Search

Document-level Relation Extraction(DocRE) aims to extract relations between entities from documents. In contrast to sentence-level relation extraction, it requires extracting semantic relations from multiple sentences. It is necessary to further improve the performance of the above algorithm in order to extract document-level relation. Therefore, the DocRE algorithms have to deal with more complex entity structure relationships and the need to unite semantic relationships between different sentences when reasoning about relationships between entities. The proposed algorithms fail to infer relationships between entities when dealing with complex entity structure relationships. In this paper, we propose an entity mentions deep attention framework that efficiently infers entity relationships through entity structure and contextual information. Firstly, a structural dependency module of entities is designed to achieve interaction between different mentions of the entity. Secondly, a deep contextual attention component proposed to enrich the semantic information between entities by entity-related contexts. Finally, we use a distance mapping component to solve the problem of entity pairs that are far away from each other. According to our implementation results, our model outperforms the state-ofthe-art models on three public datasets DocRED, DGA, and CDR. • We created dependencies for entities and various mentions to enable structural reasoning. • The deep contextual attention module is designed to enhance context semantic information and improve model reasoning. • A location mapping module enhances entity representations for accurate capture of remote entity relationships. • The experimental results demonstrate the superiority of our EMDA model. [ABSTRACT FROM AUTHOR]

With the development of pre-trained language models and large-scale datasets, automatic text summarization has attracted much attention from the community of natural language processing, but the progress of automatic summarization evaluation has stagnated. Although there have been efforts to improve automatic summarization evaluation, ROUGE has remained one of the most popular metrics for nearly 20 years due to its competitive evaluation performance. However, ROUGE is not perfect, there are studies have shown that it is suffering from inaccurate evaluation of abstractive summarization and limited diversity of generated summaries, both caused by lexical bias. To avoid the bias of lexical similarity, more and more meaningful embedding-based metrics have been proposed to evaluate summaries by measuring semantic similarity. Due to the challenge of accurately measuring semantic similarity, none of them can fully replace ROUGE as the default automatic evaluation toolkit for text summarization. To address the aforementioned problems, we propose a compromise evaluation framework (ROUGE-SEM) for improving ROUGE with semantic information, which compensates for the lack of semantic awareness through a semantic similarity module. According to the differences in semantic similarity and lexical similarity, summaries are classified into four categories for the first time, including good-summary, pearl-summary, glass-summary, and bad-summary. In particular, the back-translation technique is adopted to rewrite pearl-summary and glass-summary that are inaccurately evaluated by ROUGE to alleviate lexical bias. Through this pipeline framework, summaries are first classified by candidate summary classifier, then rewritten by categorized summary rewriter, and finally scored by rewritten summary scorer, which are efficiently evaluated in a manner consistent with human behavior. When measured using Pearson, Spearman, and Kendall rank coefficients, our proposal achieves comparable or higher correlations with human judgments than several state-of-the-art automatic summarization evaluation metrics in dimensions of coherence, consistency, fluency, and relevance. This also suggests that improving ROUGE with semantics is a promising direction for automatic summarization evaluation. [ABSTRACT FROM AUTHOR]

Background: Automated summarization of scientific literature and patient records is essential for enhancing clinical decision-making and facilitating precision medicine. Most existing summarization methods are based on single indicators of relevance, offer limited capabilities for information visualization, and do not account for user specific interests. In this work, we develop an interactive content extraction, recognition, and construction system (CERC) that combines machine learning and visualization techniques with domain knowledge for highlighting and extracting salient information from clinical and biomedical text.Methods: A novel sentence-ranking framework multi indicator text summarization, MINTS, is developed for extractive summarization. MINTS uses random forests and multiple indicators of importance for relevance evaluation and ranking of sentences. Indicative summarization is performed using weighted term frequency-inverse document frequency scores of over-represented domain-specific terms. A controlled vocabulary dictionary generated using MeSH, SNOMED-CT, and PubTator is used for determining relevant terms. 35 full-text CRAFT articles were used as the training set. The performance of the MINTS algorithm is evaluated on a test set consisting of the remaining 32 full-text CRAFT articles and 30 clinical case reports using the ROUGE toolkit.Results: The random forests model classified sentences as "good" or "bad" with 87.5% accuracy on the test set. Summarization results from the MINTS algorithm achieved higher ROUGE-1, ROUGE-2, and ROUGE-SU4 scores when compared to methods based on single indicators such as term frequency distribution, position, eigenvector centrality (LexRank), and random selection, p < 0.01. The automatic language translator and the customizable information extraction and pre-processing pipeline for EHR demonstrate that CERC can readily be incorporated within clinical decision support systems to improve quality of care and assist in data-driven and evidence-based informed decision making for direct patient care.Conclusions: We have developed a web-based summarization and visualization tool, CERC ( https://newton.isye.gatech.edu/CERC1/ ), for extracting salient information from clinical and biomedical text. The system ranks sentences by relevance and includes features that can facilitate early detection of medical risks in a clinical setting. The interactive interface allows users to filter content and edit/save summaries. The evaluation results on two test corpuses show that the newly developed MINTS algorithm outperforms methods based on single characteristics of importance. [ABSTRACT FROM AUTHOR]

