I want to work under fake news detected(Networking). Naw I need recent raw data.
Mohit Tiwari
Publicly Available Datasets: FakeNewsNet: Integrates content, social context, and temporal information from news articles and Twitter. LIAR Dataset: Contains thousands of short statements with credibility labels from PolitiFact. BuzzFeed/PolitiFact Kaggle datasets: Often used in baseline fake news detection models. CoAID (COVID-19 related): Useful if you’re exploring misinformation in health contexts. Social Media APIs (Twitter/X, Facebook Graph API): You can scrape real-time data (headlines, comments, hashtags) using keywords, hashtags, and fact-checker responses. Combine with tools like Tweepy (Python) and cross-reference with verified fact-checking sites (e.g., Snopes, PolitiFact) to label. Manual Data Collection & Annotation: If you’re focusing on regional language or country-specific content, consider building your own dataset from online news portals and classifying articles manually or via semi-supervised learning techniques. Networking Angle: If your research includes detection via network behavior (propagation graphs, bot activity, etc.), look into datasets that provide user interaction graphs, such as the PHEME dataset.
Thank you for your question, Tausif. For research on fake news detection—especially in the context of networking or cybersecurity—you’ll need recent, labeled datasets that reflect current linguistic and platform-specific patterns.
Here are a few suggestions to collect or access raw data:
Do ensure that your data collection complies with ethical guidelines and platform-specific terms of service. You may also anonymize user data to align with research ethics norms.
Best wishes on your project—fake news detection is a pressing and impactful research area.
1 votes
1 thanks
More Tausif Mahamud Ruhan's questions See All
Similar questions and discussions