DATA-BASED INTERVENTION OPPORTUNITIES IN SOCIAL SERVICES: CLUSTERING HOME CARE SERVICE RECIPIENTS USING THE K-MEANS ALGORITHM

Abstract

The aging population, increase in chronic diseases, and rise in the number of individuals living alone have all contributed to the escalating demand for home care services. While it is known that individuals who benefit from home care services exhibit different needs according to their individual characteristics, there is a limited number of data-driven studies that systematically analyze these differences. In this context, the current study aims to determine whether individuals receiving home care services exhibit similar patterns based on their demographic, socioeconomic, and service usage characteristics, thereby contributing to the development of targeted social work interventions. The study used the “Home Care Service Requests” dataset provided by the Izmir Metropolitan Municipality, covering the years 2020 to 2024. Before the analysis phase, records that were incomplete, inconsistent, or ambiguous were removed from the dataset, and the analyses were performed on a total of 5,245 valid observations. During the data preparation process, categorical variables were converted to numerical format using the one-hot encoding method, and then all variables with different scales were standardized using the StandardScaler method. In the application of the K-Means clustering algorithm, one of the methods used in the study, the optimal number of clusters was determined using the Elbow and Silhouette methods, and it was decided that the most appropriate cluster count for the analysis was five. The obtained clusters were visualized using Principal Component Analysis (PCA), and the characteristics of the clusters were interpreted based on average Z-scores. The analysis findings reveal that the clusters differ significantly based on variables such as age, gender, physical capacity, social security status, and type of service requests.

Keywords

• K-Means algorithm
• Unsupervised machine learning
• Data analysis
• Home care
• Social work

SOSYAL HİZMETLERDE VERİ TEMELLİ MÜDAHALE OLANAKLARI: EVDE BAKIM HİZMETİ ALAN BİREYLERİN K-MEANS ALGORİTMASI İLE KÜMELENMESİ

Abstract

Yaşlanan nüfus, kronik hastalıkların artışı ve yalnız yaşayan bireylerin çoğalması evde bakım hizmetlerine duyulan ihtiyacı her geçen gün artırmaktadır. Literatürde evde bakım hizmetlerinden yararlanan bireylerin bireysel özelliklerine göre farklı ihtiyaçlar sergilediği bilinmekle birlikte bu farklılıkların sistematik olarak analizine yönelik veri temelli çalışmalar sınırlıdır. Bu bağlamda mevcut araştırma evde bakım hizmeti alan bireylerin demografik, sosyoekonomik ve hizmet kullanım özelliklerine göre benzer örüntüler sergileyip sergilemediğini belirlemeyi ve bu yolla hedefe yönelik sosyal hizmet müdahaleleri geliştirilmesine katkı sunmayı amaçlamaktadır. Araştırmada İzmir Büyükşehir Belediyesi tarafından sağlanan ve 2020-2024 yıllarını kapsayan “Evde Bakım Hizmeti Talepleri” veri seti kullanılmıştır. Analiz aşamasından önce eksik, tutarsız ve anlam belirsizliği taşıyan kayıtlar veri setinden çıkarılmış ve analizler toplam 5245 geçerli gözlem üzerinden gerçekleştirilmiştir. Veri hazırlığı sürecinde kategorik değişkenler one-hot kodlama yöntemiyle sayısal formata dönüştürülmüş, ardından farklı ölçeklerdeki tüm değişkenler StandardScaler yöntemiyle standardize edilmiştir. Araştırmada kullanılan yöntemlerden olan K-Means kümeleme algoritmasının uygulanmasında optimum küme sayısı Elbow ve Silhouette yöntemleriyle belirlenmiş ve analiz için en uygun küme sayısının beş olduğuna karar verilmiştir. Elde edilen kümeler, Temel Bileşen Analizi (PCA) yardımıyla görselleştirilmiş ve kümelerin özellikleri ortalama Z-skorları üzerinden yorumlanmıştır. Analiz bulguları bireylerin yaş, cinsiyet, fiziksel yeterlilik, sosyal güvence durumu ve hizmet talepleri gibi değişkenler temelinde anlamlı şekilde farklılaşan kümeler oluşturduğunu ortaya koymaktadır.

