We need to take into account the feature relevance in the pre-diagnosis (ej. omeda) in the deparsing, so that some features might be more relevant than others.
Assigning weights to features in the data source configuration file is an implemented choice. Weights.dat file is generated after parsing process (if no weight is specified, it is automatically set to one).
We need to take into account the feature relevance in the pre-diagnosis (ej. omeda) in the deparsing, so that some features might be more relevant than others.