Local word vectors guiding keyphrase extraction

Automated keyphrase extraction is a fundamental textual information processing task concerned with the selection of representative phrases from a document that summarize its content. This work presents a novel unsupervised method for keyphrase extraction, whose main innovation is the use of local word embeddings (in particular GloVe vectors), i.e., embeddings trained from the single document under consideration. We argue that such local representation of words and keyphrases are able to accurately capture their semantics in the context of the document they are part of, and therefore can help in improving keyphrase extraction quality. Empirical results offer evidence that indeed local representations lead to better keyphrase extraction results compared to both embeddings trained on very large third corpora or larger corpora consisting of several documents of the same scientific field and to other state-of-the-art unsupervised keyphrase extraction methods.     

Web document summarization by exploiting social context with matrix co-factorization

In the context of social media, users usually post relevant information corresponding to the contents of events mentioned in a Web document. This information posses two important values in that (i) it reflects the content of an event and (ii) it shares hidden topics with sentences in the main document. In this paper, we present a novel model to capture the nature of relationships between document sentences and post information (comments or tweets) in sharing hidden topics for summarization of Web documents by utilizing relevant post information. Unlike previous methods which are usually based on hand-crafted features, our approach ranks document sentences and user posts based on their importance to the topics. The sentence-user-post relation is formulated in a share topic matrix, which presents their mutual reinforcement support. Our proposed matrix co-factorization algorithm computes the score of each document sentence and user post and extracts the top ranked document sentences and comments (or tweets) as a summary. We apply the model to the task of summarization on three datasets in two languages, English and Vietnamese, of social context summarization and also on DUC 2004 (a standard corpus of the traditional summarization task). According to the experimental results, our model significantly outperforms the basic matrix factorization and achieves competitive ROUGE-scores with state-of-the-art methods.     

Automatic generic document summarization based on non-negative matrix factorization

In existing unsupervised methods, Latent Semantic Analysis (LSA) is used for sentence selection. However, the obtained results are less meaningful, because singular vectors are used as the bases for sentence selection from given documents, and singular vector components can have negative values. We propose a new unsupervised method using Non-negative Matrix Factorization (NMF) to select sentences for automatic generic document summarization. The proposed method uses non-negative constraints, which are more similar to the human cognition process. As a result, the method selects more meaningful sentences for generic document summarization than those selected using LSA.     

From statistical methods to deep learning,automatic keyphrase prediction: A survey

Keyphrase prediction aims to generate phrases (keyphrases) that highly summarizes a given document. Recently, researchers have conducted in-depth studies on this task from various perspectives. In this paper, we comprehensively summarize representative studies from the perspectives of dominant models, datasets and evaluation metrics. Our work analyzes up to 167 previous works, achieving greater coverage of this task than previous surveys. Particularly, we focus highly on deep learning-based keyphrase prediction, which attracts increasing attention of this task in recent years. Afterwards, we conduct several groups of experiments to carefully compare representative models. To the best of our knowledge, our work is the first attempt to compare these models using the identical commonly-used datasets and evaluation metric, facilitating in-depth analyses of their disadvantages and advantages. Finally, we discuss the possible research directions of this task in the future.     

EdgeSumm: Graph-based framework for automatic text summarization

Searching the Internet for a certain topic can become a daunting task because users cannot read and comprehend all the resulting texts. Automatic Text summarization (ATS) in this case is clearly beneficial because manual summarization is expensive and time-consuming. To enhance ATS for single documents, this paper proposes a novel extractive graph-based framework “EdgeSumm” that relies on four proposed algorithms. The first algorithm constructs a new text graph model representation from the input document. The second and third algorithms search the constructed text graph for sentences to be included in the candidate summary. When the resulting candidate summary still exceeds a user-required limit, the fourth algorithm is used to select the most important sentences. EdgeSumm combines a set of extractive ATS methods (namely graph-based, statistical-based, semantic-based, and centrality-based methods) to benefit from their advantages and overcome their individual drawbacks. EdgeSumm is general for any document genre (not limited to a specific domain) and unsupervised so it does not require any training data. The standard datasets DUC2001 and DUC2002 are used to evaluate EdgeSumm using the widely used automatic evaluation tool: Recall-Oriented Understudy for Gisting Evaluation (ROUGE). EdgeSumm gets the highest ROUGE scores on DUC2001. For DUC2002, the evaluation results show that the proposed framework outperforms the state-of-the-art ATS systems by achieving improvements of 1.2% and 4.7% over the highest scores in the literature for the metrics of ROUGE-1 and ROUGE-L respectively. In addition, EdgeSumm achieves very competitive results for the metrics of ROUGE-2 and ROUGE-SU4.     

An end-to-end joint model for evidence information extraction from court record document

Information extraction is one of the important tasks in the field of Natural Language Processing (NLP). Most of the existing methods focus on general texts and little attention is paid to information extraction in specialized domains such as legal texts. This paper explores the task of information extraction in the legal field, which aims to extract evidence information from court record documents (CRDs). In the general domain, entities and relations are mostly words and phrases, indicating that they do not span multiple sentences. In contrast, evidence information in CRDs may span multiple sentences, while existing models cannot handle this situation. To address this issue, we first add a classification task in addition to the extraction task. We then formulate the two tasks as a multi-task learning problem and present a novel end-to-end model to jointly address the two tasks. The joint model adopts a shared encoder followed by separate decoders for the two tasks. The experimental results on the dataset show the effectiveness of the proposed model, which can obtain 72.36% F1 score, outperforming previous methods and strong baselines by a large margin.     

Boundaries and edges rethinking: An end-to-end neural model for overlapping entity relation extraction

Overlapping entity relation extraction has received extensive research attention in recent years. However, existing methods suffer from the limitation of long-distance dependencies between entities, and fail to extract the relations when the overlapping situation is relatively complex. This issue limits the performance of the task. In this paper, we propose an end-to-end neural model for overlapping relation extraction by treating the task as a quintuple prediction problem. The proposed method first constructs the entity graphs by enumerating possible candidate spans, then models the relational graphs between entities via a graph attention model. Experimental results on five benchmark datasets show that the proposed model achieves the current best performance, outperforming previous methods and baseline systems by a large margin. Further analysis shows that our model can effectively capture the long-distance dependencies between entities in a long sentence.     

A semantic approach to extractive multi-document summarization: Applying sentence expansion for tuning of conceptual densities

Today, due to a vast amount of textual data, automated extractive text summarization is one of the most common and practical techniques for organizing information. Extractive summarization selects the most appropriate sentences from the text and provide a representative summary. The sentences, as individual textual units, usually are too short for major text processing techniques to provide appropriate performance. Hence, it seems vital to bridge the gap between short text units and conventional text processing methods.In this study, we propose a semantic method for implementing an extractive multi-document summarizer system by using a combination of statistical, machine learning based, and graph-based methods. It is a language-independent and unsupervised system. The proposed framework learns the semantic representation of words from a set of given documents via word2vec method. It expands each sentence through an innovative method with the most informative and the least redundant words related to the main topic of sentence. Sentence expansion implicitly performs word sense disambiguation and tunes the conceptual densities towards the central topic of each sentence. Then, it estimates the importance of sentences by using the graph representation of the documents. To identify the most important topics of the documents, we propose an inventive clustering approach. It autonomously determines the number of clusters and their initial centroids, and clusters sentences accordingly. The system selects the best sentences from appropriate clusters for the final summary with respect to information salience, minimum redundancy, and adequate coverage.A set of extensive experiments on DUC2002 and DUC2006 datasets was conducted for investigating the proposed scheme. Experimental results showed that the proposed sentence expansion algorithm and clustering approach could considerably enhance the performance of the summarization system. Also, comparative experiments demonstrated that the proposed framework outperforms most of the state-of-the-art summarizer systems and can impressively assist the task of extractive text summarization.     

