Inducing Rich Interaction Structures Between Words for Document-Level Event Argument Extraction

2021 ◽  
pp. 703-715
Author(s):  
Amir Pouran Ben Veyseh ◽  
Franck Dernoncourt ◽  
Quan Tran ◽  
Varun Manjunatha ◽  
Lidan Wang ◽  
...  
Keyword(s):  
Event Argument ◽  
Rich Interaction ◽  
Document Level

scholarly journals A Prior Information Enhanced Extraction Framework for Document-level Financial Event Extraction

2021 ◽  
pp. 1-12
Author(s):  
Haitao Wang ◽  
Tong Zhu ◽  
Mingtao Wang ◽  
Guoliang Zhang ◽  
Wenliang Chen
Keyword(s):  
Event Detection ◽  
Prior Information ◽  
Event Extraction ◽  
Language Models ◽  
Event Type ◽  
Event Argument ◽  
Implicit Information ◽  
Share Task ◽  
Financial Domain ◽  
Document Level

Abstract Document-level financial event extraction (DFEE) is the task of detecting event and extracting the corresponding event arguments in financial documents, which plays an important role in information extraction in the financial domain. This task is challenging as the financial documents are generally long text and event arguments of one event may be scattered in different sentences. To address this issue, we propose a novel Prior Information Enhanced Extraction framework (PIEE) for DFEE, leveraging prior information from both event types and pre-trained language models. Specifically, PIEE consists of three components: event detection, event argument extraction, and event table filling. In event detection, we identify the event type. Then, the event type is explicitly used for event argument extraction. Meanwhile, the implicit information within language models also provides considerable cues for event arguments localization. Finally, all the event arguments are filled in an event table by a set of predefined heuristic rules. To demonstrate the effectiveness of our proposed framework, we participate the share task of CCKS2020 Task5-2: Document-level Event Arguments Extraction. On both Leaderboard A and Leaderboard B, PIEE takes the first place and significantly outperforms the other systems.

Developing a K-CDA Implementation Guide for Applying Health Information Exchange Service in South Korea (Preprint)

2020 ◽  
Author(s):  
Sung Won Jung ◽  
Sungchul Bae ◽  
Donghyeong Seong ◽  
Byoung-Kee Yi
Keyword(s):  
Information Exchange ◽  
Health Information Exchange ◽  
Expert Committee ◽  
Healthcare Information ◽  
Entry Level ◽  
Template Library ◽  
Value Sets ◽  
Definition Of ◽  
Document Level ◽  
Section Level

BACKGROUND Through several years of the healthcare information exchange based on the HIE project, some problems were found in the CDA documents generated. OBJECTIVE To fix some problems, we developed the K-CDA Implementation Guide (K means S. Korea) that conforms to the HL7 CDA, and suits the domestic conditions regarding the healthcare information. METHODS We achieved by analyzing HIE guideline and the U.S. C-CDA, and comparing each item. The items that required further discussion were reviewed by the expert committee. Based on the reviews, the previously developed templates were revised. RESULTS A total of 35 CDA templates were developed: five document-level templates, fourteen section-level templates, and sixteen entry-level templates. The 28 value sets used in the templates have been improved and the OIDs for HIE have been redefined CONCLUSIONS The K-CDA IG allows management in the form of a template library based on the definition of the General K-Header and the structured templates. This enables the K-CDA IG to respond to the expansion of national HIE templates with flexibility. For the K-CDA IG, the CDA template in current use was incorporated to the greatest extent possible, to minimize the scope of modifications. It enables the national HIE and the HIE with countries abroad.

The “bundling” hypothesis and the disparate functions of little v

2017 ◽  
Author(s):  
Heidi Harley
Keyword(s):  
External Argument ◽  
Event Argument ◽  
Accusative Case ◽  
Form Part ◽  
Korean And English

Following Pylkkänen (2002), among others, many of the functions of the vP have been distributed between two independent projections: VoiceP and vP. Pylkkänen proposed a “bundling” parameter: some languages project a single bundled Voice/vP, and all functions depend on that single projection, and others project VoiceP and vP separately, and functions are distributed. The chapter first reviews the roles ascribed to these projections: (i) external argument introduction, (ii) event argument introduction, (iii) accusative case checking, (iv) introduction of causative or inchoative semantics, (v) verbalizing of nonverbal material, and (vi) demarcating a cycle. The chapter then reviews support for Pylkkänen’s parametric view of Voice-bundling from, e.g., Hiaki, Turkish, Korean, and English. Results on causatives from Key (2013) and Jung (2014) suggest that the projection sequence dominating v may form part of a predetermined projection hierarchy. The constraint against stacking productive morphological causatives may thus be attributed to the extended verbal projection.

A Survey on Document-level Neural Machine Translation

2021 ◽  
Vol 54 (2) ◽  
pp. 1-36
Author(s):  
Sameen Maruf ◽  
Fahimeh Saleh ◽  
Gholamreza Haffari
Keyword(s):  
Machine Translation ◽  
Language Processing ◽  
Research Field ◽  
Translation Process ◽  
Future Directions ◽  
Translation Quality ◽  
Current State ◽  
Evaluation Strategies ◽  
Almost All ◽  
Document Level

Machine translation (MT) is an important task in natural language processing (NLP), as it automates the translation process and reduces the reliance on human translators. With the resurgence of neural networks, the translation quality surpasses that of the translations obtained using statistical techniques for most language-pairs. Up until a few years ago, almost all of the neural translation models translated sentences independently , without incorporating the wider document-context and inter-dependencies among the sentences. The aim of this survey article is to highlight the major works that have been undertaken in the space of document-level machine translation after the neural revolution, so researchers can recognize the current state and future directions of this field. We provide an organization of the literature based on novelties in modelling and architectures as well as training and decoding strategies. In addition, we cover evaluation strategies that have been introduced to account for the improvements in document MT, including automatic metrics and discourse-targeted test sets. We conclude by presenting possible avenues for future exploration in this research field.

Sentiment Analysis of Comparative Sentences for Chinese Document

2012 ◽  
Vol 157-158 ◽  
pp. 1079-1082
Author(s):  
Guo Shi Wu ◽  
Xiao Yin Wu ◽  
Jing Jing Wei
Keyword(s):  
Sentiment Analysis ◽  
Opinion Mining ◽  
Sentiment Classification ◽  
The Third ◽  
Sentence Patterns ◽  
Simple Sentences ◽  
Document Level ◽  
Level Analysis

One of the most widely-studied sub-problems of opinion mining is sentiment classification, which includes three study levels: word, sentence and document. At the third level, most of the existing methods ignore comparative sentences which have particular sentence patterns and may lower the precision of the document-level analysis. This paper studies sentiment analysis of comparative sentences. The aim is to determine whether opinions expressed in a comparative sentence are positive or negative. Experiments of comparing with document-level sentiment analysis based on simple sentences shows the effectiveness of the proposed method.

Sign in / Sign up

Export Citation Format

Share Document