During the past years, the increase in scientific knowledge and the massive data production have caused an exponential growth in databases and repositories. Biomedical domain represents one of the rich data domains. An extensive amount of biomedical data is currently available, ranging from details of clinical symptoms to various types of biochemical data and outputs of imaging devices. Manually extracting biomedical patterns from data and transforming them into machine-understandable knowledge is a difficult task because biomedical domain comprises huge, dynamic, and complicated knowledge. Data mining is capable of improving the quality of extracting biomedical patterns. In this research, an overview of the applications of data mining on the management of diseases is presented. The main focus is to investigate machine learning techniques (MLT) which are widely used to predict, prognose and treat important frequent diseases such as cancers, hepatitis and heart diseases. The techniques namely Artificial Neural Network, K-Nearest Neighbour, Decision Tree, and Associative Classification are illustrated and analyzed. This survey provides a general analysis of the current status of management of diseases using MLT. The achieved accuracy of the various applications ranged from 70% to 100% according to the disease, the solved problem, and the used data and technique.
- Klíčová slova
- rozhodovací strom, Algoritmus k-nejbližších sousedů, associative classification,
- MeSH
- algoritmy MeSH
- data mining MeSH
- databáze jako téma MeSH
- hepatitida * diagnóza klasifikace MeSH
- kardiovaskulární nemoci * diagnóza klasifikace MeSH
- lékařská informatika MeSH
- lidé MeSH
- management nemoci MeSH
- nádory * diagnóza klasifikace MeSH
- neuronové sítě MeSH
- prognóza MeSH
- správnost dat MeSH
- strojové učení * klasifikace MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- přehledy MeSH
