Attention enhanced capsule network for text classification by encoding syntactic dependency trees with graph convolutional neural network

Text classification is a fundamental task in many applications such as topic labeling, sentiment analysis, and spam detection. The text syntactic relationship and word sequence are important and useful for text classification. How to model and incorporate them to improve performance is one key challenge. Inspired by human behavior in understanding text. In this paper, we combine the syntactic relationship, sequence structure, and semantics for text representation, and propose an attention-enhanced capsule network-based text classification model. Specifically, we use graph convolutional neural networks to encode syntactic dependency trees, build multi-head attention to encode dependencies relationship in text sequence, merge with semantic information by capsule network at last. Extensive experiments on five datasets demonstrate that our approach can effectively improve the performance of text classification compared with state-of-the-art methods. The result also shows capsule network, graph convolutional neural network, and multi-headed attention has integration effects on text classification tasks.

Augmenting Paraphrase Generation with Syntax Information Using Graph Convolutional Networks

Paraphrase generation is an important yet challenging task in natural language processing. Neural network-based approaches have achieved remarkable success in sequence-to-sequence learning. Previous paraphrase generation work generally ignores syntactic information regardless of its availability, with the assumption that neural nets could learn such linguistic knowledge implicitly. In this work, we make an endeavor to probe into the efficacy of explicit syntactic information for the task of paraphrase generation. Syntactic information can appear in the form of dependency trees, which could be easily acquired from off-the-shelf syntactic parsers. Such tree structures could be conveniently encoded via graph convolutional networks to obtain more meaningful sentence representations, which could improve generated paraphrases. Through extensive experiments on four paraphrase datasets with different sizes and genres, we demonstrate the utility of syntactic information in neural paraphrase generation under the framework of sequence-to-sequence modeling. Specifically, our graph convolutional network-enhanced models consistently outperform their syntax-agnostic counterparts using multiple evaluation metrics.

Do dependency lengths explain constraints on crossing dependencies?

In syntactic dependency trees, when arcs are drawn from syntactic heads to dependents, they rarely cross. Constraints on these crossing dependencies are critical for determining the syntactic properties of human language, because they define the position of natural language in formal language hierarchies. We study whether the apparent constraints on crossing syntactic dependencies in natural language might be explained by constraints on dependency lengths (the linear distance between heads and dependents). We compare real dependency trees from treebanks of 52 languages against baselines of random trees which are matched with the real trees in terms of their dependency lengths. We find that these baseline trees have many more crossing dependencies than real trees, indicating that a constraint on dependency lengths alone cannot explain the empirical rarity of crossing dependencies. However, we find evidence that a combined constraint on dependency length and the rate of crossing dependencies might be able to explain two of the most-studied formal restrictions on dependency trees: gap degree and well-nestedness.

Augmenting Paraphrase Generation with Syntax Information using Graph Convolutional Networks

Paraphrase generation is an important yet challenging task in NLP. Neural network-based approaches have achieved remarkable success in sequence-to-sequence(seq2seq) learning. Previous paraphrase generation work generally ignores syntactic information regardless of its availability, with the assumption that neural nets could learn such linguistic knowledge implicitly. In this work we make an endeavor to probe into the efficacy of explicit syntactic information for the task of paraphrase generation. Syntactic information can appear in the form of dependency trees which could be easily acquired from off-the-shelf syntactic parsers. Such tree structures could be conveniently encoded via graph convolutional networks(GCNs) to obtain more meaningful sentence representations, which could improve generated paraphrases. Through extensive experiments on four paraphrase datasets with different sizes and genres, we demonstrate the utility of syntactic information in neural paraphrase generation under the framework of seq2seq modeling. Specifically, our GCN-enhanced models consistently outperform their syntax-agnostic counterparts in multiple evaluation metrics.

Syntax-Informed Self-Attention Network for Span-Based Joint Entity and Relation Extraction

Current state-of-the-art joint entity and relation extraction framework is based on span-level entity classification and relation identification between pairs of entity mentions. However, while maintaining an efficient exhaustive search on spans, the importance of syntactic features is not taken into consideration. It will lead to a problem that the prediction of a relation between two entities is related based on corresponding entity types, but in fact they are not related in the sentence. In addition, although previous works have proven that extract local context is beneficial for the task, it still lacks in-depth learning of contextual features in local context. In this paper, we propose to incorporate syntax knowledge into multi-head self-attention by employing part of heads to focus on syntactic parents of each token from pruned dependency trees, and we use it to model the global context to fuse syntactic and semantic features. In addition, in order to get richer contextual features from the local context, we apply local focus mechanism on entity pairs and corresponding context. Based on applying the two strategies, we perform joint entity and relation extraction on span-level. Experimental results show that our model achieves significant improvements on both Conll04 and SciERC dataset compared to strong competitors.