Compute and store the PLSA thesarus. More...
#include <PLSAFloatThesaurus.h>
Public Member Functions | |
| PLSAFloatThesaurus (int totalWords, BlockStats *stats, SortedSubIndex< WordSubIndex > *index, FileName *fileName, int eigenvalues, int iterations) | |
| Compute each of the PLSA thesaurus values. | |
| bool | calculateRow (int currentRealWord) |
| Calculate the row and return TRUE if it should be stored. | |
| int | listRealPosition (int position) |
| Return the real list position (based on the initial index). | |
| int | indexRealPosition (int position) |
| Return the real index position (based in the initial index). | |
| Quantise * | createQuantiser (float lower, float upper) |
| Assign a quantiser to convert floats to ints. | |
Compute and store the PLSA thesarus.
This class is used to compute the Probabilistic Latent Semantic Analysis (PLSA) values from the document index, and then store the values in a compressed thesarus. This class makes use of the FloatIndex base class.
1.6.1
