Размер шрифта: 

The importance of analyzing online news automatically gradually grows. In this paper we focus on the studying the features that define a corpus of online publications, like news, articles, inter-views, and blogs. We created a feature set on past work to recognize texts based on features approach, and analyzed them throw the series of experiments. We use cluster analysis to deter-mine the importance of each. Our results suggest that some features may represent basic infor-mation to distinguish news from news, articles, and interview.