SUMMA Title Sentence Similarity

Functionality

Adds to each sentence a feature ('title_sim' b default) representing the similarity of the title to the sentence. The cosine between a vector of the title and the vector of the sentence is the value stored in the feature.

Parameters of the Resource

• annSet: the annotation set where the sentences are.
• titleAnnSet: the annotation where the only vector of the title is to be found.
• sentAnn: the name of the annotation for which you want to compute the feature (e.g. Sentence)
• vector: the name of the vector for sentence and title

Restriction

There should be an annotation set with the title and a vector. This can usually be done by transfering tokens with statistics to the title annotation set and then using a