KMAN Publication Inc. (KMANPUB) Psychology of Woman Journal 3041-8615 7 Serial Number 26 2026 04 01 Using Machine Learning to Identify Latent Profiles of Somatosensory Amplification and Catastrophic Misinterpretation of Bodily Cues in Psychosomatic Patients Using Machine Learning to Identify Latent Profiles of Somatosensory Amplification and Catastrophic Misinterpretation of Bodily Cues in Psychosomatic Patients 1 10 10.61838/kman.pwj.5168 EN Journal Article 2026 01 09 <p><strong>Objective:</strong> To utilize a machine learning framework to evaluate and interpret the predictive power and complex interactions of perceived efficacy, social network centrality, and collective identity strength on the advocacy behavior of women.</p> <p><strong>Methods and Materials:</strong> This cross-sectional quantitative study analyzed data from a stratified random sample of exactly women in South Africa. Validated self-report scales measured advocacy behavior, internal and external perceived efficacy, social network centrality (derived via an egocentric generator), and collective identity strength. Predictive modeling compared four machine learning algorithms (Gradient Boosting Machine, Random Forest, Support Vector Machine, and Logistic Regression) using an training and testing data partition. Model performance was evaluated using and -score metrics, with SHAP values extracted from the optimal model for feature interpretability.</p> <p><strong>Findings:</strong> Tree-based ensemble methods significantly outperformed linear baselines, with the Gradient Boosting Machine achieving the highest of on the test set. SHAP analysis demonstrated that Collective Identity Strength (Mean Absolute SHAP Value of ) was the paramount predictor, followed by Social Network Centrality ( ). Furthermore, Internal Perceived Efficacy ( ) was a substantially stronger predictor than External Perceived Efficacy ( ). Bivariate analyses corroborated these algorithmic outputs, showing advocacy behavior correlated most strongly with Collective Identity Strength ( , ).</p> <p><strong>Conclusion:</strong> Advocacy behavior in women is optimally predicted by a non-linear integration dominated by collective identity strength, structural network position, and internal self-efficacy.</p> https://journals.kmanpub.com/index.php/psywoman/article/download/5168/9681