Klasifikasi Risiko No-Show Pasien Berdasarkan Karakteristik Penjadwalan dan Sosiodemografi Menggunakan Kombinasi Instance Hardness Threshold dan Symbolic Regression

Meidhika Nawa Sapta; Arief Hermawan; Donny Avianto

doi:10.70340/jirsi.v5i3.501

Penulis

Meidhika Nawa Sapta Universitas Teknologi Yogyakarta
Arief Hermawan Universitas Teknologi Yogyakarta
Donny Avianto Universitas Teknologi Yogyakarta

DOI:

https://doi.org/10.70340/jirsi.v5i3.501

Kata Kunci:

no-show pasien, Instance Hardness Threshold, Symbolic Regression, interpretabilitas, machine learning kesehatan

Abstrak

Ketidakhadiran pasien pada janji medis terjadwal (no-show) menyebabkan inefisiensi sumber daya, gangguan kontinuitas perawatan, dan kerugian finansial dalam sistem pelayanan kesehatan. Dataset janji medis yang umumnya tidak seimbang (imbalanced) menyulitkan identifikasi kelas minoritas no-show secara akurat melalui pendekatan klasifikasi konvensional. Di sisi lain, sebagian besar model prediksi yang ada bersifat black-box sehingga tidak mampu menghasilkan penjelasan yang dapat langsung ditindaklanjuti oleh tenaga kesehatan secara operasional. Penelitian ini mengusulkan pipeline klasifikasi dua tahap yang mengintegrasikan Instance Hardness Threshold (IHT) dan Symbolic Regression (SR) berbasis pemrograman genetika. Dataset yang digunakan adalah no-show Appointments Kaggle (N=110.527 rekaman, Vitória, Brasil) dengan 14 fitur sosiodemografi dan penjadwalan. IHT mengeliminasi instance mayoritas yang ambigu secara selektif, sementara SR menghasilkan formula matematika eksplisit yang dapat dibaca dan diterjemahkan langsung menjadi aturan prioritas intervensi tanpa memerlukan alat interpretabilitas tambahan. IHT mereduksi rasio ketidakseimbangan dari 3,95:1 menjadi 1,08:1 dengan menghapus 44.790 instance. SR mengidentifikasi waiting_days dan same_day_appt sebagai pasangan fitur terkritik, menghasilkan formula simbolik eksplisit dengan recall 92%, F1-Score 0,43, dan ROC-AUC 0,67 pada data uji, kompetitif terhadap Gradient Boosting terbaik (F1=0,43, ROC-AUC=0,73) dengan keunggulan interpretabilitas yang signifikan. Formula SR secara langsung dapat dioperasionalkan sebagai aturan prioritas pengiriman pengingat, pasien dengan waktu tunggu panjang yang bukan janji hari yang sama diprioritaskan untuk intervensi, sementara pasien janji hari yang sama menunjukkan risiko no-show mendekati nol. Penelitian lanjutan direkomendasikan mencakup eksplorasi multi-objective SR, penambahan fitur kontekstual, serta validasi pada dataset klinis Indonesia.

Unduhan

Data unduhan belum tersedia.

Referensi

D. Liu et al., “Machine Learning Approaches to Predicting No-Shows in Pediatric Medical Appointment,” NPJ Digit. Med., vol. 5, no. 1, pp. 1–11, Dec. 2022, doi: 10.1038/s41746-022-00594-w.

Q. A. Hathaway, N. Yanamala, T. Narumanchi, and J. Narumanchi, “Prediction of Pediatric No-Show Rates Before and During the COVID-19 Pandemic Utilizing Machine Learning,” medRxiv, pp. 1–2, Nov. 2024, doi: 10.1101/2024.11.23.24317836.

D. Valero-Bover et al., “Reducing Non-Attendance in Outpatient Appointments: Predictive Model Development, Validation, and Clinical Assessment,” BMC Health Serv. Res., vol. 22, no. 451, pp. 1–9, Dec. 2022, doi: 10.1186/s12913-022-07865-y.

A. Russotto et al., “Understanding No-Show Patterns in Healthcare: A Retrospective Study from Northern Italy,” Healthcare (Switzerland), vol. 13, no. 15, pp. 1–10, Jul. 2025, doi: 10.3390/HEALTHCARE13151869.

A. F. A. Hamdan and A. A. Bakar, “Machine Learning Predictions on Outpatient No-Show Appointments in a Malaysia Major Tertiary Hospital,” Malaysian Journal of Medical Sciences, vol. 30, no. 5, pp. 169–180, Oct. 2023, doi: 10.21315/mjms2023.30.5.14.

A. M. Husein and M. K. Harahap, “Teknik Data Science di Bidang Kesehatan: Menganalisis Faktor penting Ketidakhadiran Janji Medis Pasien,” Data Sciences Indonesia (DSI), vol. 1, no. 2, pp. 50–61, Jan. 2022, doi: 10.47709/dsi.v1i2.1314.

C. Deina, F. S. Fogliatto, G. J. C. da Silveira, and M. J. Anzanello, “Decision Analysis Framework for Predicting No-Shows to Appointments Using Machine Learning Algorithms,” BMC Health Serv. Res., vol. 24, no. 37, pp. 1–17, Dec. 2024, doi: 10.1186/s12913-023-10418-6.

