Semantic piecewise convolutional neural network with adaptive negative training for distantly supervised relation extraction

Distantly Supervised Relation Extraction (DSRE) aims to automatically identify semantic relationships within large text corpora by aligning with external knowledge bases. Despite the success of current methods in automating data annotation, they introduce two main challenges: label noise and data long-tail distribution. Label noise results in inaccurate annotations, which can undermine the quality of relation extraction. The long-tail problem, on the other hand, leads to an imbalanced model that struggles to extract less frequent, long-tail relations. In this paper, we introduce a novel relation extraction framework based on multi-level hierarchical attention. This approach utilizes Graph Attention Networks (GATs) to model the hierarchical structure of the relations, capturing the semantic dependencies between relation types and generating relation embeddings that reflect the overall hierarchical framework. To improve the classification process, we incorporate a multi-level classification structure guided by hierarchical attention, which enhances the accuracy of both head and tail relation extraction. A local probability constraint is introduced to ensure coherence across the classification levels, fostering knowledge transfer from frequent to less frequent relations. Experimental evaluations on the New York Times (NYT) dataset demonstrate that our method outperforms existing baselines, particularly in the context of long-tail relation extraction, offering a comprehensive solution to the challenges of DSRE.

Distant supervised relation extraction has been widely used to find novel relational facts from text. However, distant supervision inevitably accompanies with the wrong labelling problem, and these noisy data will substantially hurt the performance of relation extraction. To alleviate this issue, we propose a sentence-level attention-based model for relation extraction. In this model, we employ convolutional neural networks to embed the semantics of sentences. Afterwards, we build sentence-level attention over multiple instances, which is expected to dynamically reduce the weights of those noisy instances. Experimental results on real-world datasets show that, our model can make full use of all informative sentences and effectively reduce the influence of wrong labelled instances. Our model achieves significant and consistent improvements on relation extraction as compared with baselines.

Two problems arise when using distant supervision for relation extraction. First, in this method, an already existing knowledge base is heuristically aligned to texts, and the alignment results are treated as labeled data. However, the heuristic alignment can fail, resulting in wrong label problem. In addition, in previous approaches, statistical models have typically been applied to ad hoc features. The noise that originates from the feature extraction process can cause poor performance. In this paper, we propose a novel model dubbed the Piecewise Convolutional Neural Networks (PCNNs) with multi-instance learning to address these two problems. To solve the first problem, distant supervised relation extraction is treated as a multi-instance problem in which the uncertainty of instance labels is taken into account. To address the latter problem, we avoid feature engineering and instead adopt convolutional architecture with piecewise max pooling to automatically learn relevant features. Experiments show that our method is effective and outperforms several competitive baseline methods.

The Graph Convolutional Network (GCN) is a universal relation extraction method that can predict relations of entity pairs by capturing sentences’ syntactic features. However, existing GCN methods often use dependency parsing to generate graph matrices and learn syntactic features. The quality of the dependency parsing will directly affect the accuracy of the graph matrix and change the whole GCN’s performance. Because of the influence of noisy words and sentence length in the distant supervised dataset, using dependency parsing on sentences causes errors and leads to unreliable information. Therefore, it is difficult to obtain credible graph matrices and relational features for some special sentences. In this article, we present a Multi-Graph Cooperative Learning model (MGCL), which focuses on extracting the reliable syntactic features of relations by different graphs and harnessing them to improve the representations of sentences. We conduct experiments on a widely used real-world dataset, and the experimental results show that our model achieves the state-of-the-art performance of relation extraction.

Semantic information and syntactic information are two essential dimensions of language. In recent years, most researches focus on extracting semantic information or how to extract useful information from syntactic dependency tree. As the semantic representations are closely related to syntactic ones, how to effectively integrate semantic and syntactic information is still a subject worth studying. We propose attention mechanism based on graph convolutional networks. The model integrates syntactic dependence information into semantic features and automatically achieve a balance between semantic information and syntactic information. We conducted experiments on the task of sentence-level relation extraction, and the results show that our model performs better than some previous methods.

