A Case Study on Pros and Cons of Regular Expression Detection and Dependency Parsing for Negation Extraction from German Medical Documents. Technical Report

TL;DR

This paper compares regular expression-based and dependency parsing methods for negation detection in German medical documents, highlighting their effectiveness, limitations, and potential for improving information extraction.

Contribution

It evaluates the efficiency of a reduced trigger set for negation detection and assesses the viability of dependency parsing as an alternative approach.

Findings

Smaller trigger sets can achieve similar negation detection performance.

Dependency parsing has potential but also notable shortcomings.

Regular expressions remain a practical baseline for negation detection.

Abstract

We describe our work on information extraction in medical documents written in German, especially detecting negations using an architecture based on the UIMA pipeline. Based on our previous work on software modules to cover medical concepts like diagnoses, examinations, etc. we employ a version of the NegEx regular expression algorithm with a large set of triggers as a baseline. We show how a significantly smaller trigger set is sufficient to achieve similar results, in order to reduce adaptation times to new text types. We elaborate on the question whether dependency parsing (based on the Stanford CoreNLP model) is a good alternative and describe the potentials and shortcomings of both approaches.