Keywords

• K-Means algoritması
• Denetimsiz makine öğrenmesi
• Veri analizi
• Evde bakım hizmeti
• Sosyal hizmet

References

1. Aksakal, S. Ş., & Eğrioğlu, E. (2025). A research on the factors affecting the outcomes of child abuse cases using machine learning methods. Turkish Journal of Forecasting, 9(1), 18–29.
2. Allison, P. D. (2001). Missing data. Sage Publications.
3. Anbalagan, S., & Fathima, A. S. (2024). Leveraging artificial intelligence for social work: Opportunities and challenges. Indian Journal of Natural Sciences, 15(85), 79433–79436.
4. Armstrong, J. J., Zhu, M., Hirdes, J. P., & Stolee, P. (2012). K-means cluster analysis of rehabilitation service users in the Home Health Care System of Ontario: Examining the heterogeneity of a complex geriatric population. Archives of Physical Medicine and Rehabilitation, 93(12), 2198–2205. https://doi.org/10.1016/j.apmr.2012.05.026
5. Bacter, C., Săveanu, S. & Marc, C. (2021). Social work during the COVID-19 pandemic. Administrative and personal response to new challenges. Social Change Review, 19(1), 50-80. https://doi.org/10.2478/scr-2021-0004
6. Badell D., de Armas J., & Julià A. (2022). Impact of socioeconomic environment on home social care service demand and dependent users. International Journal of Environmental Research and Public Health, 19(4), 1-21. https://doi.org/10.3390/ijerph19042053
7. Başcıllar, M., Karataş, M., & Pak Güre, M. D. (2022). Dijital çağda sosyal algoritmalar: yapay zekâ ve sosyal hizmet. Sosyal Politika Çalışmaları Dergisi, 22(56), 539-565. https://doi.org/10.21560/spcd.vi.1081060
8. Bellis, M. A., Jarman, I., Downing, J., Perkins, C., Beynon, C., Hughes, K., & Lisboa, P. (2012). Using clustering techniques to identify localities with multiple health and social needs. Health & Place, 18(2), 138–143. https://doi.org/10.1016/j.healthplace.2011.08.003

