From a dataset about arrhythmias and the study on pre-processing, the repository aims to present and discuss the results obtained using the Decision Tree model J48 and the SVM Linear model in order to classify the mentioned data.
-
Describse the database attributes, numerical and categorical, classifying according to scale (nominal or ratio) and cardinality (discreet, continuous, binary).
-
Describes each of the attributes according to frequency, minimum and maximum value, day standard deviations, as appropriate.
-
After the classification process with J48 has been used, it also evaluates the results of the processes below:
🔸 Treat missing values
🔸 Recognizing outliers
🔸 Normalization
In order to compare the different outcomes, we built a accuracy x method table for both Decision Tree and SVM Linear: