Search

Processing and structuring medical texts today are still difficult tasks due to specific terminology and abbreviations, a large number of errors and a lack of labeled data for high-quality model training. One of the important tasks is event extraction from unstructured texts. The most efficient event extraction methods to date are deep learning methods, however, they require labeled data to train them. Manual marking is labor-intensive, and expert knowledge is required for high-quality marking. A good alternative is automated pre-labeling. In the public domain, it is difficult to find works on preliminary data markup for the task of extracting events. In this paper, we propose a methodology for rude event extraction based on rules that evaluate the elements of a syntax tree. This methodology can also be used for pre-labeling data for training machine learning and deep learning models. [ABSTRACT FROM AUTHOR]

Most existing methods regard event extraction as the classification task. They not only heavily rely on named entity recognition, causing error propagation, but are also inefficient in low resource scenarios. To deal with above challenges, we propose an improved machine reading comprehension (MRC) approach, namely MRCBEE. Firstly, a new paradigm is applied to redefine event extraction as MRC task by designing question templates for event detection and argument extraction tasks. Specially, Question-Context Bridging is a new graph structure drawn to reconstruct the semantic relationship between the template and the text, in order to strengthen the prompt role of question templates. Then, a cross-domain attention module is designed to integrate both the semantic feature and global feature of words. A new GNN based on gated mechanism is proposed to capture the global feature and filter the information of invalid neighbors. Finally, the results of our experiments show that MRCBEE achieves better performance than the state-of-the-art methods on ACE2005 and ERE datasets. And it outperforms prior methods in low resource and complex text scenarios. [ABSTRACT FROM AUTHOR]

Low-resource event extraction presents a significant challenge in real-world applications, particularly in domains like pharmaceuticals, military and law, where data is frequently insufficient. Data augmentation, as a direct method for expanding samples, is considered an effective solution. However, existing data augmentation methods often suffer from text fluency issues and label hallucination. To address these challenges, we propose a framework called Agent-DA, which leverages multi-agent collaboration for event extraction data augmentation. Specifically, Agent-DA follows a three-step process: data generation by the large language model, collaborative filtering by both the large language model and small language model to discriminate easy samples, and the use of an adjudicator to identify hard samples. Through iterative and selective augmentation, our method significantly enhances both the quantity and quality of event samples, improving text fluency and label consistency. Extensive experiments on the ACE2005-EN and ACE2005-EN+ datasets demonstrate the effectiveness of Agent-DA, with F1-score improvements ranging from 0.15% to 16.18% in trigger classification and from 2.2% to 15.67% in argument classification. • Multi-agent collaboration for enhancing event extraction data augmentation. • Reduces label hallucination in data generated by a large language model. • Ranked dataset used for training the adjudicator in sample judgments. [ABSTRACT FROM AUTHOR]

Event extraction has a trend in identifying event triggers and arguments in a unified framework, which has the advantage of avoiding the cascading failure in pipeline methods. The main problem is that joint models usually assume a one-to-one relationship between event triggers and arguments. It leads to the argument multiplexing problem, in which an argument mention can serve different roles in an event or shared by different events. To address this problem, we propose a multigraph-based event extraction framework. It allows parallel edges between any nodes, which is effective to represent semantic structures of an event. The framework enables the neural network to map a sentence(s) into a structurized semantic representation, which encodes multi-overlapped events. After evaluated on four public datasets, our method achieves the state-of-the-art performance, outperforming all compared models. Analytical experiments show that the multigraph representation is effective to address the argument multiplexing problem and helpful to advance the discriminability of the neural network for event extraction. [ABSTRACT FROM AUTHOR]

The task of Cross-Lingual Event Extraction (CLEE) aims to transfer event knowledge from rich-resourced languages to low-resourced languages under a zero-shot cross-lingual setting. Despite presenting successful transfer abilities, existing CLEE methods face two limitations. Firstly, a majority of works concentrate on single source CLEE ignoring the benefits of multi-source potentials. Secondly, universal dependency trees have been adopted for learning shared syntactic features across languages while leaving exploration of language-specific features (e.g. word order) insufficient. In this work, we investigate the effectiveness of multi-source CLEE and propose a Syntax-enhanced Parameter Generation Network (SPGN) for the task. SPGN mainly consists of two components, i.e., a parameter generation tree network for determining language-private syntactic knowledge, and an adversarial network for learning language-shared representations. Experiments on widely used ACE2005 and recently released MINION dataset show that our proposed method significantly outperforms existing baseline systems on total 10 languages. Specifically, we observed an improvement of 1.9% in average F1-score over the best-performing baseline on ACE2005 and 3.5% on MINION, highlighting the benefits of leveraging both language-specific and shared features in CLEE. Further analysis confirms the individual effectiveness of each component of the SPGN model, suggesting its potential applicability in other low-resourced language processing tasks. [ABSTRACT FROM AUTHOR]

