<p>Text forums are an important source of information in the modern world. When studying new technology forums become the extremely valuable source of knowledge. They contain experience and responses of people which really used technology, its features, and also negative experience - that often it is impossible to receive from official documentation. However information at forums possesses a number of negative features: informational content of posts is uneven and not always belongs directly to a subject. Therefore for studying new technologies it can be useful to allocate part of messages of a forum, and to rank them on suitability. This work focuses on the method of processing text forum for further analysis. We created technique for marking up text forum and presenting it in convenient for applying NLP algorithms form. For an example, forum «Linux for beginners» was marked down.</p>