Transfer Learning for Human & AI

We describe our work on information extraction in medical documents written in German . We employ a version of the NegExregular expression algorithm with a large set of triggers as a baseline . Weshow how a significantly smaller trigger set is sufficient to achieve similar results . We elaborate on the question whether dependency parsing (based on the Stanford CoreNLP model) is a good alternative . We describe the potentials and shortcomings of both approaches and describe potentials of bothapproaches. Weshow we show how smaller trigger sets are sufficient to reduce adaptation times to new text types .

Author(s) : Hans-Jürgen Profitlich, Daniel Sonntag

Links : PDF - Abstract

Code :


Keywords : describe - dependency - expression - parsing - sufficient -

• Bioinformatics Specialization on Coursera
• The Hundred-Page Machine Learning Book
• Kaggle Competitions
• Deeplearning.ai Specialization by Andrew Ng
• Practical Deep Learning for Coders – Part 1
• Computational Linear Algebra by Rachel Thomas
• 3Blue1Brown
• How to Win a Data Science Competition on Coursera