Distant supervision for relation extraction with hierarchical selective attention

Attention mechanisms are often used in deep neural networks for distantly supervised relation extraction (DS-RE) to distinguish valid from noisy instances. However, traditional 1- D vector attention models are insufficient for the learning of different contexts in the selection of valid instances to predict the relationship for an entity pair. To alleviate this issue, we propose a novel multi-level structured (2-D matrix) self-attention mechanism for DS-RE in a multi-instance learning (MIL) framework using bidirectional recurrent neural networks. In the proposed method, a structured word-level self-attention mechanism learns a 2-D matrix where each row vector represents a weight distribution for different aspects of an instance regarding two entities. Targeting the MIL issue, the structured sentence-level attention learns a 2-D matrix where each row vector represents a weight distribution on selection of different valid instances. Experiments conducted on two publicly available DS-RE datasets show that the proposed framework with a multi-level structured self-attention mechanism significantly outperform state-of-the-art baselines in terms of PR curves, P@N and F1 measures.

Distant supervision (DS) has been widely used for relation extraction (RE), which automatically generates large‐scale labeled data. However, there is a wrong labeling problem, which affects the performance of RE. Besides, the existing method suffers from the lack of useful semantic features for some positive training instances. To address the above problems, we propose a novel RE model with sentence selection and interaction representation for distantly supervised RE. First, we propose a pattern method based on the relation trigger words as a sentence selector to filter out noisy sentences to alleviate the wrong labeling problem. After clean instances are obtained, we propose the interaction representation using the word‐level attention mechanism‐based entity pairs to dynamically increase the weights of the words related to entity pairs, which can provide more useful semantic information for relation prediction. The proposed model outperforms the strongest baseline by 2.61 in F1‐score on a widely used dataset, which proves that our model performs significantly better than the state‐of‐the‐art RE systems.

In document-level relation extraction (DocRE), graph structure is generally used to encode relation information in the input document to classify the relation category between each entity pair, and has greatly advanced the DocRE task over the past several years. However, the learned graph representation universally models relation information between all entity pairs regardless of whether there are relationships between these entity pairs. Thus, those entity pairs without relationships disperse the attention of the encoder-classifier DocRE for ones with relationships, which may further hind the improvement of DocRE. To alleviate this issue, we propose a novel encoder-classifier-reconstructor model for DocRE. The reconstructor manages to reconstruct the ground-truth path dependencies from the graph representation, to ensure that the proposed DocRE model pays more attention to encode entity pairs with relationships in the training. Furthermore, the reconstructor is regarded as a relationship indicator to assist relation classification in the inference, which can further improve the performance of DocRE model. Experimental results on a large-scale DocRE dataset show that the proposed model can significantly improve the accuracy of relation extraction on a strong heterogeneous graph-based baseline. The code is publicly available at https://github.com/xwjim/DocRE-Rec.

In recent years, distantly-supervised relation extraction has achieved a certain success by using deep neural networks. Distant Supervision (DS) can automatically generate large-scale annotated data by aligning entity pairs from Knowledge Bases (KB) to sentences. However, these DS-generated datasets inevitably have wrong labels that result in incorrect evaluation scores during testing, which may mislead the researchers. To solve this problem, we build a new dataset NYTH, where we use the DS-generated data as training data and hire annotators to label test data. Compared with the previous datasets, NYT-H has a much larger test set and then we can perform more accurate and consistent evaluation. Finally, we present the experimental results of several widely used systems on NYT-H. The experimental results show that the ranking lists of the comparison systems on the DS-labelled test data and human-annotated test data are different. This indicates that our human-annotated data is necessary for evaluation of distantly-supervised relation extraction.

The semantic social network is a complex system composed of nodes, links, and documents. Traditional semantic social network community detection algorithms only analyze network data from a single view, and there is no effective representation of semantic features at diverse levels of granularity. This paper proposes a multi-view integration method for community detection in semantic social network. We develop a data feature matrix based on node similarity and extract semantic features from the views of word frequency, keyword, and topic, respectively. To maximize the mutual information of each view, we use the robustness of L21-norm and F-norm to construct an adaptive loss function. On this foundation, we construct an optimization expression to generate the unified graph matrix and output the community structure with multiple views. Experiments on real social networks and benchmark datasets reveal that in semantic information analysis, multi-view is considerably better than single-view, and the performance of multi-view community detection outperforms traditional methods and multi-view clustering algorithms.

