Machine Learning Based Predictive Models and Its Performance Evaluation on Healthcare Systems
In previous days, the growth in various technologies and the huge data generation has produced a drastic development in database and sources. Medicinal area characterizes a rich data field. A wide-ranging quantity of medicinal data is presently offered, reaching from details of medical indications to several kinds of medicinal data and creation of image capturing components. The manual extraction of medicinal designs is a hard job due to the nature of medicinal field includes enormous, lively, and difficult information. DM is accomplished to improve the value to extract medicinal designs. This paper outlines the applications of DM on the groups of diseases is projected. The key effort is to examine machine learning (ML) models commonly utilized for the prediction, prognosis and treatment of significant standard diseases like heart diseases, cancer, hepatitis, and diabetes. A set of different classifier models is applied to examine their efficiency. This examination distributes a complete investigation of the recent position of diagnosing diseases by the use of ML models. The attained detection rate of the numerous applications ranges between 70–100% based on diseases, utilized data and methods.
KeywordsClassification Disease diagnosis Diabetes Machine learning
- 1.National Library of Medicine (2017). http://www.nlm.nih.gov/tsd/acquisitions/cdm/subjects58.html. Accessed 14 Nov 2017
- 2.American Medical Informatics Association (2017). http://www.amia.org/informatics/. Accessed 12 Nov 2017
- 3.World Health Organization: Cardiovascular diseases (CDV’s) (2017). http://www.who.int/mediacentre/factsheets/fs317/en/. Accessed 15 May 2017
- 4.World Health Organization: Cancer (2017). http://www.who.int/mediacentre/factsheets/fs297/en/. Accessed 13 Feb 2017
- 5.Medjahed, S.A., Saadi, T.A., Benyettou, A.: Breast cancer diagnosis by using k-nearest neighbour with different distances and classification rules. Int. J. Comput. Appl. 62(1), 1–5 (2013)Google Scholar