Main Article Content

Abstract

Because of the huge information in biomedical and healthcare communities, correct study of medical data benefits early disease detection, community services and patient care. The exactness of study is reduced when the value of medical data is incomplete. Moreover, various regions exhibit unique appearances of particular regional diseases, those results in weakening the prediction of disease outbreaks. In the proposed system, it provides machine learning algorithms for effective prediction of thyroid disease occurrences in disease-frequent societies. It experiment the changed models over real-life hospital data collected. To overcome the difficulty of incomplete data, it uses a latent factor model to rebuild the missing data. It experiment on a thyroid diseases using structured and unstructured data from hospital it use Naïve Bayes and MLP algorithm. Compared to numerous approximation algorithms, the calculation exactness of our proposed MLP algorithm reaches 98.8% with a convergence speed which is faster than that of the Naïve Bayes algorithm on disease risk prediction on thyroid using Weka tool.

Article Details