Seed Word Selection for Weakly-Supervised Text Classification with Unsupervised Error Estimation.
Yiping JinAkshay BhatiaDittaya WanvariePublished in: NAACL-HLT (Student Research Workshop) (2021)
Keyphrases
- weakly supervised
- error estimation
- text classification
- n gram
- relation extraction
- fully supervised
- supervised methods
- superpixels
- model selection
- labeled data
- semantic features
- object class
- semi supervised
- text mining
- topic models
- feature selection
- text categorization
- bag of words
- named entities
- co occurrence
- unlabeled data
- generalization error
- object detection
- feature extraction
- machine learning
- domain specific
- multi label
- automatic extraction
- support vector machine
- feature space
- natural language
- image processing