1
$\begingroup$

The forest coverType Dataset Contains the following attributes distributed in many boolean features.

Wilderness_Area (4 binary columns) / qualitative / 0 (absence) or 1 (presence) / Wilderness area designation
Soil_Type (40 binary columns) / qualitative / 0 (absence) or 1 (presence) / Soil Type designation 

Training the SVC classifier with RBF kernel without attributes re-engineering takes too much time. Can we merge all 40 binary columns for form a Soil_Type Attribute representing those columns. What are the pros and cons of this approach ?

Also, do we have any direct approach to achieve this in weka or sklearn ?

$\endgroup$
1
$\begingroup$

On Some Analysis I found out the columns were mutually Exclusive and therefore created one column named Soil_Type(0..40 possible values) replacing 40 binary columns and Wilderness_Area(0..4 possible values) replacing 4 binary columns. This did improve performance of the classifiers (Random Forest, Decision Tree ) with respect to time and prediction accuracy as well.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.