Extracting Explicit and Implicit Causal Relations from Sparse, Domain-Specific Texts

Automatically extracting causal relations from texts is a challenging task in Natural Language Processing (NLP). Most existing methods focus on extracting intra-sentence or explicit causality, while neglecting the causal relations that expressed implicitly or hidden in inter-sentences. In this paper, we propose Cascaded multi-Structure Neural Network (CSNN), a novel and unified model that extract intersentence or implicit causal relations from Chinese Corpus, without relying on external knowledge. The model employs Convolutional Neural Network (CNN) to capture important features as well as causal structural pattern. Self-attention mechanism is designed to mine semantic and relevant characteristics between different features. The output of CNN and self-attention structure are concatenated as higher-level phrase representations. Then Conditional Random Field (CRF) layer is employed to calculate the label of each word in inter-sentence or implicit causal relation sentences, which improves the performance of inter-sentence or implicit causality extraction. Experimental results show that the proposed model achieves state-of-the-art results, improved on three datasets, when compared with other methods.

show abstract

“…Rule-based [9] methods or traditional machine learning methods [1] contribute a lot on causality mining. However, there are still many drawbacks.…”

Section: Formsmentioning

confidence: 99%

“…Numerous rule-based [9] methods have been dedicated on causality mining. Kontos and Sidiropoulou [13] used causal language patterns and hand-crafted causal relation templates to detect causal relations which hidden in contexts.…”

Section: Non-statistical Approachesmentioning

confidence: 99%

Inter-sentence and Implicit Causality Extraction from Chinese Corpus

Jin

Wang

Luo

et al. 2020

Advances in Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

show abstract

“…However, this method is not able to detect the causes and the effects. Ittoo and Bouma [17] present a semi-supervised method for automatic extraction of high quality causal relations from domain-specific, sparse corpora. In this work, they initially acquire a set of explicit and implicit lexico-syntactic patterns from Wikipedia.…”

Section: Semi-automatic Causal Pattern Learningmentioning

confidence: 99%

Event Causality Extraction from Natural Science Literature

Barik¹,

Marsi²,

Öztürk³

2016

RCS

View full text Add to dashboard Cite

We aim to develop a text mining framework capable of identifying and extracting causal dependencies among changing variables (or events) from scientific publications in the cross-disciplinary field of oceanographic climate science. The extracted information can be used to infer new knowledge or to find out unknown hypotheses through reasoning, which forms the basis of a knowledge discovery support system. Automatic extraction of causal knowledge from text content is a challenging task. Generally, the approaches of causal relation identification proposed in the literature target specific domain such as online news or biomedicine as the domain has significant influence on causality expressions found in the domain texts. Therefore, the existing models of causality extraction may not be directly portable to other/new domains. In this paper, we describe the nature of causation observed in climate science domain, review the state-of-the-art approaches in causal knowledge extraction from text and carefully select the methods and resources most likely to be applicable to the considered domain.

show abstract

“…3. The third paper, Extracting Explicit and Implicit Causal Relations from Sparse, Domain-Specific Texts, extended version of the one in [4]) by Ashwin Ittoo and Gosse Bouma, from the University of Groningen, deals with the task of mining casual relations from domain specific corpora, in order to discover new knowledge from nonstructured databases. Causality is a complex phenomenon through which there is a connection between an agent event (cause of something) and a resultant effect (the principle of cause-effect).…”

Section: The First Paper Emotinet a Knowledge Base For Emotion Detementioning

confidence: 99%

An overview of the Applications of Natural Language to Information Systems

Martínez-Barco

Métais

Llopis

et al. 2013

Data & Knowledge Engineering

View full text Add to dashboard Cite

This introduction provides an overview of the state-of-the-art technology in Applications of Natural Language to Information Systems. Specifically, we analyze the need for such technologies to successfully address the new challenges of modern information systems, in which the exploitation of the Web as a main data source on business systems becomes a key requirement. It will also discuss the reasons why Human Language Technologies themselves have shift their focus onto new areas of interest very directly linked to the development of technology for the treatment and understanding of Web 2.0. These new technologies are expected to be future interfaces for the new information systems to come. Moreover, we will review current topics of interest to this research community, and will present the selection of manuscripts that have been chosen by the program committee of the NLDB 2011 conference as representative cornerstone research works, especially highlighting their contribution to the advancement of such technologies.

show abstract

Extracting Explicit and Implicit Causal Relations from Sparse, Domain-Specific Texts

Cited by 41 publications

References 11 publications

Inter-sentence and Implicit Causality Extraction from Chinese Corpus

Inter-sentence and Implicit Causality Extraction from Chinese Corpus

Event Causality Extraction from Natural Science Literature

An overview of the Applications of Natural Language to Information Systems

Contact Info

Product

Resources

About