Gated Neural Network for Sentence Compression Using Linguistic Knowledge.

Previous works have recognized that linguistic features such as part of speech and dependency labels are helpful for sentence compression that aims to simplify a text while leaving its underlying meaning. In this work, we introduce a gating mechanism and propose a gated neural network that selectively exploits linguistic knowledge for deletion-based sentence compression. Experimental results on two popular datasets show that the proposed gated neural network equipped with selectively fused linguistic features leads to better compressions upon both automatic metric and human evaluation, compared with a previous competitive compression system. We also investigate the gating mechanism through visualization analysis.