L. H. A. Salazar, V. R. Q. Leithardt, W. D. Parreira, A. M. da Rocha Fernandes, J. L. V. Barbosa, and S. D. Correia, “Application of Machine Learning Techniques to Predict A Patient’s No-Show in The Healthcare Sector,” Future Internet, vol. 14, no. 3, pp. 1–21, Dec. 2022, doi: 10.3390/fi14010003.

J. Joseph, S. Senith, A. Alfred Kirubaraj, and S. R. Jino Ramson, “Machine Learning for Prediction of Clinical Appointment No-Shows,” International Journal of Mathematical, Engineering and Management Sciences, vol. 7, no. 4, pp. 558–574, Aug. 2022, doi: 10.33889/IJMEMS.2022.7.4.036.

G. Fan, Z. Deng, Q. Ye, and B. Wang, “Machine Learning Based Prediction Models for Patients No-Show in Online Outpatient Appointments,” Data Science and Management, vol. 2, pp. 45–52, Jun. 2021, doi: 10.1016/j.dsm.2021.06.002.

A. Alshammari, R. Almalki, and R. Alshammari, “Developing a Predictive Model of Predicting Appointment No-Show by Using Machine Learning Algorithms,” Journal of Advances in Information Technology, vol. 12, no. 3, Aug. 2021.

K. M. Toffaha, M. C. E. Simsekler, M. A. Omar, and I. ElKebbi, “Predicting Patient No-Shows Using Machine Learning: A Comprehensive Review and Future Research Agenda,” Intell. Based. Med., vol. 11, pp. 2666–5212, Feb. 2025, doi: 10.1016/j.ibmed.2025.100229.

E. F. Agyemang et al., “Addressing Class Imbalance Problem in Health Data Classification: Practical Application From an Oversampling Viewpoint,” Applied Computational Intelligence and Soft Computing, vol. 2025, no. 1, pp. 1–0, 2025, doi: 10.1155/acis/1013769.

S. Al-Ahmari and F. Nadeem, “Improving Surgical Site Infection Prediction Using Machine Learning: Addressing Challenges of Highly Imbalanced Data,” Diagnostics, vol. 15, no. 4, pp. 1–34, Feb. 2025, doi: 10.3390/diagnostics15040501.

M. Yadav Mamilla, R. Al-Haddad, and S. Chowdhury, “Resampling Imbalanced Healthcare Data for Predictive Modelling,” IJACSA) International Journal of Advanced Computer Science and Applications, vol. 16, no. 2, p. 2025, 2025, [Online]. Available: www.ijacsa.thesai.org. [Accessed: Apr. 23, 2026]

A. Malik and C. Tepe, “Improving Early Diabetes Diagnosis, A Machine Learning Approach for Class Imbalance Mitigation,” International Journal of Scientific Research in Computer Science, Engineering and Information Technology, vol. 11, no. 3, pp. 861–874, Jun. 2025.

J. Dunstan et al., “Predicting No-Show Appointments in A Pediatric Hospital in Chile Using Machine Learning,” Health Care Manag. Sci., vol. 26, no. 2, pp. 313–329, Jan. 2023, doi: 10.1007/s10729-022-09626-z.

M. R. Smith, T. Martinez, and C. Giraud-Carrier, “An instance Level Analysis of Data Complexity,” Mach Learn, vol. 95, no. 2, pp. 225–256, 2014, doi: 10.1007/s10994-013-5422-z.

W. G. La Cava et al., “A Flexible Symbolic Regression Method for Constructing Interpretable Clinical Prediction Models,” NPJ Digit. Med., vol. 6, no. 107, pp. 1–14, Dec. 2023, doi: 10.1038/s41746-023-00833-8.

F. O. de Franca et al., “Interpretable Symbolic Regression for Data Science: Analysis of the 2022 Competition,” pp. 1–13, Jul. 2023, [Online]. Available: http://arxiv.org/abs/2304.01117. [Accessed: Jun. 05, 2026]

D. Ferrari, V. Guidetti, Y. Wang, and V. Curcin, “Multi-objective Symbolic Regression to Generate Data-driven, Non-fixed Structure and Intelligible Mortality Predictors using EHR: Binary Classification Methodology and Comparison with State-of-the-art,” Germany, 2022. [Online]. Available: https://github.com/davideferrari92/multiobjective_symbolic_regression. [Accessed: Jun. 05, 2026]

L. H. A. Salazar, W. D. Parreira, A. M. da R. Fernandes, and V. R. Q. Leithardt, “No-Show in Medical Appointments with Machine Learning Techniques: A Systematic Literature Review,” Information (Switzerland), vol. 13, no. 11, pp. 1–13, Nov. 2022, doi: 10.3390/info13110507.

A. Chatzimparmpas, F. V. Paulovich, and A. Kerren, “HardVis: Visual Analytics to Handle Instance Hardness Using Undersampling and Oversampling Techniques,” Computer Graphics Forum, vol. 42, no. 1, pp. 135–154, Feb. 2023, doi: 10.1111/cgf.14726.

Y. Alabdulkarim, M. Almukaynizi, A. Alameer, B. Makanati, R. Althumairy, and A. Almaslukh, “Predicting No-Shows for Dental Appointments,” PeerJ Comput. Sci., vol. 8, pp. 1–24, Nov. 2022, doi: 10.7717/PEERJ-CS.1147.