Mutual-reinforcement document summarization using embedded graph based sentence clustering for storytelling

In this paper, a document summarization framework for storytelling is proposed to extract essential sentences from a document by exploiting the mutual effects between terms, sentences and clusters. There are three phrases in the framework: document modeling, sentence clustering and sentence ranking. The story document is modeled by a weighted graph with vertexes that represent sentences of the document. The sentences are clustered into different groups to find the latent topics in the story. To alleviate the influence of unrelated sentences in clustering, an embedding process is employed to optimize the document model. The sentences are then ranked according to the mutual effect between terms, sentence as well as clusters, and high-ranked sentences are selected to comprise the summarization of the document. The experimental results on the Document Understanding Conference (DUC) data sets demonstrate the effectiveness of the proposed method in document summarization. The results also show that the embedding process for sentence clustering render the system more robust with respect to different cluster numbers.     

A progressive sentence selection strategy for document summarization

Saliency and coverage are two of the most important issues in document summarization. In most summarization methods, the saliency issue is usually of top priority. Many studies are conducted to develop better sentence ranking methods to identify the salient sentences for summarization. It is also well acknowledged that sentence selection strategies are very important, which mainly aim at reducing the redundancy among the selected sentences to enable them to cover more concepts. In this paper, we propose a novel sentence selection strategy that follows a progressive way to select the summary sentences. We intend to ensure the coverage of the summary first by an intuitive idea, i.e., considering the uncovered concepts only when measuring the saliency of the sentences. Moreover, we consider the subsuming relationship between sentences to define a conditional saliency measure of the sentences instead of the general saliency measures used in most existing methods. Based on these ideas, a progressive sentence selection strategy is developed to discover the “novel and salient” sentences. Compared with traditional methods, the saliency and coverage issues are more integrated in the proposed method. Experimental studies conducted on the DUC data sets demonstrate the advantages of the progressive sentence selection strategy.     

An unsupervised aspect extraction strategy for monitoring real-time reviews stream

One of the most important opinion mining research directions falls in the extraction of polarities referring to specific entities (aspects) contained in the analyzed texts. The detection of such aspects may be very critical especially when documents come from unknown domains. Indeed, while in some contexts it is possible to train domain-specific models for improving the effectiveness of aspects extraction algorithms, in others the most suitable solution is to apply unsupervised techniques by making such algorithms domain-independent and more efficient in a real-time environment. Moreover, an emerging need is to exploit the results of aspect-based analysis for triggering actions based on these data. This led to the necessity of providing solutions supporting both an effective analysis of user-generated content and an efficient and intuitive way of visualizing collected data. In this work, we implemented an opinion monitoring service implementing (i) a set of unsupervised strategies for aspect-based opinion mining together with (ii) a monitoring tool supporting users in visualizing analyzed data. The aspect extraction strategies are based on the use of an open information extraction strategy. The effectiveness of the platform has been tested on benchmarks provided by the SemEval campaign and have been compared with the results obtained by domain-adapted techniques.     

Sentence modeling via multiple word embeddings and multi-level comparison for semantic textual similarity

Recently, using a pretrained word embedding to represent words achieves success in many natural language processing tasks. According to objective functions, different word embedding models capture different aspects of linguistic properties. However, the Semantic Textual Similarity task, which evaluates similarity/relation between two sentences, requires to take into account of these linguistic aspects. Therefore, this research aims to encode various characteristics from multiple sets of word embeddings into one embedding and then learn similarity/relation between sentences via this novel embedding. Representing each word by multiple word embeddings, the proposed MaxLSTM-CNN encoder generates a novel sentence embedding. We then learn the similarity/relation between our sentence embeddings via Multi-level comparison. Our method M-MaxLSTM-CNN consistently shows strong performances in several tasks (i.e., measure textual similarity, identify paraphrase, recognize textual entailment). Our model does not use hand-crafted features (e.g., alignment features, Ngram overlaps, dependency features) as well as does not require pre-trained word embeddings to have the same dimension.     

