Improvement the Accuracy of Six Applied Classification Algorithms through Integrated Supervised and Unsupervised Learning ApproachReportar como inadecuado




Improvement the Accuracy of Six Applied Classification Algorithms through Integrated Supervised and Unsupervised Learning Approach - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

We have presented an integrated approachbased on supervised and unsupervised learning tech- nique to improve theaccuracy of six predictive models. They are developed to predict outcome oftuberculosis treatment course and their accuracy needs to be improved as theyare not precise as much as necessary. The integrated supervised and unsupervisedlearning method ISULM has been proposed as a new way to improve modelaccuracy. The dataset of 6450 Iranian TB patients under DOTS therapy wasapplied to initially select the significant predictors and then develop six predictivemodels using decision tree, Bayesian network, logistic regression, multilayerperceptron, radial basis function, and support vector machine algorithms.Developed models have integrated with k-mean clustering analysis to calculatemore accurate predicted outcome of tuberculosis treatment course. Obtainedresults, then, have been evaluated to compare prediction accuracy before andafter ISULM application. Recall, Precision, F-measure, and ROC area are othercriteria used to assess the models validity as well as change percentage toshow how different are models before and after ISULM. ISULM led to improve theprediction accuracy for all applied classifiers ranging between 4% and 10%.The most and least improvement for prediction accuracy were shown by logisticregression and support vector machine respectively. Pre-learning by k- meanclustering to relocate the objects and put similar cases in the same group canimprove the classification accuracy in the process of integrating supervisedand unsupervised learning.

KEYWORDS

ISULM; Integration Supervised and Unsupervised Learning; Classification; Accuracy; Tuberculosis

Cite this paper

Kalhori, S. and Zeng, X. 2014 Improvement the Accuracy of Six Applied Classification Algorithms through Integrated Supervised and Unsupervised Learning Approach. Journal of Computer and Communications, 2, 201-209. doi: 10.4236-jcc.2014.24027.





Autor: Sharareh R. Niakan Kalhori, Xiao-Jun Zeng

Fuente: http://www.scirp.org/



DESCARGAR PDF




Documentos relacionados