In our research on disinformation analysis using machine learning, we created numerous models gauging individual manipulative aspects of news texts. Be it clickbait, hate speech, or political bias, the chances are you'll find a model for its analysis in our toolkit. As we realize these models' potential benefits for research in the fields of machine learning and data science, as well as social, journalism, and cognitive sciences, we decided to make them available to the research community free of charge.
> WHAT'S IN THE PACKAGE?
We'll provide you with the code of 5 machine learning models analyzing different news texts' attributes, along with their weights trained for the Czech language. They can be easily retrained for other languages as well, provided you have training data for the respective task.
> HOW DO YOU ENSURE SAFETY OF SUCH A PRACTICE?
We require you to fill in the application below, including prompts about your project, intended use, and funding. If approved, we'll issue you a license for using and editing our models under our ethical research conditions. If we discover any violations of the terms, we will revoke the license.
> RESOURCES
[Read Verifee Dataset research paper]
[Read Verifee Methodology]
[About Verifee Research Initiative & Resources]