When converting the original Heart Disease data set data file Cleveland.data, containing
all 76 attributes to an csv-file, that was to be inserted and run in WEKA,
21 instances (patients) needed to be removed because they contained corrupt data.
To get a better understanding of how this would affect our results, a comparison
was made between the original Cleveland data set (with all 303 instances) and the
Cleveland data set (with 282 instances), that is used in all other tests in this report.
When running tests with both of these files, the original Cleveland data set produced
higher accuracy of correctly classified instances as can be seen in the table in
Figure 4.1.
When converting the original Heart Disease data set data file Cleveland.data, containingall 76 attributes to an csv-file, that was to be inserted and run in WEKA,21 instances (patients) needed to be removed because they contained corrupt data.To get a better understanding of how this would affect our results, a comparisonwas made between the original Cleveland data set (with all 303 instances) and theCleveland data set (with 282 instances), that is used in all other tests in this report.When running tests with both of these files, the original Cleveland data set producedhigher accuracy of correctly classified instances as can be seen in the table inFigure 4.1.
การแปล กรุณารอสักครู่..
