Context-Sensitive Spelling Correction of Clinical Text via Conditional Independence

Juyong Kim, Jeremy C Weiss, Pradeep Ravikumar
Proceedings of the Conference on Health, Inference, and Learning, PMLR 174:234-247, 2022.

Abstract

Spelling correction is a particularly important problem in clinical natural language processing because of the abundant occurrence of misspellings in medical records. However, the scarcity of labeled datasets in a clinical context makes it hard to build a machine learning system for such clinical spelling correction. In this work, we present a probabilistic model of correcting misspellings based on a simple conditional independence assumption, which leads to a modular decomposition into a language model and a corruption model. With a deep character-level language model trained on a large clinical corpus, and a simple edit-based corruption model, we can build a spelling correction model with small or no real data. Experimental results show that our model significantly outperforms baselines on two healthcare spelling correction datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v174-kim22b, title = {Context-Sensitive Spelling Correction of Clinical Text via Conditional Independence}, author = {Kim, Juyong and Weiss, Jeremy C and Ravikumar, Pradeep}, booktitle = {Proceedings of the Conference on Health, Inference, and Learning}, pages = {234--247}, year = {2022}, editor = {Flores, Gerardo and Chen, George H and Pollard, Tom and Ho, Joyce C and Naumann, Tristan}, volume = {174}, series = {Proceedings of Machine Learning Research}, month = {07--08 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v174/kim22b/kim22b.pdf}, url = {https://proceedings.mlr.press/v174/kim22b.html}, abstract = {Spelling correction is a particularly important problem in clinical natural language processing because of the abundant occurrence of misspellings in medical records. However, the scarcity of labeled datasets in a clinical context makes it hard to build a machine learning system for such clinical spelling correction. In this work, we present a probabilistic model of correcting misspellings based on a simple conditional independence assumption, which leads to a modular decomposition into a language model and a corruption model. With a deep character-level language model trained on a large clinical corpus, and a simple edit-based corruption model, we can build a spelling correction model with small or no real data. Experimental results show that our model significantly outperforms baselines on two healthcare spelling correction datasets.} }
Endnote
%0 Conference Paper %T Context-Sensitive Spelling Correction of Clinical Text via Conditional Independence %A Juyong Kim %A Jeremy C Weiss %A Pradeep Ravikumar %B Proceedings of the Conference on Health, Inference, and Learning %C Proceedings of Machine Learning Research %D 2022 %E Gerardo Flores %E George H Chen %E Tom Pollard %E Joyce C Ho %E Tristan Naumann %F pmlr-v174-kim22b %I PMLR %P 234--247 %U https://proceedings.mlr.press/v174/kim22b.html %V 174 %X Spelling correction is a particularly important problem in clinical natural language processing because of the abundant occurrence of misspellings in medical records. However, the scarcity of labeled datasets in a clinical context makes it hard to build a machine learning system for such clinical spelling correction. In this work, we present a probabilistic model of correcting misspellings based on a simple conditional independence assumption, which leads to a modular decomposition into a language model and a corruption model. With a deep character-level language model trained on a large clinical corpus, and a simple edit-based corruption model, we can build a spelling correction model with small or no real data. Experimental results show that our model significantly outperforms baselines on two healthcare spelling correction datasets.
APA
Kim, J., Weiss, J.C. & Ravikumar, P.. (2022). Context-Sensitive Spelling Correction of Clinical Text via Conditional Independence. Proceedings of the Conference on Health, Inference, and Learning, in Proceedings of Machine Learning Research 174:234-247 Available from https://proceedings.mlr.press/v174/kim22b.html.

Related Material