Keep Up With Top Trending Papers In CS / AI / ML ...
Computation and Language
A Multi-Instance Learning Reformulation for Biomedical Relation Extraction
Abstractified Multi-instance Learning (AMIL) for Biomedical Relation Extraction
We propose a novel reformulation of multi-instance learning (mil) for biomedical relation extraction that abstractifies biomedical entities into their corresponding semantic types.
By grouping entities by types, we are better able to take advantage of the benefits of mil and further denoise the training signal.
We show this reformulation, which we refer to as abstractified multi-instance learning (amil), improves performance in biomedical relationship extraction.
We also propose a novel relationshipembedding architecture that further improves model performance.
William Hogan, Molly Huang, Yannis Katsis, Tyler Baldwin, Ho-Cheol Kim, Yoshiki Vazquez Baeza, Andrew Bartko, Chun-Nan Hsu
Abstractified multi-instance learning
Relationship embedding architecture
Read the Paper
◐ Recommended Members
◐ Latest Activity