A knowledge-rich approach to identifying semantic relations between nominals

This paper describes a state-of-the-art supervised, knowledge-intensive approach to the automatic identification of semantic relations between nominals in English sentences. The system employs a combination of rich and varied sets of new and previously used lexical, syntactic, and semantic features extracted from various knowledge sources such as WordNet and additional annotated corpora. The system ranked first at the third most popular SemEval 2007 Task – Classification of Semantic Relations between Nominals and achieved an F-measure of 72.4% and an accuracy of 76.3%. We also show that some semantic relations are better suited for WordNet-based models than other relations. Additionally, we make a distinction between out-of-context (regular) examples and those that require sentence context for relation identification and show that contextual data are important for the performance of a noun–noun semantic parser. Finally, learning curves show that the task difficulty varies across relations and that our learned WordNet-based representation is highly accurate so the performance results suggest the upper bound on what this representation can do.     

Deep Learning-based Extraction of Algorithmic Metadata in Full-Text Scholarly Documents

The advancements of search engines for traditional text documents have enabled the effective retrieval of massive textual information in a resource-efficient manner. However, such conventional search methodologies often suffer from poor retrieval accuracy especially when documents exhibit unique properties that behoove specialized and deeper semantic extraction. Recently, AlgorithmSeer, a search engine for algorithms has been proposed, that extracts pseudo-codes and shallow textual metadata from scientific publications and treats them as traditional documents so that the conventional search engine methodology could be applied. However, such a system fails to facilitate user search queries that seek to identify algorithm-specific information, such as the datasets on which algorithms operate, the performance of algorithms, and runtime complexity, etc. In this paper, a set of enhancements to the previously proposed algorithm search engine are presented. Specifically, we propose a set of methods to automatically identify and extract algorithmic pseudo-codes and the sentences that convey related algorithmic metadata using a set of machine-learning techniques. In an experiment with over 93,000 text lines, we introduce 60 novel features, comprising content-based, font style based and structure-based feature groups, to extract algorithmic pseudo-codes. Our proposed pseudo-code extraction method achieves 93.32% F1-score, outperforming the state-of-the-art techniques by 28%. Additionally, we propose a method to extract algorithmic-related sentences using deep neural networks and achieve an accuracy of 78.5%, outperforming a Rule-based model and a support vector machine model by 28% and 16%, respectively.     

Candidate document retrieval for cross-lingual plagiarism detection using two-level proximity information

The rapid growth of documents in different languages, the increased accessibility of electronic documents, and the availability of translation tools have caused cross-lingual plagiarism detection research area to receive increasing attention in recent years. The task of cross-language plagiarism detection entails two main steps: candidate retrieval and assessing pairwise document similarity. In this paper we examine candidate retrieval, where the goal is to find potential source documents of a suspicious text. Our proposed method for cross-language plagiarism detection is a keyword-focused approach. Since plagiarism usually happens in parts of the text, there is a requirement to segment the texts into fragments to detect local similarity. Therefore we propose a topic-based segmentation algorithm to convert the suspicious document to a set of related passages. After that, we use a proximity-based model to retrieve documents with the best matching passages. Experiments show promising results for this important phase of cross-language plagiarism detection.     

An extract-then-abstract based method to generate disaster-news headlines using a DNN extractor followed by a transformer abstractor

