Predicting generalized anxiety disorder among women using Shapley value.

Area of Science:

• Psychiatry and Data Science
• Computational Mental Health

Background:

• Mental illnesses, including anxiety disorders, are prevalent but often misunderstood.
• Existing data mining research inadequately addresses specific conditions like generalized anxiety disorder (GAD).
• There is a need for improved methods to identify GAD, particularly in women.

Purpose of the Study:

• To develop and implement a novel data mining system for predicting GAD in women.
• To apply Shapley value as a feature selection method within a data mining classifier.
• To enhance the early detection of GAD for improved mental health outcomes.

Main Methods:

• The study adapted the Knowledge Discovery in Databases (KDD) methodology.
• Key phases included data acquisition, pre-processing, feature selection (Shapley value), classification, and evaluation.
• A novel feature selection and data mining classifier system was implemented.

Main Results:

• The proposed method successfully implements Shapley value for feature selection in GAD prediction.
• The data mining approach facilitates the identification of potential GAD cases in women.
• The system aims to improve the accuracy and efficiency of GAD detection.

Conclusions:

• The developed data mining system offers a promising tool for identifying generalized anxiety disorder in women.
• This research addresses a gap in data mining applications for specific mental health conditions.
• The study aims to increase mental health awareness and empower women to seek necessary services.