Computer implementation of legal acts classification

Text classification is one of Data Mining function that assigns items in a collection to target categories or classes. The goal of classification is to accurately predict the target class for each case in the data. Realization of the Russian Federation’s Legal acts classification is presented in article. Legal acts are formed by legal drafting requirements, have a certain form and are different in comparing with usual texts, and generate hierarchical structure. Nowadays classification problem is not solved in legal reference systems. 16 main directions of Russian Federation government work as categories. Legal acts of Russian Federation that regulate relations in main directions of government working are used as texts which define categories. Task description looks like this: we have a set of categories; we have a set of texts for each category; we should define category for new text. Stages of implementation: text analysis, category definition and learning; text comparing with defined categories. Task realized by implementation in Java. MS SQL Server used as database. Classification problem is solved and realized successfully, as a result we have accurate tool for Legal acts classifying and defining executive branch of government by competence. Next purpose of the study is application rework for incoming correspondence classification in Federal antimonopoly service. Bibliogr. 9. Il. 4.