ReliAble dependency arc recognition

•We propose a novel natural language processing task, ReliAble Dependency Arc Recognition (RADAR).•We model RADAR as a binary classification problem with imbalanced data.•We design three sorts of features to express reliability of arcs and evaluated the contributions of these features.•A logistic regression classifier is trained to recognize reliable dependency arcs.•The classification method can outperform a probabilistic baseline method.

We propose a novel natural language processing task, ReliAble dependency arc recognition (RADAR), which helps high-level applications better utilize the dependency parse trees. We model RADAR as a binary classification problem with imbalanced data, which classifies each dependency parsing arc as correct or incorrect. A logistic regression classifier with appropriate features is trained to recognize reliable dependency arcs (correct with high precision). Experimental results show that the classification method can outperform a probabilistic baseline method, which is calculated by the original graph-based dependency parser.