Search

The article considers an important scientific and applied task of identifying irrelevant and unreliable information on web resources, which is an important area of development and implementation of methods of data mining. The analysis of modern methods and means of estimation of irrelevant and unreliable information from the point of view of estimation of information sources is carried out and the basic problem directions which arise in the course of their functioning are allocated. A system of indicators for filtering unreliable and irrelevant information, which is obtained on the basis of several sources, is proposed. Based on this system, a method of checking information from web resources for relevance and reliability has been implemented. This approach is based on the possibility of using a predefined resource, the data from which are only reliable. A method of detecting inaccurate and irrelevant information has been developed, taking into account the peculiarities of its distribution through relevant pages in social networks and the use of multitasking classification of information obtained from various data sources. The use of the proposed intelligent data processing methods together with other methods of intellectual analysis used to evaluate information obtained from the Internet, will significantly increase the efficiency of the process of establishing irrelevance and inaccuracy of information, and will build an assessment of a particular web resource for publishing and disseminating such information. [ABSTRACT FROM AUTHOR]