Predicting the Quality of Life Index: A Comparative XGBoost–LSTM Study with SHAP-Based Explainable AI

Yıl 2025, Cilt: 14 Sayı: 5, 2208 - 2224, 31.12.2025

İbrahim Budak

https://doi.org/10.15869/itobiad.1791179

Öz

In this study, the prediction performance of different artificial intelligence algorithms was examined using quality of life data from 2016 to 2025. The analysis compared gradient-boosted tree-based XGBoost with LSTM, which has the capacity to model time series and sequential dependencies. In addition, SHAP analysis was applied to ensure the model's explainability and to identify the key factors affecting quality of life. The findings show that both models successfully capture quality of life patterns, with the LSTM model achieving higher out-of-sample accuracy than XGBoost (higher R² and lower MAE, RMSE, and MAPE). SHAP analysis revealed that Purchasing Power and Pollution are the factors with the strongest impact on quality of life. The decisive effect of Purchasing Power indicates that macroeconomic conditions such as real income level, price stability, and Purchasing Power Parity -adjusted welfare indicators directly reflect quality of life. Other factors, such as cost of living, housing price/income ratio, security, healthcare services, climate, and commute time, were found to have varying degrees of importance across countries. These findings emphasize the priority of designing macroeconomic frameworks targeting income/wage policies and price stability alongside policies aimed at improving environmental conditions. The results obtained indicate that policy makers should focus on the efficient allocation of resources. The results obtained provide policymakers with an evidence-based roadmap for the efficient allocation of resources and demonstrate that more detailed analyses can be conducted using different explainable artificial intelligence methods for future research. Additionally, to test the robustness of the model, different training/testing splits, alternative error metrics, and hyperparameter sensitivity analyses were performed; the direction and magnitude of the main findings were found to be consistent across these scenarios. Finally, SHAP-based findings provide a starting framework for policy simulations, enabling the quantitative prediction of potential welfare gains from targeted improvements in specific sub-indices.

Anahtar Kelimeler

Quality of Life , Purchasing Power , XGBoost , LSTM , SHAP , Explainable AI

Yaşam Kalitesi Endeksinin Tahmini: SHAP Tabanlı Açıklanabilir Yapay Zeka ile Karşılaştırmalı Bir XGBoost–LSTM Çalışması

Öz

Bu çalışmada, 2016–2025 yılları arasındaki yaşam kalitesi verileri kullanılarak farklı yapay zekâ algoritmalarının tahmin performansları incelenmiştir. Analizde, gradyan artırmalı ağaç tabanlı XGBoost ile zaman serisi ve ardışık bağımlılıkları modelleme kapasitesine sahip LSTM karşılaştırılmıştır. Ayrıca, modelin açıklanabilirliğini sağlamak ve yaşam kalitesini etkileyen temel faktörleri belirlemek amacıyla SHAP analizi uygulanmıştır. Bulgular, her iki modelin de yaşam kalitesi eğilimlerini başarılı şekilde yakaladığını, ancak LSTM’in daha yüksek doğruluk (R²=0.98) ve daha düşük hata değerleri (MAE, RMSE, MAPE) ile XGBoost’a göre üstün performans sergilediğini göstermektedir. SHAP analizi, yaşam kalitesi üzerinde en güçlü etkiye sahip faktörlerin Satın Alma Gücü ve Kirlilik olduğunu ortaya koymuştur. Satın Alma Gücü’nün belirleyici etkisi, reel gelir düzeyi, fiyat istikrarı ve satın alma gücü paritesi-düzeltilmiş refah göstergeleri gibi makroekonomik koşulların yaşam kalitesine doğrudan yansıdığını göstermektedir. Yaşam maliyeti, konut fiyat/gelir oranı, güvenlik, sağlık hizmetleri, iklim ve trafik süresi gibi diğer faktörlerin ise ülkeler arasında değişen önem derecelerine sahip olduğu belirlenmiştir. Bu bulgular, gelir/ücret politikaları ve fiyat istikrarını hedefleyen makroekonomik çerçeveler ile çevresel koşulların iyileştirilmesine yönelik politikaların birlikte tasarlanmasının öncelikli olması gerektiğini vurgulamaktadır. Elde edilen sonuçlar, politika yapıcılar için kaynakların verimli yönlendirilmesine yönelik kanıta dayalı bir yol haritası sunmakta ve gelecekteki araştırmalar için farklı açıklanabilir yapay zekâ yöntemleri ile daha ayrıntılı analizlerin yapılabileceğini göstermektedir. Ek olarak, model sağlamlığını sınamak amacıyla farklı eğitim/test bölmeleri, alternatif hata metrikleri ve hiperparametre duyarlılık analizleri gerçekleştirilmiş; ana bulguların yönü ve büyüklüğü bu senaryolarda da tutarlı bulunmuştur. Son olarak, SHAP tabanlı bulgular politika simülasyonları için bir başlangıç çerçevesi sağlayarak, belirli alt endekslerde hedeflenen iyileştirmelerin olası refah kazanımlarını nicel olarak öngörmeye imkân tanımaktadır.

