1Assistant Professor, Department of Computer Science and Engineering, KGiSL Institute of Technology, Coimbatore, India
Email: janani.r@kgkite.ac.in (Corresponding Author)
2Associate Professor, Department of Computer Science and Engineering, KGiSL Institute of Technology, Coimbatore, India
3,4,5,6Department of Computer Science and Engineering, KGiSL Institute of Technology, Coimbatore, India
Mental well-being is closely linked with hormonal cycles, emotional fluctuation, as well as lifestyle patterns that women undergo throughout their various stages in life. The ability to accurately detect stress levels and deliver appropriate assistance is a significant issue in the mental health systems. Since various machine learning models have been used in the prediction of stress, overreliance on one model normally constrains prediction quality and difficulty. The gap in this paper is the absence of effective hybrid algorithm systems that can be applied to predict stress and optimize interference to the mental well-being of women considering both behavioral, emotional, and lifestyle-related data. This work aims at comparing single algorithms, developing a hybrid prediction model, enhancing the accuracy of stress classification, and assessing optimized involvement results. The targeted approach is to merge learning with the help of the Random Forest method and gradient boosting with the XGBoost algorithm to combine the advantages of both systems. This will be done with aim of developing the reliability of stress prediction in order to justify individual stress reduction plans. The originality of this work is the hybrid algorithm comparison and the outcome-based stress reduction analysis through the stages of the life of women. It is experimentally evaluated that the hybrid model is more accurate and stable and able to generalize as compared to individual algorithms and therefore it can be applied to applications in real time centric to women mental well-being.
Index Terms: Women mental health, Hybrid learning model, Stress prediction, Random forest, XGBoost.
How to cite this article: Janani R, Senathipathi K, Nandakumar A, Arumugam C, Bharath J, Kaviyashree S. A Hybrid AI Framework Integrating XGBoost and Random Forest for Analyzing Hormonal Cycles and Women's Mental Well-Being. Int J Drug Deliv Technol. 2026;16(8s): 210-218; DOI: 10.25258/ijddt.16.8s.32
Source of support: None
Conflict of interest: None