demo56.cpp File Reference

Example 56: Revealing feature selection results' bias. More...

#include <boost/smart_ptr.hpp>
#include <exception>
#include <iostream>
#include <cstdlib>
#include <string>
#include <vector>
#include "error.hpp"
#include "global.hpp"
#include "subset.hpp"
#include "data_intervaller.hpp"
#include "data_splitter.hpp"
#include "data_splitter_5050.hpp"
#include "data_splitter_cv.hpp"
#include "data_scaler.hpp"
#include "data_scaler_void.hpp"
#include "data_accessor_splitting_memTRN.hpp"
#include "data_accessor_splitting_memARFF.hpp"
#include "criterion_wrapper.hpp"
#include "criterion_wrapper_bias_estimate.hpp"
#include "distance_euclid.hpp"
#include "classifier_knn.hpp"
#include "seq_step_straight.hpp"
#include "search_seq_dos.hpp"

Include dependency graph for demo56.cpp:

Functions
int	main ()

---

Detailed Description

Example 56: Revealing feature selection results' bias.

---

Function Documentation

int main

(

)

Example 56: Revealing feature selection results' bias

FST enables classifier bias estimation, i.e., estimation of the difference between classification accuracy on dependent (training) and independent (test) data. Values close to 0 are preferable. High estimated bias may indicate severe over-fitting on the evaluated feature subspace and thus may help to avoid poor performance on previously unseen data. In this example features are selected using DOS algorithm and 3-NN wrapper classification accuracy as FS criterion. Classification accuracy (i.e, FS wrapper criterion value) is estimated on the first 50% of data samples by means of 3-fold cross-validation. The final classification performance on the selected subspace is eventually validated on the second 50% of data.

References FST::Search_DOS< RETURNTYPE, DIMTYPE, SUBSET, CRITERION, EVALUATOR >::search(), FST::Search_DOS< RETURNTYPE, DIMTYPE, SUBSET, CRITERION, EVALUATOR >::set_delta(), and FST::Search< RETURNTYPE, DIMTYPE, SUBSET, CRITERION >::set_output_detail().