Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

FakingRecipe: Detecting Fake News on Short Video Platforms from the Perspective of Creative Process

Yuyan Bu, Qiang Sheng, Juan Cao, Peng Qi, Danding Wang, Jintao Li

TL;DR

This work tackles fake news on short video platforms by shifting focus from what is presented to how it is produced. It introduces FakingRecipe, a dual-branch, creative process-aware detector capturing material selection and editing cues, and validates it on two datasets (FakeSV and the newly created FakeTT) with state-of-the-art performance. Key contributions include the MSAM and MEAM modules, a Hierarchical Temporal Structure Extractor for temporal editing, and the FakeTT English dataset, enabling cross-language evaluation and robust detection. The results demonstrate the practical impact of modeling production processes for misinformation detection, with implications for real-time moderation and investigative journalism.

Abstract

As short-form video-sharing platforms become a significant channel for news consumption, fake news in short videos has emerged as a serious threat in the online information ecosystem, making developing detection methods for this new scenario an urgent need. Compared with that in text and image formats, fake news on short video platforms contains rich but heterogeneous information in various modalities, posing a challenge to effective feature utilization. Unlike existing works mostly focusing on analyzing what is presented, we introduce a novel perspective that considers how it might be created. Through the lens of the creative process behind news video production, our empirical analysis uncovers the unique characteristics of fake news videos in material selection and editing. Based on the obtained insights, we design FakingRecipe, a creative process-aware model for detecting fake news short videos. It captures the fake news preferences in material selection from sentimental and semantic aspects and considers the traits of material editing from spatial and temporal aspects. To improve evaluation comprehensiveness, we first construct FakeTT, an English dataset for this task, and conduct experiments on both FakeTT and the existing Chinese FakeSV dataset. The results show FakingRecipe's superiority in detecting fake news on short video platforms.

FakingRecipe: Detecting Fake News on Short Video Platforms from the Perspective of Creative Process

TL;DR

This work tackles fake news on short video platforms by shifting focus from what is presented to how it is produced. It introduces FakingRecipe, a dual-branch, creative process-aware detector capturing material selection and editing cues, and validates it on two datasets (FakeSV and the newly created FakeTT) with state-of-the-art performance. Key contributions include the MSAM and MEAM modules, a Hierarchical Temporal Structure Extractor for temporal editing, and the FakeTT English dataset, enabling cross-language evaluation and robust detection. The results demonstrate the practical impact of modeling production processes for misinformation detection, with implications for real-time moderation and investigative journalism.

Abstract

As short-form video-sharing platforms become a significant channel for news consumption, fake news in short videos has emerged as a serious threat in the online information ecosystem, making developing detection methods for this new scenario an urgent need. Compared with that in text and image formats, fake news on short video platforms contains rich but heterogeneous information in various modalities, posing a challenge to effective feature utilization. Unlike existing works mostly focusing on analyzing what is presented, we introduce a novel perspective that considers how it might be created. Through the lens of the creative process behind news video production, our empirical analysis uncovers the unique characteristics of fake news videos in material selection and editing. Based on the obtained insights, we design FakingRecipe, a creative process-aware model for detecting fake news short videos. It captures the fake news preferences in material selection from sentimental and semantic aspects and considers the traits of material editing from spatial and temporal aspects. To improve evaluation comprehensiveness, we first construct FakeTT, an English dataset for this task, and conduct experiments on both FakeTT and the existing Chinese FakeSV dataset. The results show FakingRecipe's superiority in detecting fake news on short video platforms.
Paper Structure (34 sections, 13 equations, 15 figures, 5 tables)

This paper contains 34 sections, 13 equations, 15 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Empirical Analysis
  3. Phase I : Material Selection
  4. Phase II : Material Editing
  5. Method
  6. Overview
  7. Material Selection-Aware Modeling (MSAM)
  8. Material Editing-Aware Modeling (MEAM)
  9. Predication and Optimization
  10. Prediction
  11. Optimization
  12. Experiments
  13. Datasets
  14. Experimental Setup
  15. Overall Performance
  16. ...and 19 more sections

Figures (15)

  • Figure 1: A fake news video about residents hanging national flags amid the COVID-19 pandemic in China, exhibited along with the speculated creative process. The text was translated into English.
  • Figure 2: Sentiment analysis of audio material.
  • Figure 3: JS divergence between textual and visual materials.
  • Figure 4: Color richness of on-screen text.
  • Figure 5: On-screen Text Dynamics.
  • ...and 10 more figures