A survey of the extraction and applications of causal relations

Abstract

Causationin written natural language can express a strong relationship between events and facts. Causation in the written form can be referred to as a causal relation where a cause event entails the occurrence of an effect event. A cause and effect relationship is stronger than a correlation between events, and therefore aggregated causal relations extracted from large corpora can be used in numerous applications such as question-answering and summarisation to produce superior results than traditional approaches. Techniques like logical consequence allow causal relations to be used in niche practical applications such as event prediction which is useful for diverse domains such as security and finance. Until recently, the use of causal relations was a relatively unpopular technique because the causal relation extraction techniques were problematic, and the relations returned were incomplete, error prone or simplistic. The recent adoption of language models and improved relation extractors for natural language such as Transformer-XL (Dai et al. (2019). Transformer-xl: Attentive language models beyond a fixed-length context. arXiv preprint arXiv:1901.02860) has seen a surge of research interest in the possibilities of using causal relations in practical applications. Until now, there has not been an extensive survey of the practical applications of causal relations; therefore, this survey is intended precisely to demonstrate the potential of causal relations. It is a comprehensive survey of the work on the extraction of causal relations and their applications, while also discussing the nature of causation and its representation in text.

Causal Extraction to Measure Media Bias? A Comparative Analysis of Causal Attributions in Reporting of the Israel-Palestine Conflict

“…Causality mining is the task of identifying causal language that connects events in a text, it is "an information extraction step that acquires causal relations from a collection of documents" [8]. Causal language can be expressed explicitly or implicitly.…”

“…It's downstream applications include biomedical discovery, emergency management, and financial narrative summaries. Causal mining used to be based on pattern matching by identifying causal connectors [8]. Advances in Machine Learning allowed for statistical pattern recognition models, such as SVMs and Bayesian models [23,24].…”

“…In this paper, we introduce a novel method to study news slant, by applying a Causal Extraction (CE) model to identify cause-effect pairs in news events. Causal relation extraction models classify and identify causal language in text [8]. We propose using causal relation extraction models to measure false cause attribution and missing cause in news media.…”

NLP as a Lens for Causal Analysis and Perception Mining to Infer Mental Health on Social Media

“…Such cause and effect relationships deduce discourse relations to examine reliability and hence, trustworthiness of decision making by AI models. Discourse Relations: There has been a recent surge in the use of discourse analysis, and its potential is demonstrated in a recent survey [36]. Discourse analysis determines the connectivity among different text segments to map causeand-effect relationships.…”

CALM-Bench: A Multi-task Benchmark for Evaluating Causality-Aware Language Models

“…Extracted causal relations are often useful for generating causal knowledge graphs (Heindorf et al, 2020) and developing causal knowledge representations (Sharp et al, 2016;Dalal et al, 2021a) which can be used to improve model performance in CQA tasks. CRI tasks have been studied extensively in the computational linguistics and NLP domain (Yang et al, 2022;Drury et al, 2022). Early methods relied on lexical triggers and linguistic cues (Khoo et al, 1998;Girju et al, 2007;Neeleman et al, 2012).…”