catalpa-cl / escrito

Apache License 2.0
5 stars 5 forks source link

Feature Extractor for surface overlap with source text #52

Open andreahorbach opened 5 years ago

andreahorbach commented 5 years ago

For source-based essays: One factor for scoring is how much material from the source text is verbatim lifted. We can model that e.g. via n-gram overlap with external text.