Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Zhengjia, Wang, Danding, Sheng, Qiang, Cao, Juan, Ma, Siyuan, Cheng, Haonan
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2312.16490
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911007856656384
author Wang, Zhengjia
Wang, Danding
Sheng, Qiang
Cao, Juan
Ma, Siyuan
Cheng, Haonan
author_facet Wang, Zhengjia
Wang, Danding
Sheng, Qiang
Cao, Juan
Ma, Siyuan
Cheng, Haonan
contents Understanding the intent behind information is crucial. However, news as a medium of public discourse still lacks a structured investigation of perceived news intent and its application. To advance this field, this paper reviews interdisciplinary studies on intentional action and introduces a conceptual deconstruction-based news intent understanding framework (NINT). This framework identifies the components of intent, facilitating a structured representation of news intent and its applications. Building upon NINT, we contribute a new intent perception dataset. Moreover, we investigate the potential of intent assistance on news-related tasks, such as significant improvement (+2.2% macF1) in the task of fake news detection. We hope that our findings will provide valuable insights into action-based intent cognition and computational social science.
format Preprint
id arxiv_https___arxiv_org_abs_2312_16490
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Exploring news intent and its application: A theory-driven approach
Wang, Zhengjia
Wang, Danding
Sheng, Qiang
Cao, Juan
Ma, Siyuan
Cheng, Haonan
Computation and Language
Artificial Intelligence
Computers and Society
Understanding the intent behind information is crucial. However, news as a medium of public discourse still lacks a structured investigation of perceived news intent and its application. To advance this field, this paper reviews interdisciplinary studies on intentional action and introduces a conceptual deconstruction-based news intent understanding framework (NINT). This framework identifies the components of intent, facilitating a structured representation of news intent and its applications. Building upon NINT, we contribute a new intent perception dataset. Moreover, we investigate the potential of intent assistance on news-related tasks, such as significant improvement (+2.2% macF1) in the task of fake news detection. We hope that our findings will provide valuable insights into action-based intent cognition and computational social science.
title Exploring news intent and its application: A theory-driven approach
topic Computation and Language
Artificial Intelligence
Computers and Society
url https://arxiv.org/abs/2312.16490