Event Extraction (EE) is an essential task in natural language processing that aims to mine events occurring in event mentions represent events using event records, which usually consist of event types, trigger words, argument elements corresponding to roles in the event types. Recently, prompt-based generative models have been developed to extract events. However, these prompt-based generative studies have ignored the fact that the strong language comprehension capability of the pre-trained language model (PLM) can analyze extract the potential role relationships in multiple templates for more information that can help extract argument elements. To determine the extended templates that can help the model for event extraction, we propose the multiple template choice model (MTCM), which designs an extended event type mining module to automatically mine the extended event types in the event mention uses the templates corresponding to the extended event types to interact with the template, corresponding to the currently to-be-extracted event type of event mention, in a multi-template information interaction, which gives the model more information guides the PLM for event extraction. To validate our model, we used two widely used datasets in the event extraction domain, ACE2005-EN ERE-EN. The experimental results show that our model achieves state-of-the-art performance on the ACE2005-EN dataset significantly improves the ERE dataset. In addition, according to the results, our model can be effectively adapted to low-resource environments. [ABSTRACT FROM AUTHOR]

The object and evidence are two important parts of financial fraud detection (FFD). In traditional FFD tasks, the detected object generally refers to the enterprise involved in typical financial fraud events, and the evidence comes from corporate statements such as auditing report, etc. Thus, previous FFD methods based on financial statements are limited by the scope of detected objects and the availability of evidence. In this study, we design a financial event library to enlarge the detection scope and expand the source of evidence to financial news, and thus financial entities involved in high-risk events are extracted. The financial event library contains common negative events, each of which corresponds to a negative behaviour that may increase financial risk. Moreover, we propose a novel method to convert the event detection task into the Q&A mode task, which also contributes to the enlargement of the original hand-built dataset. Compared with existing methods on the enlargement of financial dataset, our approach does not require additional annotation. We apply our method into Chinese financial news corpus, and achieves good performance in the extraction task. [ABSTRACT FROM AUTHOR]

Event extraction aims to determine entity mentions, event triggers and argument roles in text, such as news articles. For this task, dependency syntax has been recognized as a valuable source of information. Previous work integrates dependency trees by using one-best results from a parser, where potential incorrect dependencies may impact the event extraction performance. We propose to use syntax features in an implicit approach, by adopting Soft Head Vectors (SHV) drawn from a well-trained parser as an adjacency matrix, densely connecting all words with the probabilistic head scores. SHV can be also regarded as dependency forests where multiple possible structures can be taken into account. A two-phase graph neural network is adopted to represent the forests, automatically differentiating the salient syntactic information from noisy parsers. Experiments on the two public datasets show that our model outperforms various baselines including graph-based one-best tree as well as recent transition-based decoding, giving the state-of-the-art results in the literature. [ABSTRACT FROM AUTHOR]

Event extraction is a challenging problem in information extraction, designed to extract structured information from unstructured text. The existing event extraction methods are mostly based on the pipeline model and use the traditional language representation model for word embedding. It is impossible to model the polysemy and the features is very simple, besides, the pipeline model is also easy to generate cascading errors. Aiming at the above problems, we propose a joint event extraction model, which combines part-of-speech features, entity features and word representation of BERT model. Because of the BERT can't capture local features, we use multi-CNN to capture the local features of different scales. and further use BiLSTM to capture context features to assist joint event extraction. Experimental results show that the precision, recall and F 1 of the method on the CEC dataset are 88.83%, 90.91% and 89.86%, respectively, which are 1.61%, 3.6% and 2.6% higher than the PLMEE model. [ABSTRACT FROM AUTHOR]

Social networks have become an essential part of daily life, and hence every real-world activity finds its place in this virtual world. The present paper proposes a methodology to find localized micro-events from the social network stream. The method is named CommunityINDICATOR. A concept of 'separation of concerns' from the software design principle is incorporated in the methodology to reduce the execution time drastically from existing state-of-the-art methods of event detection. In order to reduce the execution time, the algorithm first generates an interaction graph from the social stream and applies community detection followed by a clustering algorithm onto it to detect micro-level events. Experiments have been conducted on Twitter data stream of 5 different cities on three different continents with the size of 2 million tweets. We have used well known quality metrics such as precision, recall, F1-score, accuracy, and execution time to compare performance with other state-of-the-art methodologies. The proposed CommunityINDICATOR provides up to 30% higher accuracy than EvenTweet and SEDTWik in similar execution times. An improvement of 11% to 51% and 17% to 57% in execution time is observed for the proposed algorithm in comparison with TwiiterNews and EventX, respectively, for different datasets. • An e2e event detection pipeline by introducing "separation of concerns" concept. • Achieving time benefit in existing unsupervised method by creating interaction graph. • Identifying very localized events with high inference speed and detection accuracy. • Experiments performed to check the effect of time and partition algorithm in pipeline. [ABSTRACT FROM AUTHOR]

