession types and compulsion types.
• Obsession types explained 11% of the variance in compulsion types, as indicated by the coefficient of determination.
• The accuracy of different classification algorithms was assessed for predictive modeling.

Conclusions:

• The study supports the hypothesis that obsession types significantly influence compulsion types in OCD.
• Data mining provides a valuable approach for analyzing complex relationships in psychiatric disorders.
• Findings can inform more targeted therapeutic interventions for individuals with obsessive-compulsive disorder.