Automatic text summarization is important in this era due to the exponential growth of documents available on the Internet. In the Vietnamese language, VietnameseMDS is the only publicly available dataset for this task. Although the dataset has 199 clusters, there are only three documents in each cluster, which is small compared to typical datasets in English. This motivates us to construct ViMs—a big and high-quality Vietnamese dataset for abstractive multi-document summarization. To that end, we recruited 29 annotators and enhanced MDSWriter—an open-source annotation tool, to support the annotators in creating gold standard summaries. As a result, ViMs has 600 summaries corresponding to 300 clusters of 1,945 documents. We have verified the reliability of our dataset by using a variety of metrics including conventional Cohen's κ , relaxed Cohen's κ —a new metric that we propose to make it more suitable for abstractive summarization, and ROUGE scores. A relaxed κ score of 0.55 indicate that ViMs could attain moderate agreement between annotators. Meanwhile, ROUGE scores are 0.729 of ROUGE-1, 0.507 of ROUGE-2 and 0.524 of ROUGE-SU4. We have further evaluated ViMs by using three different summarization systems: TextRank, CFVi and MUSEEC. Their performances are 0.628, 0.711 and 0.732 of ROUGE-1, respectively. These results show that the ViMs dataset is suitable for both training and evaluating multi-document summarization systems. We have made the dataset and evaluation results of this work publicly available for research community. It is noted that unlike previous work that only published the final summarization dataset, we also publish intermediate annotation results, which can be used in other NLP problems such as sentence classification. [ABSTRACT FROM AUTHOR]

To inform the development of automated summarization of clinical conversations, this study sought to estimate the proportion of doctor-patient communication in general practice (GP) consultations used for generating a consultation summary. Two researchers with a medical degree read the transcripts of 44 GP consultations and highlighted the phrases to be used for generating a summary of the consultation. For all consultations, less than 20% of all words in the transcripts were needed for inclusion in the summary. On average, 9.1% of all words in the transcripts, 26.6% of all medical terms, and 27.3% of all speaker turns were highlighted. The results indicate that communication content used for generating a consultation summary makes up a small portion of GP consultations, and automated summarization solutions—such as digital scribes—must focus on identifying the 20% relevant information for automatically generating consultation summaries. [ABSTRACT FROM AUTHOR]

This paper presented a novel model for the lower accuracy issue in Chinese automatic summary which merged copying mechanism and input-feeding approach into the decoder of sequence-to-sequence ( seq2seq) basic model with attention mechanism. Firstly, it used pointer networks to extend the source's OOV words to a fixed dictionary to copy OOV words from the source into the generated sequence. Secondly, it used the input-feeding approach to track the attention decision information of generated sequence for improving the model output accuracy. Experimental results on NLPCC2018 datasets show that the proposed model obtains a higher ROUGE score than the basic model, which confirms the feasibility of this model. [ABSTRACT FROM AUTHOR]

In order to tackle the problem of abstractive summarization of long multi-sentence texts, it is critical to construct an efficient model, which can learn and represent multiple compositionalities better. In this paper, we introduce a temporal hierarchical pointer generator network that can represent multiple compositionalities in order to handle longer sequences of texts with a deep structure. We demonstrate how a multilayer gated recurrent neural network organizes itself with the help of an adaptive timescale in order to represent the compositions. The temporal hierarchical network is implemented with a multiple timescale architecture where the timescale of each layer is also learned during the training process through error backpropagation through time. We evaluate our proposed model using an Introduction-Abstract summarization dataset from scientific articles and the CNN/Daily Mail summarization benchmark dataset. The results illustrate that, we successfully implement a summary generation system for long texts by using the multiple timescale with adaptation concept. We also show that we have improved the summary generation system with our proposed model on the benchmark dataset. [ABSTRACT FROM AUTHOR]

This research leads to the development of a system that generates minutes while the system and user interactively structure and summarize the discussion. The system can look back on a meeting from a specific viewpoint or purpose by sending an inquiry to the minutes, receiving the result, and performing a new operation. To improve the reusability of the minutes, it is necessary to understand the semantic relationships among utterances at a deep level. We therefore propose a novel tree structure called the discussion time-span tree for representing a hierarchical discussion based on the relative importance of utterances derived from the meeting. The discussion time-span tree is a binary tree, each leaf of which corresponds to an utterance. Adjacent relevant utterances are hierarchically grouped, and the most important utterance becomes a head. In this way, the tree represents the implicit meaning and intention contained in meeting records, and can be used for summarization. Our system aims to grasp the main point of the meeting by interactively manipulating this tree structure and giving different viewpoints to each system user. An evaluation of the important utterance extraction using the proposed system showed that the score of ROUGE-2 was higher than that of the existing document summarizing technique. In a subsequent user evaluation conducted with ten participants, we found that the system can be used to grasp conference contents efficiently. The results of both evaluations demonstrate the usefulness of the system. [ABSTRACT FROM AUTHOR]