Generating news headlines has been one of the predominant problems in Natural Language Processing research. Modern transformer models, if fine-tuned, can present a good headline with attention to all the parts of a disaster-news article. A disaster-news headline generally focuses on the event, its effect, and the major impacts, which a transformer model lacks when generating the headline. The extract-then-abstract based method proposed in this article improves the performance of a state-of-the-art transformer abstractor to generate a good-quality disaster-news headline. In this work, a Deep Neural Network (DNN) based sentence extractor and a transformer-based abstractive summarizer work sequentially to generate a headline. The sentence extraction task is formulated as a binary classification problem where the DNN model is trained to generate two binary labels: one corresponding to the sentence similarity with ground truth headlines and the other corresponding to the presence of disaster and its impact related information in the sentence. The transformer model generates the headline from the sentences extracted by the DNN. ROUGE scores of the headlines generated using the proposed method are found to be better than the scores of the headlines generated directly from the original documents. The highest ROUGE 1, 2, and 3 score improvements are observed in the case of the Text-To-Text Transfer Transformer (T5) model by 17.85%, 38.13%, and 21.01%, respectively. Such improvements suggest that the proposed method can have a high utility for finding effective headlines from disaster related news articles.     

Query-focused multi-document summarization using hypergraph-based ranking

General graph random walk has been successfully applied in multi-document summarization, but it has some limitations to process documents by this way. In this paper, we propose a novel hypergraph based vertex-reinforced random walk framework for multi-document summarization. The framework first exploits the Hierarchical Dirichlet Process (HDP) topic model to learn a word-topic probability distribution in sentences. Then the hypergraph is used to capture both cluster relationship based on the word-topic probability distribution and pairwise similarity among sentences. Finally, a time-variant random walk algorithm for hypergraphs is developed to rank sentences which ensures sentence diversity by vertex-reinforcement in summaries. Experimental results on the public available dataset demonstrate the effectiveness of our framework.     

A cognitive emotion model enhanced sequential method for social emotion cause identification

Social emotion refers to the emotion evoked to the reader by a textual document. In contrast to the emotion cause extraction task which analyzes the cause of the author's sentiments based on the expressions in text, identifying the causes of social emotion evoked to the reader from text has not been explored previously. Social emotion mining and its cause analysis is not only an important research topic in Web-based social media analytics and text mining but also has a number of applications in multiple domains. As the focus of social emotion cause identification is on analyzing the causes of the reader's emotions elicited by a text that are not explicitly or implicitly expressed, it is a challenging task fundamentally different from the previous research. To tackle this, it also needs a deeper level understanding of the cognitive process underlying the inference of social emotion and its cause analysis. In this paper, we propose the new task of social emotion cause identification (SECI). Inspired by the cognitive structure of emotions (OCC) theory, we present a Cognitive Emotion model Enhanced Sequential (CogEES) method for SECI. Specifically, based on the implications of the OCC model, our method first establishes the correspondence between words/phrases in text and emotional dimensions identified in OCC and builds the emotional dimension lexicons with 1,676 distinct words/phrases. Then, our method utilizes lexicons information and discourse coherence for the semantic segmentation of document and the enhancement of clause representation learning. Finally, our method combines text segmentation and clause representation into a sequential model for cause clause prediction. We construct the SECI dataset for this new task and conduct experiments to evaluate CogEES. Our method outperforms the baselines and achieves over 10% F1 improvement on average, with better interpretability of the prediction results.     

Multilingual sentence categorization and novelty mining

A challenge for sentence categorization and novelty mining is to detect not only when text is relevant to the user’s information need, but also when it contains something new which the user has not seen before. It involves two tasks that need to be solved. The first is identifying relevant sentences (categorization) and the second is identifying new information from those relevant sentences (novelty mining). Many previous studies of relevant sentence retrieval and novelty mining have been conducted on the English language, but few papers have addressed the problem of multilingual sentence categorization and novelty mining. This is an important issue in global business environments, where mining knowledge from text in a single language is not sufficient. In this paper, we perform the first task by categorizing Malay and Chinese sentences, then comparing their performances with that of English. Thereafter, we conduct novelty mining to identify the sentences with new information. Experimental results on TREC 2004 Novelty Track data show similar categorization performance on Malay and English sentences, which greatly outperform Chinese. In the second task, it is observed that we can achieve similar novelty mining results for all three languages, which indicates that our algorithm is suitable for novelty mining of multilingual sentences. In addition, after benchmarking our results with novelty mining without categorization, it is learnt that categorization is necessary for the successful performance of novelty mining.