The article titled Data Mining and Clinical Decision Support System by Michael Hardin and David Chhieng provides an insight into a process of searching and converting data. First of all, authors define the notion of data mining in its broader terms. It is aimed at transforming the data into meaningful information and at searching for new patterns for the users. Pattern recognition is a part of the data mining. It was developed to look for analytical models, which would be able to classify data. Data mining is also related to the clinical decision support system. The progress in computing and medical technology resulted in many applications, where decision support system and data mining are used. The decision support systems are computer-based tools aimed at the decision making. Typically, such system consists of the data management, the model management, the knowledge engine, the user interface and the users (Hardin & Chhieng 45).
The decision support system with data mining tools does not require prior knowledge. Further authors provide information of learning from data, which can be supervised and unsupervised. The first type is also called directed data mining. It is aimed at developing of predictive model between the dependent and independent variables. In supervised learning the training set may be unrepresentative regarding the general population and require adjustment. Unsupervised learning target variables and less information is provided at the beginning. Therefore, the authors focus on the supervised learning, which require the proper choice of classifier. The decision tree system is often used in such type of learning. The more complicated systems are logistic regression and neural networks development. The classifiers are evaluated though receiver operating characteristic graphs and Kolmogorov-Smirnov Tests. Unsupervised training is based on cluster analysis. The authors provide a lot of examples and include a short characteristic of other techniques such as genetic algorithms and biological computing.
This article is very useful in the learning process and in the future perspective. It provides a lot of information on data mining and decision support systems and their interrelation. Also, the article is full of examples, which are aimed how the described systems can be used in healthcare. I have learnt that the clinical decision support system has applications in many fields of healthcare including cancer diagnosis and screening. Besides, such systems could help to estimate the number of hospitalized patients and take the necessary measures. Also, the decision support systems are helpful in improving the managerial decisions. Furthermore, the data mining is very important for the decision making process. The availability of precise information and proper classifiers is crucial for developing a working decision support system. Besides, the unsupervised and supervised learning can be applied to different areas depending on the amount of information and variables you have.
The article is related to the healthcare informatics and its studies. The subject deals with the use of computer technology to the healthcare and the computerization of the whole healthcare system. The article provides an insight into the real life application of the computer-based decision systems in the corresponding field. Furthermore, article can be used a practical material as it has a lot of examples. Besides, authors focus on the statistical pattern recognition and data mining process. These fields are important part of the healthcare informatics as they are aimed at digitalization of the system.
Works Cited
Hardin, J. Michael, and David C. Chhieng. "Data mining and clinical decision support systems." Clinical Decision Support Systems. Springer New York, 2007. 44-63.
