On 26 Mar 2016, at 06:51, Andria Lan
It's clear that Weka has several evaluation methods that are existed in the
"Classify" panel. Cross-validation applies the cross-validation method, while
"Percentage split" method splits the data into training and test sets based on
percentage ratio, however, my question:
What is the evaluation method technique that is used for the "Supplied test
set" option? Precisely, if I loaded 2 files; training set file through the
"Preprocess" panel, and test set file using "Supplied test set", what
evaluation method will be used to estimate instances class of the file that is located in
the "Supplied test set"?
The classifier will be built from the data that is loaded into the Preprocess panel and
used to classify the instances in the supplied test set.