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.