System documentation of the GNU Image-Finding Tool

CComparisonResults Class Reference

The general scheme of implementing a PicHunter like algorithm in GIFT uses the following scheme:. More...

#include <CComparisonResults.h>

List of all members.

Public Member Functions

 CComparisonResults (lCContainerType &inContainer, const CWeightingFunction &inPrototype, TID inID, const CRelevanceLevel &inQuery)
double calculateImageScore (TID inDID, const CDocumentFrequencyList &inList) const
double addImageScore (TID inDID, const CDocumentFrequencyList &inList) const
CBitVector compare (const CComparisonResults &inResults) const
double getAsymmetry (const CComparisonResults &inResults) const
CWeightingFunctionPointerHashgetWeighters ()
void setRelevanceLevel (double)
TID getID () const
CRelevanceLevel const & getQuery () const


Detailed Description

The general scheme of implementing a PicHunter like algorithm in GIFT uses the following scheme:.

If you want a suggestion of n< > elements get 2n< > elements. Call them the s_i.

For the same dbSize/10 (or something) elements (which we will call the e_i) calculate the distance (s_i,d_j) for all i and j possible. (Leading to something like 500 comparisons in our case).

For each s_i the delta_{ij} will be stored in a structure of the type CComparisonResults


The documentation for this class was generated from the following file:

Need for discussion? Want to contribute? Contact
help-gift@gnu.org Generated using Doxygen