9. Bholowalia, P., & Kumar, A. (2014). EBK-means: A clustering technique based on elbow method and k-means in WSN. International Journal of Computer Applications, 105(9), 17–24. https://doi.org/10.5120/18405-9674
10. Brand, J. E., Xu, J., Koch, B., & Geraldo, P. (2021). Uncovering sociological effect heterogeneity using tree-based machine learning. Sociol Methodol, 51(2), 189–223. https://doi.org/10.1177/0081175021993503
11. Bretos-Azcona, P. E., Sánchez-Iriso, E., & Cabasés Hita, J. M. (2020). Tailoring integrated care services for high-risk patients with multiple chronic conditions: A risk stratification approach using cluster analysis. BMC Health Services Research, 20, 806. https://doi.org/10.1186/s12913-020-05668-7
12. Castillo de Mesa, J. (2021). Digital social work: Towards digital disruption in social work. Journal of Sociology & Social Welfare, 48(3), 117–140.
13. Centre for Social Data Analytics. (2025, 10 Nisan). Allegheny family screening tool. https://csda.aut.ac.nz/research/our-projects/all-projects/allegheny-family-screening-tool
14. Cerrah, L., Karakuş Umar, E., Oral, M., & Pak, M. (2024). Evde bakım hizmeti alan yaşlı bireylerin ve bakım verenlerinin karşılaştıkları sosyal sorunlar üzerine nitel bir çalışma: Oltu örneği. Sosyal Politika Çalışmaları Dergisi, 24(62), 89–110. https://doi.org/10.21560/spcd.vi.1461554
15. Çoban, M., & Esatoğlu, A. E. (2004). Evde bakım hizmetlerine genel bir bakış. Türkiye Klinikleri Tıp Etiği-Hukuku-Tarihi Dergisi, 12(2), 109–120.
16. Erçorumlu, M. (2025). Dijital sosyal koruma için çok katmanlı bir yapay zeka modeli: Türkiye örneği. Sosyal Güvence (Özel Sayı), 187-209. https://doi.org/10.21441/sosyalguvence.1682319
17. Erdurak, Y. (2025). Z-score heatmap of variables across clusters. Figshare. https://doi.org/10.6084/m9.figshare.29483393.v1
18. GeeksforGeeks. (2023). Demonstration of k‑means assumptions. https://www.geeksforgeeks.org/machine-learning/demonstration-of-k-means-assumptions
19. Genet, N., Boerma, W., Kroneman, M., Hutchinson, A., & Saltman, R. (2011). Home care across Europe: Current structure and future challenges. WHO Regional Office for Europe.
20. Han, J., Kamber, M., & Pei, J. (2012). Data mining: Concepts and techniques (3rd ed.). Elsevier.
21. Henman, P. (2019). Of algorithms, apps and advice: Digital social policy and service delivery. Journal of Asian Public Policy, 12(1), 44–58.
22. İzmir Büyükşehir Belediyesi. (2025, 3 Mart). Evde bakım hizmeti talepleri veri seti. https://acikveri.bizizmir.com/dataset/ba4845f8-0611-46de-83b4-f728e352406f/resource/16a83c54-c513-47b3-93bb-c890d2244dd6/download/acik-veri-tablo.xlsx
23. Jacob, S., & Souissi, S. (2023). Social workers in the context of digital transformations: International literature review on the evolution of roles, skills, and professional competencies. Revue des Politiques Sociales et Familiales, (4), 207–225. https://doi.org/10.3917/rpsf.149.0207
24. Jain, A. K. (2010). Data clustering: 50 years beyond k-means. Pattern Recognition Letters, 31(8), 651–666. https://doi.org/10.1016/j.patrec.2009.09.011
25. Kuhn, M., & Johnson, K. (2019). Feature engineering and selection: A practical approach for predictive models. CRC Press.
26. Kurban, S. (2024). Sosyal hizmetin örgütlenmesinde ve politikaların geliştirilmesinde yapay zekâ katkısı ve Türkiye üzerine bir değerlendirme. Yönetim Bilimleri Dergisi, 22 (Özel Sayı: Endüstri 4.0 ve Dijitalleşmenin Sosyal Bilimlerde Yansımaları), 1667-1690. https://doi.org/10.35408/comuybd.1514386
27. Maleki, M., & Ghahari, S. (2024). Comprehensive clustering analysis and profiling of COVID-19 vaccine hesitancy and related factors across U.S. counties: Insights for future pandemic responses. Healthcare, 12(15), 1458. https://doi.org/10.3390/healthcare12151458
28. McKinney, W. (2010). Data structures for statistical computing in python. In Proceedings of the 9th Python in Science Conference, 56–61.
29. Minguela R., M. A., Lafuente, P. H., & Mota M., J. M. (2021). Home visit training in social work with virtual reality. Journal of Sociology & Social Welfare, 48(3), 53–73. https://doi.org/10.15453/0191-5096.4555
30. Miyahara, M. (2022). A pilot exploratory study to form subgroups using cluster analysis of family needs survey scores for providing tailored support to parents caring for a population-based sample of 5-year-old children with developmental concerns. International Journal of Environmental Research and Public Health, 19(2), 744. https://doi.org/10.3390/ijerph19020744
31. Molenaar, E.A., Barten, DJ.J., de Hoop, A., Bleijenberg, N., de Wit, N. J., & Veenhof, C. (2022). Cluster analysis of functional independence in community-dwelling older people. BMC Geriatr, 22(996), 1-11. https://doi.org/10.1186/s12877-022-03684-2
32. Naslund, J. A., Aschbrenner, K. A., Marsch, L. A., & Bartels, S. J. (2016). The future of mental health care: peer-to-peer support and social media. Epidemiology and Psychiatric Sciences, 25(02), 113–122. https://doi.org/10.1017/s2045796015001067
33. Nnoaham, K. E., & Cann, K. F. (2020). Can cluster analyses of linked healthcare data identify unique population segments in a general practice–registered population? BMC Public Health, 20(798), 1-10 https://doi.org/10.1186/s12889-020-08930-z
34. Nordesjö, K., & Scaramuzzino, G. (2023). Digitalization, stress, and social worker–client relationships during the COVID-19 pandemic. Journal of Social Work, 23(6), 1080-1098. https://doi.org/10.1177/14680173231180309
