The application of machine learning algorithms for forecasting the quality of life index of the population

   Aim. The work aimed to investigate the possibilities of applying various machine learning algorithms to forecast the quality of life index of the population.

   Objectives. The work seeks to develop predictive models for analyzing the quality of life index of the population of selected countries (Germany, India, the Netherlands, Russia) using various machine learning algorithms based on historical data from the Numbeo website from 2012 to 2025; as well as to systematize and analyze the results of machine learning models for these countries.

   Methods. The study used machine learning models such as random forest, linear regression, gradient boosting, k-nearest neighbors, and support vector machine. Forecasting the quality of life index of the population is based on data on socio-economic factors for various countries presented in the Numbeo database.

   Results. A comparative analysis of the results of forecasting the quality of life index of the population of selected countries was performed using machine learning algorithms based on historical data from 2012 to 2025. Particular attention is paid to adjusting the hyperparameters of the models and cross-validation to improve the accuracy of predictions. The analysis demonstrated that the most reliable results can be obtained using an ensemble of machine learning models without taking into account linear regression forecasts.

   Conclusion. The calculations performed revealed that the gradient boosting model demonstrates the best results. However, in order to improve accuracy and reduce deviations, it is recommended to use an ensemble of models. The use of machine learning in forecasting offers new opportunities for the development of social government programs aimed at improving the quality of life of the population.

1. Suvorov A.V. Forecasting the standard of living: Methodological problems. Studies on Russian Economic Development. 2000;11(1):1-10. (In Russ.: Problemy prognozirovaniya. 2000;(1):22-38.).

2. Derkachenko O.V. Clustering of regions of Russia by the quality of life of the population and the level of the digital economy. Voprosy regional’noi ekonomiki = Problems of Regional Economy. 2020;(2):59-65. (In Russ.). DOI: 10.21499/2078-4023-2020-43-2-59-65

3. Enikeeva L.A., Shirshikova M.S. Prognosis models of quality of life on the basis of international indexes. Sovremennye problemy nauki i obrazovaniya = Modern Problems of Science and Education. 2015;(1-1):656-664. (In Russ.).

4. Quality of life index by country. Numbeo. URL: https://www.numbeo.com/quality-of-life/rankings_by_country.jsp? (accessed on 01. 02. 2025).

5. Petrova E.A. Living standards control in the region: Modeling of factors of external and internal environment. Vestnik Volgogradskogo gosudarstvennogo universiteta. Seriya 3: Ekonomika. Ekologiya = Science Journal of VolSU. Global Economic System. 2018;20(1): 30-39. (In Russ.). DOI: 10.15688/jvolsu3.2018.1.4

6. Lipesa B.A., Okango E., Omolo B.O., Omondi E.O. An application of a supervised machine learning model for predicting life expectancy. SN Applied Sciences. 2023;5(7):189. DOI: 10.1007/s42452-023-05404-w

7. Shchekotin E.V., Goiko V.L., Basina P.A., Bakulin V.V. Using machine learning to study the population life quality: Methodological aspects. Tsifrovaya sotsiologiya = Digital Sociology. 2022;5(1):87-97. (In Russ.). DOI: 10.26425/2658-347X-2022-5-1-87-97

8. Chistiakov S.P. Random forests : An overview. Trudy Karel’skogo nauchnogo tsentra Rossiiskoi akademii nauk = Transactions of the Karelian Research Centre of the Russian Academy of Sciences. 2013;(1):117-136. (In Russ.).

9. Rotova O.M., Shibanova A.D. Basic principles of the linear regression method. Teoriya i praktika sovremennoi nauki = Theory and Practice of Modern Science. 2020;(1):483-487. (In Russ.).

10. Druzhkov P.N., Zolotykh N.Yu., Polovinkin A.N. Parallel implementation of prediction algorithm in gradient boosting trees method. Vestnik Yuzhno-Ural’skogo gosudarstvennogo universiteta. Seriya: Matematicheskoe modelirovanie i programmirovanie = Bulletin of the South Ural State University. Series: Mathematical Modelling, Programming & Computer Software. 2011;(37):82-89. (In Russ.).

11. Pyatkova A.A. Application of k-Nearest Neighbours method to representation of histogram time series. Dostizheniya vuzovskoi nauki. 2014;(8):206-210. (In Russ.).

12. Yi W., Gerasimov I.V., Kuzmin S.A., He H., Yang H. Application of support vector regression in modeling and data processing of rice leaves. Izvestiya SPbGETU LETI = Proceedings of Saint Petersburg Electrotechnical University. 2016;(2):17-22. (In Russ.).

13. SMOTE. Data Science. URL: https://datascience.eu/computer-programming/smote/(accessed on 01. 02. 2025).

14. Machine Learning in Python. Scikit-learn. URL: https://scikit-learn.org/ (accessed on 01. 02. 2025).

15. About pandas. Pandas. URL: https://pandas.pydata.org/about/ (accessed on 01. 02. 2025).

16. NumPy user guide. NymPy. URL: https://numpy.org/doc/stable/user/ (accessed on 01. 02. 2025).

17. Visualization with Python. Matplotlib. URL: https://matplotlib.org/ (accessed on 01. 02. 2025).

18. About SciPy. SciPy. https://scipy.org/about/ (accessed on 01. 02. 2025).

19. Project Jupyter Documentation. Project Jupyter. URL: https://docs.jupyter.org/ (accessed on 01. 02. 2025).

20. GridSearchCV. Scikit-learn developers. URL: https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html (accessed on 01. 02. 2025).

21. Noble J. What are ARIMA models? IBM. URL: https://www.ibm.com/think/topics/arima-model (accessed on 01. 02. 2025).

22. How can we perform the Nemenyi Post-Hoc Test in Python? Psychological Scales. URL: https://scales.arabpsychology.com/stats/how-can-we-perform-the-nemenyi-post-hoc-test-in-python/ (accessed on 01. 02. 2025).