Investigating the Psychological Outcomes Through Yoga for Stress Reduction Using Machine Learning and Statistical Techniques

Yoga has become crucial mode of intervention in enhancing mental health through empirical and data-supported methodologies. This study investigates the effects of yoga interventions after cognitive task on physiological and psychological parameters, specifically blood pressure (BP) and State-Trait Anxiety Inventory (STAI) scores, using optimized machine learning techniques. The primary objective was to analyze and predict how different yoga practices impact individual health outcomes through stress reduction. The initial phase involved collecting baseline data on BP and STAI parameters from participants before and after interventions. The various machine learning algorithms were employed to model the data, with a focus on their performance in predicting changes in BP and anxiety levels as measured by the STAI. In like manner, subsequent analysis using DT and RF models demonstrated promising results in terms of accuracy, with RF and DT particularly effective in handling the complex patterns in the combined BP and STAI data. Further, the Artificial Bee Colony algorithm was applied to optimize the hyper-parameters of the DT and RF models. The optimized models, termed BP-STAI Combined-Optimized DT and BP-STAI Combined-Optimized RF, achieved accuracies of 69.90% and 75.43%, respectively. These results highlight a significant improvement over the non-optimized models and underscore the effectiveness of combining physiological and psychological data in predictive analytics. The findings provide valuable insights into cognitive-driven stress reduction and the broader implications of employing advanced data-driven approaches in health and wellness research.