Text summarization is an active research area in the field of natural language processing. Huge amount of information in the internet necessitates the development of automatic summarization systems. There are two types of summarization techniques: Extractive and Abstractive. Extractive summarization selects important sentences from the text and produces summary as it is present in the original document. Abstractive summarization systems will provide a summary of the input text as is generated by human beings. Abstractive summary requires semantic analysis of text. Limited works have been carried out in the area of abstractive summarization in Indian languages especially in Malayalam. Only extractive summarization methods are proposed in Malayalam. In this paper, an abstractive summarization system for Malayalam documents using clause identification method is proposed. As part of this research work, a POS tagger and a morphological analyzer for Malayalam words in cricket domain are also developed. The clauses from input sentences are identified using a modified clause identification algorithm. The clauses are then semantically analyzed using an algorithm to identify semantic triples - subject, object and predicate. The score of each clause is then calculated by using feature extraction and the important clauses which are to be included in the summary are selected based on this score. Finally an algorithm is used to generate the sentences from the semantic triples of the selected clauses which is the abstractive summary of input documents. [ABSTRACT FROM AUTHOR]

The article deals with the relevant problem of automatic arbitrary text summary obtaining. Several works in this field are analyzed using comparison and classification methods. The problem of obtaining a text summary by means of an answer to a random question is emphasized. The identification of semantic relations between sentences using a set of rules based on syntax and semantics of a language is described. These rules are represented in the form of regular expressions – patterns that consist of characters and metacharacters and set search rules. Taking into account semantic coherence features, an improved method of sentences similarity calculation to identify the measure of inclusion of one sentence into another one is developed. This method helps to define more precisely logical stress on the words within automatic summarization and detect contradictions. A modified automatic summarization method, oriented at a specific problem is suggested. It is concluded that the proposed method is quiet effective in the process of automatic search for answers to questions in the Internet. [ABSTRACT FROM AUTHOR]

Emerging Artificial Intelligence (AI) technology has brought machines to reach an equal or even superior level compared to human capabilities in several fields; nevertheless, among many other fields, making a computer able to understand human language still remains a challenge. When dealing with speech understanding, Automatic Speech Recognition (ASR) is used to generate transcripts, which are processed with text-based tools targeting Spoken Language Understanding (SLU). Depending on the ASR quality (which further depends on speech quality, the complexity of the topic, environment etc.), transcripts contain errors, which propagate further into the processing pipeline. Subjective tests show on the other hand, that humans understand quite well ASR-closed captions, despite the word and punctuation errors. Through word embedding based semantic parsing, the present paper is interested in quantifying the semantic bias introduced by ASR error propagation. As a special use case, speech summarization is also evaluated with regard to ASR error propagation. We show, that despite the higher word error rates seen with the highly inflectional Hungarian, the semantic space suffers least impact than the difference in Word Error Rate would suggest. [ABSTRACT FROM AUTHOR]

A long-running event represents a continuous stream of information on a given topic, such as natural disasters, stock market updates, or even ongoing customer relationship. These news stories include hundreds of individual, time-dependent texts. Simultaneously, new technologies have profoundly transformed the way we consume information. The need to obtain quick, relevant, and digest updates continuously has become a crucial issue and creates new challenges for the task of automatic document summarization. To that end, we introduce an innovative unsupervised method based on two competing sequence-to-sequence models to produce short updated summaries. The proposed architecture relies on several parameters to balance the outputs from the two autoencoders. This relation enables the overall model to correlate generated summaries with relevant information coming from both current and previous news iterations. Depending on the model configuration, we are then able to control the novelty or the consistency of terms included in generated summaries. We evaluate our method on a modified version of the TREC 2013, 2014, and 2015 datasets to track continuous events from a single source. We not only achieve state-of-the-art performance similar to other more complex unsupervised sentence compression approaches, but also influence the information included in the model in the summaries. • Incremental summarization of real-time news events. • Unsupervised autoencoder for summarization of short texts related to the same event. • Semi-extractive method of accounting for information updates in the sentence compression task. • Promoting novelty or cohesion of information in incremental and update summarization processes. [ABSTRACT FROM AUTHOR]

