## Volume 10, No. 2, 2015

- Molodtsov D.A.

**«Soft topological structures»**

For each point in space, a finite number of neighborhoods are considered. For such construction an analogy of topological concepts are presented. These concepts allow to work without the assumption of infinite divisibility of space.Keywords: soft topological notions.

Resources: title page of the article.

Bibliographic citationMolodtsov D.A. Soft topological structures.*Nechetkie Sistemy i Myagkie Vychisleniya*[Fuzzy Systems and Soft Computing], 2015, vol. 10, no. 2, pp. 115-153. (in Russian) - Filchenkov A.A., Zolotin A.A., Tulupyev A.L.

**«Graph structures in relational databases, constraint satisfaction and Bayesian networks»**

The paper is devoted to the comparative analysis of systems of knowledge representation based on graph structures. Such systems include relational databases, constraint satisfaction problems, Bayesian belief networks and algebraic Bayesian networks. The article examines the application of the principle of decomposition for each of these systems. The given comparative analysis of the graph structuress shows that in acyclic case all these structures are equivalent whereas in general the requirements for a graph structure of algebraic Bayesian networks are more stringent than for the other three structures.Keywords: probabilistic graphical models, secondary structure, primary structure, knowledge with uncertainty, decomposition of the system, Bayesian networks, constraint satisfaction problems, relational database, join graphs.

Resources: title page of the article.

Bibliographic citationFilchenkov A.A., Zolotin A.A., Tulupyev A.L. Graph structures in relational databases, constraint satisfaction and Bayesian networks.*Nechetkie Sistemy i Myagkie Vychisleniya*[Fuzzy Systems and Soft Computing], 2015, vol. 10, no. 2, pp. 155-179. (in Russian) - Sorokina I.V., Sorokin S.V.

**«Parameter estimation of multidimensional possibilistic distributions for a given risk level»**

Based on the work of Dug Hun Hong, we present a solution to parameters estimation problem of fuzzy distributions in m-dimensional case and show that this estimator is consistent, sufficient and maximum likelihood. Our method is based on geometrical approach, where minimal area enclosing ellipsoid is constructed around the sample.Keywords: fuzzy distribution, parameter estimation.

Resources: title page of the article.

Bibliographic citationSorokina I.V., Sorokin S.V. Parameter estimation of multidimensional possibilistic distributions for a given risk level.*Nechetkie Sistemy i Myagkie Vychisleniya*[Fuzzy Systems and Soft Computing], 2015, vol. 10, no. 2, pp. 181-193. (in Russian) - Volkov A.V.

**«Method of calculating the coefficients of semantic proximity between scientific articles»**

The article describes the method of calculating the coefficients of semantic proximity between scientific articles based on the use of semantic analysis of the text elements, as well as analysis of the structural elements of a scientific article. The scientific article is presented as a fuzzy set whose elements are the terms where each term is assigned a grade of membership, which characterizes the degree of its importance for this document. The described technique offers a new way to calculate the weight coefficients of the terms in the text. It also introduces a metric measuring the distance between scientific articles, which takes into account implicit semantic links between the compared documents. To obtain semantic characteristics of the terms in the study Wiktionary is used as a thesaurus.Keywords: semantic analysis, a measure of proximity documents Wiktionary, the semantic net of a document, fuzzy sets.

Resources: title page of the article.

Bibliographic citationVolkov A.V. Method of calculating the coefficients of semantic proximity between scientific articles.*Nechetkie Sistemy i Myagkie Vychisleniya*[Fuzzy Systems and Soft Computing], 2015, vol. 10, no. 2, pp. 195-207. (in Russian) - Azarov A.A., Abramov M.V., Tulupyeva T.V., Filchenkov A.A.

**«The use of probabilistic relational models of complex "critical documents - information system - the user - the attacker" for the analysis of security of information systems users from social engineering attacks»**

We considered a problem of consistency diagnostics for socially significant behavior model based on the data about behavior episodes. We described the extention of the model and provided the examples of consistency estimates. We suggested possible improvements for this model.Keywords: diagnostics, consistency evaluation, socially significant behavior, last episodes, Bayesian belief networks.

Resources: title page of the article.

Bibliographic citationAzarov A.A., Abramov M.V., Tulupyeva T.V., Filchenkov A.A. The use of probabilistic relational models of complex "critical documents - information system - the user - the attacker" for the analysis of security of information systems users from social engineering attacks.*Nechetkie Sistemy i Myagkie Vychisleniya*[Fuzzy Systems and Soft Computing], 2015, vol. 10, no. 2, pp. 209-221. (in Russian)