Performance Evaluation of the Tree Augmented Naïve Bayes Classifier for Knowledge Discovery in Healthcare Databases

• Mário W. L. Moreira UBI / IFCE
• Joel J. P. C. Rodrigues UBI / Inatel / UNIFOR
• Antonio M. B. Oliveira IFCE
• Kashif Saleem KSU
• Augusto J. Venâncio Neto UFRN

DOI: https://doi.org/10.5753/sbcas.2017.3730

Resumo

Smart decision support systems (DSSs) have been successfully employed in several areas. In healthcare, these systems offer solutions for uncertain reliably acts and moments. Systems based on Bayesian networks (BNs) can generate predictions even in information lack situations. This paper proposes the modeling and presents a performance evaluation study of the Bayesian classifier named Tree Augmented Naı̈ve Bayes (TAN). Results show that the proposed algorithm obtained good performance for a pregnancy database, presenting F-measure 0.92, Kappa statistic 0.8932, and ROC area 0.993. The proposed method allows representing more complex connections between variables. Nevertheless, it requires major computational effort and time that are not needed in other Bayesian algorithms.

Referências

Al-Khasawneh, A. (2016). Diagnosis of breast cancer using intelligent information systems techniques. International Journal of E-Health and Medical Communications (IJEHMC), 7(1):65–75.

Anand, T., Pal, R., and Dubey, S. K. (2016). Data mining in healthcare informatics: Techniques and applications. In 2016 3rd International Conference on Computing for Sustainable Global Development (INDIACom), March 16-18, New Delhi, India, pages 4023–4029. IEEE.

Arisha, A. and Rashwan, W. (2016). Modeling of healthcare systems: past, current and future trends. In Proceedings of the 2016 Winter Simulation Conference, December 11-14, Washington, DC, USA, pages 1523–1534. IEEE Press.

Arshad, A., Noordin, M. F. B., and Othman, R. B. (2016). A comprehensive knowledge management process framework for healthcare information systems in healthcare industry of pakistan. In 2016 6th International Conference on Information and Communication Technology for The Muslim World (ICT4M), November 22-24, Jakarta, Indonesia, pages 30–35. IEEE.

Chakraborty, C., Gupta, B., and Ghosh, S. K. (2016). Chronic wound characterization using bayesian classifier under telemedicine framework. International Journal of E-Health and Medical Communications (IJEHMC), 7(1):76–93.

Chiroma, H., Gital, A., Abubakar, A., and Zeki, A. (2014). Comparing performances of markov blanket and tree augmented naive-bayes on the iris dataset. In International MultiConference of Engineers and Computer Scientists (IMECS), March 12-14, Hong Kong, China, pages 1–4.

Duda, R. O., Hart, P. E., and Stork, D. G. (2012). Pattern classification. John Wiley & Sons.

Huang, K., King, I., and Lyu, M. R. (2002). Constructing a large node chow-liu tree based on frequent itemsets. In Proceedings of the 9th International Conference on Neural Information Processing (ICONIP’02), November 18-22, Singapore, volume 1, pages 498–502. IEEE.

Khan, K. S., Wojdyla, D., Say, L., Gülmezoglu, A. M., and Van Look, P. F. (2006). Who analysis of causes of maternal death: a systematic review. The lancet, 367(9516):1066–1074.

Kovács, G. and Hajdu, A. (2011). Extraction of vascular system in retina images using averaged one-dependence estimators and orientation estimation in hidden markov random fields. In 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, March 30 - April 2, Chicago, IL, USA, pages 693–696. IEEE.

Magee, L. A., Pels, A., Helewa, M., Rey, E., von Dadelszen, P., Committee, S. H. G., et al. (2014). Diagnosis, evaluation, and management of the hypertensive disorders of pregnancy: executive summary. Journal of Obstetrics and Gynaecology Canada, 36(7):575–576.

Manirabona, A., Fourati, L. C., and Boudjit, S. (2017). Investigation on healthcare monitoring systems: Innovative services and applications. International Journal of E-Health and Medical Communications (IJEHMC), 8(1):1–18.

Moore, L., Kambhampati, C., and Cleland, J. G. (2014). Classification of a real live heart failure clinical dataset-is tan bayes better than other bayes? In 2014 IEEE International Conference on Systems, Man and Cybernetics (SMC), October 5-8, San Diego, CA, USA, pages 882–887. IEEE.

Moreira, M. W., Rodrigues, J. J., Oliveira, A. M., Saleem, K., and Neto, A. V. (2016). An inference mechanism using bayes-based classifiers in pregnancy care. In 2016 IEEE 18th International Conference on e-Health Networking, Applications and Services (Healthcom), September 14-17, Munich, Germany, pages 1–5. IEEE.

Mushcab, H., Kernohan, W. G., Wallace, J., Harper, R., and Martin, S. (2017). Self-management of diabetes mellitus with remote monitoring: A retrospective review of 214 cases. International Journal of E-Health and Medical Communications (IJEHMC), 8(1):52–61.

Orphanou, K., Dagliati, A., Sacchi, L., Stassopoulou, A., Keravnou, E., and Bellazzi, R. (2016). Combining naive bayes classifiers with temporal association rules for coronary heart disease diagnosis. In 2016 IEEE International Conference on Healthcare Informatics (ICHI), October 4-7, Chicago, IL, USA, pages 81–92. IEEE.

Raffaeli, L., Spinsante, S., and Gambi, E. (2016). Integrated smart tv-based personal e-health system. International Journal of E-Health and Medical Communications (IJEHMC), 7(1):48–64.

Ren, P. (2015). A tree augmented naive bayesian network experiment for breast cancer prediction. ArXiv e-prints, pages 1–24.

Wan, C. and Freitas, A. A. (2016). A new hierarchical redundancy eliminated tree augmented naive bayes classifier for coping with gene ontology-based features. In International Conference on Machine Learning (ICML 2016), June 19-24, New York, NY, USA, pages 1–5.

Wang, J., Hu, X., and Zhu, D. (2008). Minimizing the minus sides of mining data. In Data Mining and Knowledge Discovery Technologies, pages 254–279. IGI Global.