Automated Text Simplification (ATS) aims to transform complex texts into their simpler variants which are easier to understand to wider audiences and easier to process with natural language processing (NLP) tools. While simplification can be applied on lexical, syntactic, and discourse level, all previously proposed ATS systems only operated on the first two levels, thus failing at simplifying texts on the discourse level. We present a semantically-motivated ATS system which is the first system that is applied on the discourse level. By exploiting the state-of-the-art event extraction system, it is the first ATS system able to eliminate large portions of irrelevant information from texts, by maintaining only those parts of the original text that belong to factual event mentions. A few handcrafted rules ensure that the output of the system is syntactically simple, by placing each factual event mention in a separate short sentence, while the state-of-the-art unsupervised lexical simplification module, based on using word embeddings, replaces complex and infrequent words with their simpler variants. We perform a thorough evaluation, both automatic and manual, showing that our system produces more readable and simpler texts than the state-of-the-art ATS systems. Our newly proposed post-editing evaluation further reveals that our system requires less human effort for correcting grammaticality and meaning preservation on news articles than the state-of-the-art ATS system. [ABSTRACT FROM AUTHOR]

Document-level event argument extraction (EAE), a critical task for event knowledge acquisition, seeks to find event arguments beyond the sentence level. Previous approaches for this task adopt supervised learning, which suffers from the data scarcity issue, especially in low-resource situations. To address this challenge, we propose a new method for document-level EAE that addresses the data scarcity from two perspectives: self-augmentation and cross-domain joint training. On the one hand, our method can generate additional training samples from an existing dataset using pre-trained language models, considerably expanding the training set; on the other hand, our method can learn with out-of-domain datasets with varying event schema, using a cross-domain joint training framework. To address data quality problems and domain mismatches, we also provide a new noise filtering method based on a teacher–student framework. The extensive experiments demonstrate that our approach delivers state-of-the-art performance on benchmark datasets, and that it is especially effective in low-resource situations. [Display omitted] • A new approach for resolving the data sparsity issue in document-level event argument extraction from two complementing viewpoints. • A self-augmentation technique that combines pre-trained language models and a label-conditioned pre-training procedure to retain word-label consistency. • A cross-domain joint training framework for knowledge transfer from datasets of varied granularity, task specifications, and the event schema description language. • A novel noise filtering method in teacher–student framework to mitigate the data quality issue. • State-of-the-art performance on standard datasets. [ABSTRACT FROM AUTHOR]

In this article, we describe a system that reads news articles in four different languages and detects what happened, who is involved, where and when. This event-centric information is represented as episodic situational knowledge on individuals in an interoperable RDF format that allows for reasoning on the implications of the events. Our system covers the complete path from unstructured text to structured knowledge, for which we defined a formal model that links interpreted textual mentions of things to their representation as instances. The model forms the skeleton for interoperable interpretation across different sources and languages. The real content, however, is defined using multilingual and cross-lingual knowledge resources, both semantic and episodic. We explain how these knowledge resources are used for the processing of text and ultimately define the actual content of the episodic situational knowledge that is reported in the news. The knowledge and model in our system can be seen as an example how the Semantic Web helps NLP. However, our systems also generate massive episodic knowledge of the same type as the Semantic Web is built on. We thus envision a cycle of knowledge acquisition and NLP improvement on a massive scale. This article reports on the details of the system but also on the performance of various high-level components. We demonstrate that our system performs at state-of-the-art level for various subtasks in the four languages of the project, but that we also consider the full integration of these tasks in an overall system with the purpose of reading text. We applied our system to millions of news articles, generating billions of triples expressing formal semantic properties. This shows the capacity of the system to perform at an unprecedented scale. [ABSTRACT FROM AUTHOR]

Knowledge graphs have gained increasing popularity in the past couple of years, thanks to their adoption in everyday search engines. Typically, they consist of fairly static and encyclopedic facts about persons and organizations–e.g. a celebrity’s birth date, occupation and family members–obtained from large repositories such as Freebase or Wikipedia. In this paper, we present a method and tools to automatically build knowledge graphs from news articles. As news articles describe changes in the world through the events they report, we present an approach to create Event-Centric Knowledge Graphs (ECKGs) using state-of-the-art natural language processing and semantic web techniques. Such ECKGs capture long-term developments and histories on hundreds of thousands of entities and are complementary to the static encyclopedic information in traditional knowledge graphs. We describe our event-centric representation schema, the challenges in extracting event information from news, our open source pipeline, and the knowledge graphs we have extracted from four different news corpora: general news (Wikinews), the FIFA world cup, the Global Automotive Industry, and Airbus A380 airplanes. Furthermore, we present an assessment on the accuracy of the pipeline in extracting the triples of the knowledge graphs. Moreover, through an event-centered browser and visualization tool we show how approaching information from news in an event-centric manner can increase the user’s understanding of the domain, facilitates the reconstruction of news story lines, and enable to perform exploratory investigation of news hidden facts. [ABSTRACT FROM AUTHOR]