35. Nuwasiima, M., Ahonon, M. P., & Kadiri, C. (2024). The role of artificial intelligence (AI) and machine learning in social work practice. World Journal of Advanced Research and Reviews. 24(1), 80-97. https://doi.org/10.30574/wjarr.2024.24.1.2998
36. Patton, D. U., Mathiyazhagan, S., & Landau, A. Y. (2021). Meet them where they are: Social work informed considerations for youth inclusion in AI violence prevention. Social Work, 66(4), 357–360. https://doi.org/10.1093/sw/swab030
37. Pazer, S. (2024). Digitalization in social work: Opportunities and challenges for practitioners. International Journal of Research Publication and Reviews, 5(9), 3531–3536. https://doi.org/10.55248/gengpi.5.0924.2704
38. Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M. & Duchesnay, E. (2011). Scikit-learn: Machine learning in python. Journal of Machine Learning Research, 12, 2825–2830.
39. Petering, R., Um, M. Y., Alipourfard, N., Tavabi, N., Kumari, R., & Gilani, S. N. (2018). Artificial intelligence to predict intimate partner violence perpetration. In M. Tambe, & E. Rice. (Eds.), Artificial intelligence and social Work (pp. 195-210). Cambridge University Press
40. Pınar, R. (2010). Türkiye’de evde bakımda mevcut durum. 3. Akademik Geriatri Kongresi Bildiri Kitabı (ss. 153–156). https://akademikgeriatri.org/files/Akademik_Geriatri_2010/Konusma_Metinleri/16.pdf
41. Reamer, F. G. (2023). Artificial intelligence in social work: Emerging ethical issues. Journal of Social Work Values and Ethics, 20(2), 52–71. https://doi.org/10.55521/10-020-205
42. Recchia, D. R., Cramer, H., Wardle, J., Lee, D. J., Ostermann, T., & Lauche, R. (2022). Profiles and predictors of healthcare utilization: Using a cluster-analytic approach to identify typical users across conventional, allied and complementary medicine, and self-care. BMC Health Services Research, 22, 29. https://doi.org/10.1186/s12913-021-07426-9
43. Rice, E., & Tambe, M. (2018). Merging social work science and computer science for social good. In M. Tambe, & E. Rice. (Eds.). Artificial intelligence and social work (pp. 3-15). Cambridge University Press.
44. Rousseeuw, P. J. (1987). Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. Journal of Computational and Applied Mathematics, 20, 53-65.
45. Salganik, M. J. (2018). Bit by bit: Social research in the digital age. Princeton University Press.
46. Scikit-learn. (2025). StandardScaler. https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html
47. Srivastava, A., Petering, R., & Misyrlis, M. (2018). Minimizing violence in homeless youth. In M. Tambe, & E. Rice. (Eds.), Artificial intelligence and social work (pp. 119-135). Cambridge University Press.
48. Sterne, J. A. C., White, I. R., Carlin, J. B., Spratt, M., Royston, P., Kenward, M. G., Wood, A. M., & Carpenter, J. R. (2009). Multiple imputation for missing data in epidemiological and clinical research: Potential and pitfalls. BMJ, 338, 157-160. https://doi.org/10.1136/bmj.b2393
49. Tan, P.N., Steinbach, M., & Kumar, V. (2006). Introduction to data mining. Pearson Education.
50. Tan, Y., Xin, S. K., Zhang, R., Lee, J., Han, M., Sen, B., & Lee, Y.-C. (2025). Empowering social service with AI: Insights from a participatory design study with practitioners. In Extended Abstracts of the CHI Conference on Human Factors in Computing Systems (CHI EA ’25) (pp. 1–10). ACM. https://doi.org/10.1145/3706599.3719736
51. Tashobya, C. K., Dubourg, D., Ssengooba, F., Speybroeck, N., Macq, J., & Criel, B. (2016). A comparison of hierarchical cluster analysis and league table rankings as methods for analysis and presentation of district health system performance data in Uganda. Health Policy and Planning, 31(2), 217–228. https://doi.org/10.1093/heapol/czv045
52. Vuik, S. I., Mayer, E., & Darzi, A. (2016). Enhancing risk stratification for use in integrated care: A cluster analysis of high-risk patients in a retrospective cohort study. BMJ Open, 6(12). https://doi.org/10.1136/bmjopen-2016-012903
53. Wassal, K., Ashurst, C., Hron, J., & Zilka, M. (2024). Reimagining AI in social work: Practitioner perspectives on incorporating technology in their practice. https://doi.org/10.48550/arXiv.2407.10244
54. WHO. (2015). World report on ageing and health. WHO Press.
55. Wilder, B. (2018). Algorithmic social intervention. In Proceedings of the 27th International Joint Conference on Artificial Intelligence (IJCAI-18) (pp. 5793–5794). AAAI Press. https://doi.org/10.24963/ijcai.2018/837
56. Yoon, S., Goh, H., Kwan, Y. H., Thumboo, J., & Low, L. L. (2020). Identifying optimal indicators and purposes of population segmentation through engagement of key stakeholders: A qualitative study. Health Research Policy and Systems, 18(26), 1-13. https://doi.org/10.1186/s12961-019-0519-x
57. Zamora-Sánchez J. J., Zabaleta-Del-Olmo E., Fernández-Bertolín S., Gea-Caballero V., Julián-Rochina I, Pérez-Tortajada G., & Amblàs-Novellas J. (2021). Profiles of frailty among older people users of a home-based primary care service in an urban area of Barcelona (Spain): An observational study and cluster analysis. J Clin Med, 10(10), 1-12. https://doi.org/10.3390/jcm10102106
58. Zengin, O. (2018). Evde bakım hizmetleri. M. Oral (Ed.), Tıbbi Sosyal Hizmet içinde (ss. 224–243). Atatürk Üniversitesi AÖF Yayını.