Guess whether a search engine snippet contains possibly criminal content.
The challenge's dataset consists of search engine snippets that have been identified by human annotators as potentially containing criminal content. Labels were assigned solely on the basis of text content.