Identifying Diabetes in Clinical Notes in Hebrew: A Novel Text Classification Approach Based on Word Embedding.
Maxim TopazLudmila MurgaChagai GrossmanDaniella DaliyotShlomit JacobsonNoa RozendornEyal ZimlichmanNadav FuriePublished in: MedInfo (2019)
Keyphrases
- text classification
- n gram
- term frequency
- training corpus
- bag of words
- text categorization
- text mining
- machine learning
- word segmentation
- text documents
- distributional clustering
- feature selection
- text data
- text compression
- computational linguistics
- naive bayes
- word sense disambiguation
- co occurrence
- text classifiers
- sentiment analysis
- language modeling
- health care
- semantic features
- document classification
- labeled data
- multi label
- diabetic patients
- diabetes mellitus
- hidden markov models
- training data
- knowledge discovery
- data cleaning
- high risk
- sentiment classification
- knn
- vector space
- semi supervised learning