Anahtar Kelimeler

Yaşam Kalitesi , Satın Alma Gücü , XGBoost , LSTM , SHAP , Açıklanabilir Yapay Zekâ

Kaynakça

• Abegaz, T. M., Ahmed, M., Ali, A. A., & Bhagavathula, A. S. (2025). Predicting Health-Related Quality of Life Using Social Determinants of Health: A Machine Learning Approach with the All of Us Cohort. Bioengineering, 12(2), 166.
• Acolin, A., & Reina, V. (2022). Housing cost burden and life satisfaction. Journal of Housing and the Built Environment, 37(4), 1789-1815.
• Alkaya, A. (2024). OECD Ülkelerinin Çok Boyutlu Ölçekleme Analiziyle Daha İyi Yaşam Endeksine Göre Konumları. İnsan ve Toplum Bilimleri Araştırmaları Dergisi, 13(3), 1130-1156.
• Arrieta, A. B., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., ... & Herrera, F. (2020). Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information fusion, 58, 82-115.
• Bai, S., Kolter, J. Z., & Koltun, V. (2018). An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv preprint arXiv:1803.01271.
• Bengio, Y., Simard, P., & Frasconi, P. (1994). Learning long-term dependencies with gradient descent is difficult. IEEE transactions on neural networks, 5(2), 157-166.
• Bingöl, U., & Ayhan, F. (2020). The Impact of NEET and Labor Market Indicators on Human Development: A Panel Data Analysis for EU-28 Countries. In Journal of Social Policy Conferences (No. 79, pp. 441-468). Istanbul University.
• Chen, T., & Guestrin, C. (2016, August). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785-794).
• Cummins, R. A. (1996). The domains of life satisfaction: An attempt to order chaos. Social Indicators Research, 38(3), 303–328.
• Diener, E., Oishi, S., & Tay, L. (2018). Advances in subjective well-being research. Nature human behaviour, 2(4), 253-260.
• Doshi-Velez, F., & Kim, B. (2017). Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608.
• Friedman, J. H. (2001). Greedy function approximation: a gradient boosting machine. Annals of statistics, 1189-1232.
• Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
• Jung, J. Y., & Yun, Y. H. (2025). The critical effects of self-management strategies on predicting cancer survivors’ future quality of life and health status using machine learning techniques. PLoS One, 20(8), e0330570.
• Kahneman, D., & Deaton, A. (2010). High income improves evaluation of life but not emotional well-being. Proceedings of the national academy of sciences, 107(38), 16489-16493.
• Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., ... & Liu, T. Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems, 30, 1-9.
• Luechinger, S. (2009). Valuing air quality using the life satisfaction approach. The Economic Journal, 119(536), 482-515.
• Lundberg, S. M., & Lee, S. I. (2017). A unified approach to interpreting model predictions. Advances in neural information processing systems, 30.
• Molnar, Christoph. Interpretable Machine Learning: A Guide for Making Black Box Models Explainable (2. ed.). 2022. ISBN: 9798411463330.
• Mostafa, F. (2025). A Statistical Framework for Model Selection in LSTM Networks. arXiv preprint arXiv:2506.06840.
• Numbeo (2025). Quality of Life Index methodology. https://www.numbeo.com/quality-of-life/indices_explained.jsp. Access Date: 12.05.2025.
• OECD. (2013). OECD guidelines on measuring subjective well-being. OECD Publishing.
• Pacione, M. (2003). Urban environmental quality and human wellbeing—a social geographical perspective. Landscape and urban planning, 65(1-2), 19-30.
• Panat, T., & Chandra, R. (2025). Global Ease of Living Index: a machine learning framework for longitudinal analysis of major economies. arXiv preprint arXiv:2502.06866.
• Petch, J., Di, S., & Nelson, W. (2022). Opening the black box: the promise and limitations of explainable machine learning in cardiology. Canadian Journal of Cardiology, 38(2), 204-213.
• Rahman, T., Mittelhammer, R. C., & Wandscheider, P. (2005). Measuring the quality of life across countries: A sensitivity analysis of well-being indices (No. 2005/06). WIDER Research Paper.
• Ribeiro, M. T., Singh, S., & Guestrin, C. (2016, August). " Why should i trust you?" Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining (pp. 1135-1144).
• Rizzo, T. (2019). A panel data study of the determinants of life expectancy in low income countries. Bryant University, 1-30.
• Shu, Z., Carrasco, R. A., García-Miguel, J. P., & Sánchez-Montañés, M. (2022). Multiple scenarios of quality of life index using fuzzy linguistic quantifiers: the case of 85 countries in numbeo. Mathematics, 10(12), 2091.
• Shwartz-Ziv, R., & Armon, A. (2022). Tabular data: Deep learning is not all you need. Information Fusion, 81, 84-90.
• Štrumbelj, E., & Kononenko, I. (2014). Explaining prediction models and individual predictions with feature contributions. Knowledge and information systems, 41(3), 647-665.
• Whoqol Group. (1998). Development of the World Health Organization WHOQOL-BREF quality of life assessment. Psychological medicine, 28(3), 551-558.
• Zhu, Z., Wang, C., Shi, L., Li, M., Li, J., Liang, S., ... & Xue, Y. (2024). Integrating machine learning and the SHapley Additive exPlanations (SHAP) framework to predict lymph node metastasis in gastric cancer patients based on inflammation indices and peripheral lymphocyte subpopulations. Journal of Inflammation Research, 9551-9566.