Evaluating Web Content Using the W3C Credibility Signals

The credibility and trustworthiness of online content has become a major societal issue as human communication information exchange continues to evolve digitally. prevalence misinformation, circulated by fraudsters, trolls, political activists state-sponsored actors, motivated heightened interest in automated evaluation curation tools. We present an system aid users assessments web pages, focusing on the analysis 23 mostly language- content-related signals content. find that emotional characteristics, various morphological syntactical properties language, exclamation mark all caps usage are particularly indicative credibility. Less credible pages have more emotional, shorter less complex texts, put greater emphasis headline, which is longer, contains frequently clickbait. Our achieves 63% accuracy fake news classification, 28% predicting rating five-point Likert scale.