TUDA-CCL at SemEval-2021 Task 1: Using Gradient-boosted Regression Tree Ensembles Trained on a Heterogeneous Feature Set for Predicting Lexical Complexity.
Sebastian GombertSabine BartschPublished in: SemEval@ACL/IJCNLP (2021)
Keyphrases
- feature set
- tree ensembles
- random forests
- random forest
- word sense disambiguation
- syntactic features
- feature selection
- classification accuracy
- feature space
- feature vectors
- conformal prediction
- feature extraction
- selected features
- part of speech
- wordnet
- class labels
- decision trees
- learning algorithm
- semantic role labeling
- semantic features
- ensemble methods
- support vector
- kernel function
- feature selection algorithms
- machine translation
- regression model
- prediction accuracy
- image processing
- machine learning algorithms