Event extraction (EE) occupies an important position in information extraction. Recently, deep neural network methods have been demonstrated to learn potential features well. However, existing networks for EE suffer from the following challenges: (1) Event argument extraction relies heavily on the classification of event triggers. (2) Previous works fail to exploit the predefined label representations for EE. (3) The interactive information between candidate arguments has not been fully exploited. To address the above mentioned problems, in this paper, we propose an advanced multi-task learning framework, named TNC , based on a fresh concept proposed by us: T rigger is N on- C entral, in which event argument extraction no longer depends on the event triggers but is performed synchronously with it. Our TNC extracts multiple event triggers and arguments simultaneously by adopting label representations and an auxiliary task, named Sentence Event Identification (SEI), which is devised to extract the event types contained in a sentence. In addition, we design a special symbol to merge the representation of candidate arguments over the Transformer encoder. We experiment on the widely used ACE 2005 corpora and TAC-KBP 2015, and the experimental results have proved that our model achieves state-of-the-art compared to other models, with higher effectiveness and adaptability. [ABSTRACT FROM AUTHOR]

With the number of documents describing real-world events and event-oriented information needs rapidly growing on a daily basis, the need for efficient retrieval and concise presentation of event-related information is becoming apparent. Nonetheless, the majority of information retrieval and text summarization methods rely on shallow document representations that do not account for the semantics of events. In this article, we present event graphs, a novel event-based document representation model that filters and structures the information about events described in text. To construct the event graphs, we combine machine learning and rule-based models to extract sentence-level event mentions and determine the temporal relations between them. Building on event graphs, we present novel models for information retrieval and multi-document summarization. The information retrieval model measures the similarity between queries and documents by computing graph kernels over event graphs. The extractive multi-document summarization model selects sentences based on the relevance of the individual event mentions and the temporal structure of events. Experimental evaluation shows that our retrieval model significantly outperforms well-established retrieval models on event-oriented test collections, while the summarization model outperforms competitive models from shared multi-document summarization tasks. [ABSTRACT FROM AUTHOR]

An Ontology-based Fuzzy Event Extraction (OFEE) agent for Chinese e-news summarization is proposed in this article. The OFEE agent contains Retrieval Agent (RA), Document Processing Agent (DPA) and Fuzzy Inference Agent (FIA) to perform the event extraction for Chinese e-news summarization. First, RA automatically retrieves Internet e-news periodically, stores them into the e-news repository, and sends them to DPA for document processing. Then, the DPA will utilize the Chinese Part-of-speech (POS) tagger provided by Chinese knowledge information processing group to process the retrieved e-news and filter the Chinese term set by Chinese term filter. Next, the FIA and Event Ontology Filter (EOF) extract the e-news event ontology based on the Chinese term set and domain ontology. Finally, the Summarization Agent (SA) will summarize the e-news by the extracted-event ontology. By the simulation, the proposed method can summarize the Chinese weather e-news effectively. [Copyright &y& Elsevier]

Event extraction (EE) is one of the core information extraction tasks, whose purpose is to automatically identify and extract information about incidents and their actors from texts. This may be beneficial to several domains such as knowledge base construction, question answering and summarization tasks, to name a few. The problem of extracting event information from texts is longstanding and usually relies on elaborately designed lexical and syntactic features, which, however, take a large amount of human effort and lack generalization. More recently, deep neural network approaches have been adopted as a means to learn underlying features automatically. However, existing networks do not make full use of syntactic features, which play a fundamental role in capturing very long-range dependencies. Also, most approaches extract each argument of an event separately without considering associations between arguments which ultimately leads to low efficiency, especially in sentences with multiple events. To address the above-referred problems, we propose a novel joint event extraction framework that aims to extract multiple event triggers and arguments simultaneously by introducing shortest dependency path in the dependency graph. We do this by eliminating irrelevant words in the sentence, thus capturing long-range dependencies. Also, an attention-based graph convolutional network is proposed, to carry syntactically related information along the shortest paths between argument candidates that captures and aggregates the latent associations between arguments; a problem that has been overlooked by most of the literature. Our results show a substantial improvement over state-of-the-art methods on two datasets, namely ACE 2005 and TAC KBP 2015. [ABSTRACT FROM AUTHOR]