SNU IDS at SemEval-2019 Task 3: Addressing Training-Test Class Distribution Mismatch in Conversational Classification.
Sanghwan BaeJihun ChoiSang-goo LeePublished in: SemEval@NAACL-HLT (2019)
Keyphrases
- class distribution
- training set
- training samples
- class imbalance
- test set
- roc analysis
- cost sensitive
- highly skewed
- supervised learning
- test data
- class labels
- training examples
- classification accuracy
- imbalanced datasets
- cost sensitive learning
- misclassification costs
- majority class
- imbalanced data sets
- classification algorithm
- support vector
- training process
- training data
- decision boundary
- training set size
- imbalanced data
- decision trees
- support vector machine
- active learning
- feature extraction
- text classification
- image classification
- feature space
- highly imbalanced
- classification models
- svm classifier
- benchmark datasets
- unsupervised learning
- training dataset
- feature set
- nearest neighbor
- bayesian networks
- machine learning