Additional document information (ADI) is information is needed for weighting schemes which take into account properties of whole documents. More...
#include <CAdditionalDocumentInformation.h>
Public Member Functions | |
| CAdditionalDocumentInformation (const string &inName="N.N.") | |
| Constructor sets filename. | |
| void | setFileNameBase (const string &inName) |
| ...If neccesary set filename later | |
| void | resetDF () |
| Reset mMaximumDF and mDFSquareSum. | |
| void | adjustDF (double inDF) |
| Add the right things to mMaximumDF and mDFSquareSum. | |
| void | resetSquareDFLogICF () |
| reset mSquareDFLogICFSum | |
| void | adjustSquareDFLogICF (double) |
| add the right things to mSquareDFLogICFSum | |
Accessors | |
| double | getMaximumDF () const |
| double | getDFSquareSum () const |
| double | getSquareDFLogICFSum () const |
| bool | output (ostream &outStream) const |
| bool | output () const |
| bool | input (istream &inStream) |
| bool | input () |
Protected Attributes | |
| string | mFileNameBase |
| Filename of the document (from which the ADI file will be built) | |
| double | mMaximumDF |
| Maximum Document frequency of a feature for the whole document. | |
| double | mDFSquareSum |
| Sum of squared document frequencies off all features of the document. | |
| double | mSquareDFLogICFSum |
| Sum of (DF*DF*log(ICF)) for all features of the document. | |
Additional document information (ADI) is information is needed for weighting schemes which take into account properties of whole documents.
Things like e.g. the euclidean length of a vector have to be calculated beforehand.