Skip to content

Recurrent neural network with connectionist temporal classification for learning how to form abbreviations.

Notifications You must be signed in to change notification settings

gpfinley/abbreviation-ctc

Repository files navigation

abbreviation-ctc

An attempt to model how humans construct abbreviations using bidirectional long short-term memory networks and connectionist temporal classification. Currently set up to learn biomedical abbreviations (train on UMLS and test on non-UMLS abbreviations from ADAM).

Uses Python and TensorFlow. Code is still very much in the experimental stage! It's ugly, but it kind of works. A trained network (couple of days, on GPUs) achieves better performance than a rule-based initialism baseline. Lots more to do with tweaking the network and looking at different data sets.

About

Recurrent neural network with connectionist temporal classification for learning how to form abbreviations.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages