Please use this identifier to cite or link to this item: https://publication.npru.ac.th/jspui/handle/123456789/807
Full metadata record
DC FieldValueLanguage
dc.contributor.authorPhuttaraksa, Chaowat-
dc.contributor.authorSisang, Pawin-
dc.contributor.authorJuntiwad, Wassana-
dc.contributor.authorHengpraphorm, Supojn-
dc.contributor.authorHengpraphorm, Kairung-
dc.date.accessioned2020-10-12T07:13:19Z-
dc.date.available2020-10-12T07:13:19Z-
dc.date.issued2020-07-10-
dc.identifier.urihttps://publication.npru.ac.th/jspui/handle/123456789/807-
dc.description.abstractThe research aimed to study and compare the techniques classified information to the appropriate classification of diabetes. In this research was to study the technique selected for use in this study, all three techniques include decision trees techniques. Neural networks Technical and closest neighbor to find the composite images of classified information, the best value in the development of diabetes. The results showed that the technique is far better results, Neural networks. The accuracy of 76.82 with a value equal to 0.86 precision by providing accurate value that is equal to 0.80, followed by the nearest neighbor technique. The accuracy of 72.27 by 0.82 by the precision accuracy that is equal to 0.77, and the final decision trees. The accuracy of 71.09 with a value equal to 0.89 precision by providing accurate value that is equal to 0.73.en_US
dc.subjectdata classificationen_US
dc.subjectdecision treeen_US
dc.subjectk-NNen_US
dc.subjectartificial neural networken_US
dc.titleA Comparison of Data Classification Efficiency for Diabetes Mellitus using Data Mining Techniquesen_US
Appears in Collections:Proceedings of the 12th NPRU National Academic Conference

Files in This Item:
File Description SizeFormat 
The 12th NPRU_17.pdf270.39 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.