Search

The semantic representation of words is a fundamental task in natural language processing and text mining. Learning word embedding has shown its power on various tasks. Most studies are aimed at generating embedding representation of a word based on encoding its context information. However, many latent relations, such as co-occurring associative patterns and semantic conceptual relations, are not well considered. In this paper, we propose an extensible model to incorporate these kinds of valuable latent relations to increase the semantic relatedness of word pairs by learning word embeddings. To assess the effectiveness of our model, we conduct experiments on both information retrieval and text classification tasks. The results indicate the effectiveness of our model as well as its flexibility on different tasks.

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. [ABSTRACT FROM AUTHOR]

Top-k words selection is a technique used to detect and return the k most similar words to a given word from a candidate set. This is a crucial and widely used tool in various tasks. The key issue in top-k words selection is how to measure the similarity between words. One popular and effective solution is to use a word embedding-based similarity measure, which represents words as low-dimensional vectors and measures the similarities between words according to the similarity of the vectors, using a metric. However, most word embedding methods only consider the local proximity properties of two words in a corpus. To mitigate this issue. In this article, we propose to use association rules for measuring word similarity at a global level, and a fuzzy similarity measure for top-k words selection that jointly encodes the local and the global similarities. Experiments on a real-world query task with three benchmark datasets, i.e., TREC-disk 4&5, WT10G, and RCV1, demonstrate the efficiency of the proposed method compared to several state-of-the-art baselines. [ABSTRACT FROM AUTHOR]

• We present a hypergraph network model for nested entity mention recognition. • We recognize nested entities by tagging hyperedges instead of nodes. • We propose a theorem to make hyperedges be easily denoted in the program. • We solve the data imbalance problem and reduce the computing cost. We propose a hypergraph network (HGN) model to recognize the nested entity mentions in texts. This model can learn the representations for the sequence structures of natural languages and the representations for the hypergraph structures of nested entity mentions. Mainstream methods recognize an entity mention by separately tagging the words or the gaps between words, which may complicate the problem and not be favorable for capturing the overall features of the mention. To solve these issues, the HGN model treats each entity mention as a whole and tags it with one label. We represent each sentence as a hypergraph, in which nodes represent words and hyperedges represent entity mentions. Thus, entity mention recognition (EMR) is transformed into a task of classifying the hyperedges. The HGN model firstly uses encoders to extract the features and learn a hypergraph representation, and then recognizes entity mentions by tagging every hyperedge. The experiments on three standard datasets demonstrate our model outperforms the previous models for nested EMR. We openly release the source code at https://github.com/nlplab-ie/HGN. [ABSTRACT FROM AUTHOR]

Linguistic knowledge is useful for various NLP tasks, but the difficulty lies in the representation and application. We consider that linguistic knowledge is implied in a large-scale corpus, while classification knowledge, the knowledge related to the definitions of entity and relation types, is implied in the labeled training data. Therefore, a corpus subgraph is proposed to mine more linguistic knowledge from the easily accessible unlabeled data, and sentence subgraphs are used to acquire classification knowledge. They jointly constitute a relation knowledge graph (RKG) to extract relations from sentences in this paper. On RKG, entity recognition can be regarded as a property value filling problem and relation classification can be regarded as a link prediction problem. Thus, the multiple relation extraction can be treated as a reasoning process for knowledge completion. We combine statistical reasoning and neural network reasoning to segment sentences into entity chunks and non-entity chunks, then propose a novel Chunk Graph LSTM network to learn the representations of entity chunks and infer the relations among them. The experiments on two standard datasets demonstrate our model outperforms the previous models for multiple relation extraction. [ABSTRACT FROM AUTHOR]

Generating the precise semantic representation of a word or concept is a fundamental task in natural language processing. Recent studies which incorporate semantic knowledge into word embedding have shown their potential in improving the semantic representation of a concept. However, existing approaches only achieved limited performance improvement as they usually 1) model a word's semantics from some explicit aspects while ignoring the intrinsic aspects of the word, 2) treat semantic knowledge as a supplement of word embeddings, and 3) consider partial relations between concepts while ignoring rich coupling relations between them, such as explicit concept co-occurrences in descriptive texts in a corpus as well as concept hyperlink relations in a knowledge network, and implicit couplings between concept co-occurrences and hyperlinks. In human consciousness, a concept is always associated with various couplings that exist within/between descriptive texts and knowledge networks, which inspires us to capture as many concept couplings as possible for building a more informative concept representation. We thus propose a neural coupled concept representation (CoupledCR) framework and its instantiation: a coupled concept embedding (CCE) model. CCE first learns two types of explicit couplings that are based on concept co-occurrences and hyperlink relations, respectively, and then learns a type of high-level implicit couplings between these two types of explicit couplings for better concept representation. Extensive experimental results on six real-world datasets show that CCE significantly outperforms eight state-of-the-art word embeddings and semantic representation methods. [ABSTRACT FROM AUTHOR]

Sentence auto-completion is an important feature that saves users many keystrokes in typing the entire sentence by providing suggestions as they type. Despite its value, the existing sentence auto-completion methods, such as query completion models, can hardly be applied to solving the object completion problem in sentences with the form of (subject, verb, object), due to the complex natural language description and the data deficiency problem. Towards this goal, we treat an SVO sentence as a three-element triple (subject, sentence pattern, object), and cast the sentence object completion problem as an element inference problem. These elements in all triples are encoded into a unified low-dimensional embedding space by our proposed TRANSFER model, which leverages the external knowledge base to strengthen the representation learning performance. With such representations, we can provide reliable candidates for the desired missing element by a linear model. Extensive experiments on a real-world dataset have well-validated our model. Meanwhile, we have successfully applied our proposed model to factoid question answering systems for answer candidate selection, which further demonstrates the applicability of the TRANSFER model. [ABSTRACT FROM PUBLISHER]

Joint entity and relation extraction is an important task in natural language processing and knowledge graph construction. Existing studies mainly focus on three issues: redundant predictions, overlapping triples and relation connections. However, as far as we know, none of them is able to solve the three problems simultaneously in a unified architecture. To address this issue, in this paper, we propose a novel translation based unified framework. Specifically, the proposed framework contains two components: an entity tagger and a relation extractor. The former is used to recognize all candidate head entities and tail entities respectively. The latter predicts relations for every entity pair dynamically through ranking with translation mechanism. To show the superiority of the proposed framework, we instantiate it through the simplest binary entity tagger and TransE algorithm. Extensive experiments over two widely used datasets demonstrate that, even with the simplest components, the proposed framework can still achieve competitive performance with most previous baselines. Moreover, the framework is flexible. It enjoys further performance boost when employing more powerful entity tagger and knowledge graph embedding algorithm. [ABSTRACT FROM AUTHOR]