SUMMA Semantic Scorer

Functionality

Adds to each sentence a feature ('sem_score') representing the relevance of the named entities (Person, Organization, Date, etc.) contained in the sentence. A sentence with more entities has a grater score. It also adds features for each entity type considered as relevant.

Parameters of the Resource

• annSetName: the annotation set where the annotations live
• semTypes: a list of named entities to be used to score relevance (these can be computed with default GATE modules for English or with customized modules for other languages). You can customize the list at will.
• sentAnn: the name of the annotation for which you want to compute the feature (e.g. Sentence)

Restriction

Named entities of the required types may be present in the document.