Suspicious News Detection Using Micro Blog Text

TL;DR

This paper introduces a new task for detecting suspicious news using micro blog posts, providing a Japanese dataset and benchmark results to assist in reducing manual fact-checking efforts.

Contribution

It defines the suspicious news detection task using SNS posts, creates a Japanese dataset, and evaluates baseline machine learning models for this purpose.

Findings

Models can effectively identify suspicious news, reducing manual verification effort.

The Japanese dataset enables future research in suspicious news detection.

Baseline models show promising results in the new task.

Abstract

We present a new task, suspicious news detection using micro blog text. This task aims to support human experts to detect suspicious news articles to be verified, which is costly but a crucial step before verifying the truthfulness of the articles. Specifically, in this task, given a set of posts on SNS referring to a news article, the goal is to judge whether the article is to be verified or not. For this task, we create a publicly available dataset in Japanese and provide benchmark results by using several basic machine learning techniques. Experimental results show that our models can reduce the cost of manual fact-checking process.