Abstract
Numerous data mining models have been proposed to construct computer-aided medical expert systems. Bayesian network classifiers (BNCs) are more distinct and understandable than other models. To graphically describe the dependency relationships among clinical variables for thyroid disease diagnosis and ensure the rationality of the diagnosis results, the proposed k-dependence causal forest (KCF) model generates a series of submodels in the framework of maximum spanning tree (MST) and demonstrates stronger dependence representation. Friedman test on 12 UCI datasets shows that KCF has classification accuracy advantage over the other state-of-the-art BNCs, such as Naive Bayes, tree augmented Naive Bayes, and k-dependence Bayesian classifier. Our extensive experimental comparison on 4 medical datasets also proves the feasibility and effectiveness of KCF in terms of sensitivity and specificity.
Highlights
Data mining [1] [2] is used to extract unknown but potentially useful information by using available incomplete, noisy, fuzzy, and random practical application data
Each instance consists of 29 attributes, which can be classified into 20 classes
In order to minimize the bias associated with the random sampling of the training and holdout data samples in comparing the classification accuracy of two or more methods, 10-fold cross-validation is applied to compare the general performance of k-dependence causal forest (KCF) with three Bayesian network classifiers (i.e., Naive Bayes (NB), tree augmented Naive Bayes (TAN) and k-dependence BNs (KDB)) and five non-Bayesian network classifiers, i.e., IBK(k-Nearest Neighbours) [22], SMO(Support Vector Machine) [23], MultilayerPerception(Artificial Neural Network) [24], DecisionStump(Decision Tree) [25] and SimpleLogistic(linear logistic regression) [26]
Summary
Data mining [1] [2] is used to extract unknown but potentially useful information by using available incomplete, noisy, fuzzy, and random practical application data. The medical domain consists of a considerable amount of data, including complete human genetic code information; clinical information on the history of patients, diagnosis, inspection, and treatment; and drug management information. Data mining can be applied in the medical field to analyze medical data, extract implicit valuable information, provide correct diagnosis and treatment, and study the genetic law of human diseases and health [3]. While dealing with a large amount of historical information of patients in the database, data mining needs to confirm the diagnosis based on age, gender, auxiliary examination results, and physiological and biochemical indicators of patients. Data mining should eliminate interference of human factors and establish diagnosis rules with good universality, provided that large amounts of data are analyzed in the process.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
