Bachelor Thesis: Automatic Extraction of Semantic Legal Metadata Using Natural Language Processing

The goal of this thesis is to extract semantic metadata from legal texts via Natural Language Processing (NLP). Tasks include (1) building a natural-language-processing-pipeline consisting of Tokenizer, Sentence Splitter, Part-of-Speech Tagger, Named-entity Recognizer, Constituency Parser and Dependency Parser and finally (2) building rules for annotating constituents using Tregex.

Supervisor: Michael Felderer