Table 1.

The description of the data set obtained via different sampling methods. 50 per cent of all the candidates are used as training data which is denoted as ‘NonSampling’, while others are in a test set. The balanced training data set generated by undersampling, oversampling, and SMOTE are denoted as ‘UnderSample’, ‘OverSample’, and ‘SMOTE’, correspondingly.

DataSamplingPulsarNon-pulsarTotalRatio
TrainNonSampling59844 99845 5961:75
SMOTE44 99844 99889 9961:1
UnderSample59859811961:1
OverSample44 99844 99889 9961:1
Test59844 99845 5961:75
Total119689 99691 1921:75
DataSamplingPulsarNon-pulsarTotalRatio
TrainNonSampling59844 99845 5961:75
SMOTE44 99844 99889 9961:1
UnderSample59859811961:1
OverSample44 99844 99889 9961:1
Test59844 99845 5961:75
Total119689 99691 1921:75
Table 1.

The description of the data set obtained via different sampling methods. 50 per cent of all the candidates are used as training data which is denoted as ‘NonSampling’, while others are in a test set. The balanced training data set generated by undersampling, oversampling, and SMOTE are denoted as ‘UnderSample’, ‘OverSample’, and ‘SMOTE’, correspondingly.

DataSamplingPulsarNon-pulsarTotalRatio
TrainNonSampling59844 99845 5961:75
SMOTE44 99844 99889 9961:1
UnderSample59859811961:1
OverSample44 99844 99889 9961:1
Test59844 99845 5961:75
Total119689 99691 1921:75
DataSamplingPulsarNon-pulsarTotalRatio
TrainNonSampling59844 99845 5961:75
SMOTE44 99844 99889 9961:1
UnderSample59859811961:1
OverSample44 99844 99889 9961:1
Test59844 99845 5961:75
Total119689 99691 1921:75
Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close