We introduce a comparative approach to political discourse that uses english Wikipedia as a repository of notions, automatic translation as a means of projecting a discourse, and the search for targeted information as a method of contextualization. The english Wikipedia is the most complete encyclopedic resource at the moment and most used by voice assistants. Automatic translation allows you to get away from the elements of language while preserving the values. Contextualization finally makes it possible to objectively measure the relative distance between a speech and a set of concepts. We apply this methodology to the case of the uses of feminine references in the speeches of the Far right in France. [ABSTRACT FROM AUTHOR]

Firms often include summaries with earnings releases. However, manager-generated summaries may be prone to strategic tone and content management, compared to the underlying disclosures they summarize. In contrast, computer algorithms can summarize text without human intervention and may provide useful summary information with less bias. We use multiple methods to provide evidence regarding the characteristics of algorithm-based summaries of earnings releases compared to those provided by managers. Results suggest that automatic summaries are generally less positively biased, often without sacrificing relevant information. We then conduct an experiment to test whether these differing attributes of automatic and management summaries affect individual investors' judgments. We find that investors who receive an earnings release accompanied by an automatic summary arrive at more conservative (i.e., lower) valuation judgments and are more confident in those judgments. Overall, our results suggest that summaries affect investors' judgments and that these effects differ for management and automatic summaries. [ABSTRACT FROM AUTHOR]

Automatic text summarization technology has a very broad application prospect in many fields ? such as web search and browsing recommendation. The classic text summarization algorithm uses statistical methods to extract article keywords and topic sentences. It ignores semantic and grammatical information of the text to some extent. As distributed word vector embedding technology has been widely used in text summarization in recent years，we propose an automatic text summarization method based on word vector generation. This method mainly includes four modules： word vector generation，paragraph vector generation based on word vector，keyword extraction? and topic sentence extraction， through which an automatic text summarization of the document can finally be achieved. Experimental results show that the improved automatic text summarization method can extract topic sentences effectively. [ABSTRACT FROM AUTHOR]

Copyright of Procesamiento del Lenguaje Natural is the property of Sociedad Espanola para el Procesamiento del Lenguaje Natural and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Concept maps are graphical tools for representation and construction of knowledge. The manual construction of a concept map requires time and cognitive effort, this being increased when the map should not represent the cognitive structure of the author, but rather, the information expressed in a text written by another author. Therefore, we propose a computational approach for concept map mining from texts in Portuguese that aims to represent the text in summary form through concepts and relationships. To this end, we define a technological architecture that includes the services of: (i) text formatting, removing characters and designing of the text; (ii) domain identification, information retrieval techniques to identify the domain to which refers the text; (iii) elements extractor, natural language processing techniques on the text to extract concept-relation-concept propositions; (iv) element summarizer, supported by graph analysis to identify the relevant concepts on the map; and (v) map visualization, presentation of the propositions in graphic form. The approach developed presents satisfactory results and contributes exceptionally to the summarization of texts to identify the relevant concepts of the text while maintaining its several and most important characteristics. Furthermore, this research introduces the specification of a project to provide computational resources for processing, handling and extraction of conceptual maps. [ABSTRACT FROM AUTHOR]

Automatic Text Summarization is the process of creating a compressed representation of one or more related documents, keeping only the most valuable information. The extractive approach for summarization is the most studied and aims to generate a compressed version of a document by identifying, ranking, and selecting the most relevant sentences or phrases from a text. The selected sentences go verbatim into the summary. However, this strategy may yield incoherent summaries, as pronominal coreferences may appear unbound. To alleviate this problem, this paper proposes a method that solves unbound pronominal anaphoric expressions, automatically enabling the cohesiveness of the extractive summaries. The proposed method can be applied to two distinct scenarios. The first one aims to find and fix unbound anaphoric expressions present in the generated summaries at a post-processing stage; whereas the second one is performed at the preprocessing stage of the proposed pipeline and generates an intermediate version of the input document that resolves the unbound pronominal coreferences. The proposed solution was evaluated on the CNN news corpus using the seventeen summarization techniques most widely acknowledged in the literature and four state-of-the-art summarization systems. Moreover, it also provides a comparative evaluation concerning two distinct assessment scenarios which are compared to a baseline. The experiments performed achieved very encouraging quantitative and qualitative results. [ABSTRACT FROM AUTHOR]

Recent years have witnessed a drastic growth of various videos in real-life scenarios, and thus there is an increasing demand for a quick view of such videos in a constrained amount of time. In this paper, we focus on automatic summarization of surveillance videos and present a new key-frame selection method for this task. We first introduce a dissimilarity measure based on f-divergence by a symmetric strategy for multiple change-point detection and then use it to segment a given video sequence into a set of non-overlapping clips. Key frames are extracted from the resulting video clips by a typical clustering procedure for final video summary. Through experiments on a wide range of testing data, excellent performances, outperforming given state-of-the-art competitors, have been demonstrated which suggests good potentials of the proposed method in real-world applications. [ABSTRACT FROM AUTHOR]