A Method of Modern Chinese Irony Detection

An-Ran Li, Chu Ren Huang

Research output: Chapter in book / Conference proceedingChapter in an edited book (as author)Academic researchpeer-review

3 Citations (Scopus)

Abstract

Irony is a kind of expressions whose literal meaning is the reversal of its real meaning. Although the understanding of ironies is considered as highly depend on the contextual information, there should be some cues in grammatical and semantical level. In this research, we try to find these linguistic cues by the observation of large scale corpora. We will find the frequently-used ironic constructions, then analyze the features and the generation mechanism of them. We notice that the intensity of ironic expressions relies on its immediacy of coercing the listener to experience the reversal. We conclude seven kinds of reversal in Chinese irony and summarize their formalized features. We also design an Irony Identification Procedure (IIP) to help us to detect ironies. In the future, we plan to classify the features and compare the efficiency of them by computational methods to get the quantized data, and then finally find an effective way to detect irony automatically.

Original languageEnglish
Title of host publicationFrom Minimal Contrast to Meaning Construct
Subtitle of host publicationCorpus-based, Near Synonym Driven Approaches to Chinese Lexical Semantics
EditorsQi Su, Weidong Zhan
PublisherSpringer Science and Business Media B.V.
Chapter19
Pages273-288
Number of pages16
ISBN (Electronic)978-981-32-9240-6
ISBN (Print)978-981-32-9239-0, 978-981-32-9242-0
DOIs
Publication statusPublished - 26 Sept 2019

Publication series

NameFrontiers in Chinese Linguistics
Volume9
ISSN (Print)2522-5308
ISSN (Electronic)2522-5316

Keywords

  • Construction
  • Irony detection
  • Irony identification procedure
  • Reversal theory

ASJC Scopus subject areas

  • Language and Linguistics

Fingerprint

Dive into the research topics of 'A Method of Modern Chinese Irony Detection'. Together they form a unique fingerprint.

Cite this