SERRANT: a syntactic classifier for English Grammatical Error Types.
Leshem ChoshenMatanel OrenDmitry NikolaevOmri AbendPublished in: CoRR (2021)
Keyphrases
- natural language
- parse tree
- error rate
- decision trees
- syntactic structures
- support vector machine
- feature selection
- feature space
- semantic representation
- minimum error
- training samples
- training data
- class labels
- multi class
- semantic analysis
- syntactic semantic
- word order
- semantic role labeling
- maximum entropy classifier
- classification algorithm
- context free grammars
- false negative
- part of speech
- cross language
- data sets
- svm classifier
- structured data
- learning algorithm