The task of entity relation extraction discovers new relation facts and enables broader applications of knowledge graph. Distant supervision is widely adopted for relation extraction, which requires large amounts of texts containing entity pairs as training data. However, in some specific domains such as medical-related applications, entity pairs that have certain relations might not appear together, thus it is difficult to meet the requirement for distantly supervised relation extraction. In the light of this challenge, we propose a novel path-based model to discover new entity relation facts. Instead of finding texts for relation extraction, the proposed method extracts path-only information for entity pairs from the current knowledge graph. For each pair of entities, multiple paths can be extracted, and some of them are more useful for relation extraction than others. In order to capture this observation, we employ attention mechanism to assign different weights for different paths, which highlights the useful paths for entity relation extraction. To demonstrate the effectiveness of the proposed method, we conduct various experiments on a large-scale medical knowledge graph. Compared with the state-of-the-art relation extraction methods using the structure of knowledge graph, the proposed method significantly improves the accuracy of extracted relation facts and achieves the best performance.

Neural models for distantly supervised relation extraction (DS-RE) encode each sentence in an entity-pair bag separately. These are then aggregated for bag-level relation prediction. Since, at encoding time, these approaches do not allow information to flow from other sentences in the bag, we believe that they do not utilize the available bag data to the fullest. In response, we explore a simple baseline approach (PARE) in which all sentences of a bag are concatenated into a passage of sentences, and encoded jointly using BERT. The contextual embeddings of tokens are aggregated using attention with the candidate relation as query -- this summary of whole passage predicts the candidate relation. We find that our simple baseline solution outperforms existing state-of-the-art DS-RE models in both monolingual and multilingual DS-RE datasets.

The fusion of hyperspectral imagery (HSI) and light detection and ranging (LiDAR) data is widely used for land cover classification. However, due to different imaging mechanisms, HSI and LiDAR data always present significant image differences, and the dimensions and feature distributions of HSI and LiDAR are highly dissimilar. This makes it challenging to represent and correlate semantic information from multimodal data. Current methods for classifying pixel-by-pixel features, which rely on cascaded or attention-based fusion, cannot effectively use multimodal features. To achieve accurate classification results, extracting and fusing similar high-order semantic information and complementary discriminative information contained in multimodal data is vital. In this paper, we propose a cross-modal semantic enhancement network (CMSE) for multimodal semantic information mining and fusion. Our proposed CMSE framework extracts features from the image on multiple scales, capturing more representative local sparse features with different sizes of convolution kernels. To represent high-level semantic features related to land cover, we establish a Gaussian-weighted matrix and semantically transform the spatial and spectral features of distinct branches. Finally, we build a multi-level residual fusion module to incrementally fuse spectral features from HSI and elevation features from LiDAR. Additionally, we introduce a cross-modal semantically constrained loss to guide multimodal semantic feature alignment. We evaluate our approach on three multimodal remote sensing datasets, namely the Houston2013, Trento, and MUUFL datasets. The experimental results demonstrate that our proposed CMSE model achieves superior performance in terms of accuracy and robustness compared to other related deep networks.

How semantic representations are manifest over the brain remains a topic of active debate. A semantic representation may be determined by specific semantic features (e.g. sensorimotor information), or may abstract away from specific features and represent generalized semantic characteristics (general semantic representation). Here we tested whether nodes of the semantic system code for a general semantic representation and/or possess representational spaces linked to particular semantic features. In an fMRI study, eighteen participants performed a typicality judgment task with written words drawn from sixteen different categories. Multivariate pattern analysis (MVPA) and representational similarity analysis (RSA) were adopted to investigate the sensitivity of the brain regions to semantic content and the type of semantic representation coded (general or feature-based). We replicated previous findings of sensitivity to general semantic similarity in posterior middle/inferior temporal gyrus (pMTG/ITG) and precuneus (PC) and additionally observed general semantic representations in ventromedial prefrontal cortex (PFC). Finally, two brain regions of the semantic network were sensitive to semantic features: the left pMTG/ITG was sensitive to haptic perception and the left ventral temporal cortex (VTC) to size. This finding supports the involvement of both general semantic representation and feature-based representations in the brain’s semantic system.