A Comparative Study of Data Classification Algorithms for Survival Prognosis in Breast Cancer Patients

• Manuelly Victor UFRN
• Isa Pereira UFRN
• Felipe Silva UFRN
• Erick Trindade UFRN
• Flavius Gorgônio UFRN
• Karliane Vale UFRN
• Yasmin Rebeca UERN
• Maria de Lourdes Morais UERN

DOI: https://doi.org/10.5753/ercas.2024.238523

Abstract

The use of classification algorithms has been an important ally in activities that require prognostics, including in the healthcare field. An example of this is the use of these algorithms in analyzing the survival time of patients diagnosed with cancer. In this work, a comparative analysis is conducted between the Naive Bayes, J48, and KNN algorithms to verify the effectiveness of the treatment concerning the patient’s survival time. To carry out this study, a database of patients diagnosed with breast cancer was used, containing behavioural data and the respective treatments to which the patients were subjected. Preliminary results indicate that the J48 algorithm outperformed the other tested algorithms in all analyzed metrics (accuracy, precision, recall, and F1-Score), demonstrating its potential to assist in research involving the planning of treatments for patients based on survival time.

References

Faceli, K. (2011). Inteligência artificial: uma abordagem de aprendizado de máquina. Grupo Gen - LTC.

Fonseca, A. S. M. et al. (2021). Mineração de dados de problemas relacionados a medicamentos registrados pela farmácia clínica de um hospital universitário. Arquivos Catarinenses de Medicina, 50(2):142–155.

Santos, P. D. d., Yahata, E., Piheiro, T. S., Oliveira, F. S. d., and Simões, P. W. (2023). Algoritmos de machine learning para predição da sobrevida do câncer de mama. Journal of Health Informatics, 15(Especial):1–12.

Silva, M. E. A. (2022). Proposta e avaliação de um modelo híbrido de seleção de características para o prognóstico do câncer de mama. Master’s thesis, Universidade Federal de Alagoas, Maceió, AL.

Wickeinecki, A. S. Z. and Viegas, S. C. (2020). A tecnologia e saúde uma simbiose. ReFAQI-Revista de Gestão Educação e Tecnologia, 8(2):1–18.