# Alexander TulupyevSaint Petersburg State University | SPBU · Department of Computer Science

Alexander Tulupyev

PhD, DrSc

## About

120

Publications

3,510

Reads

**How we measure 'reads'**

A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more

302

Citations

Citations since 2017

## Publications

Publications (120)

This paper considers the problem of automating the previously presented scheme of classification of users’ posts in social media. Different variations of the embeddings obtained from the RuBERT and ELMo language models are used as input features of the classification. Theoretical significance lies in the adaptation of existing language models BERT...

A sharp deterioration of the patient’s condition against the backdrop of the development of life-threatening arrhythmias with symptoms of acute heart failure (AHF), multiple organ dysfunction syndrome (MODS) or cerebral edema (CE) can lead to the death of the patient. Since the known methods of automated diagnostics currently cannot accurately and...

Algebraic Bayesian networks and Bayesian belief networks are one of the probabilistic graphical models. One of the main tasks which need to be solved during the networks’ handling is the model structure training. This paper is dedicated to the automation of this process for algebraic Bayesian networks. This work relates to the PC-algorithm for alge...

The article presents methods for solving the problem of global a priori inference in algebraic Bayesian networks (ABN) for scalar and interval estimates in knowledge patterns of network. The practical significance is that the ability to solve these problems allows us to evaluate the probability of the truth of logical formulas based on the already...

The article presents the results of the study of the upper assessment of the sensitivity of solving the second problem of local posterior inference. The result is the found upper exponential estimate. The theoretical and practical significance of the study is to determine the permissible error in the data when working with algebraic Bayesian networ...

The article is aimed at summarizing the concepts of a derivative graph and a primitive graph for graphs with backbone connectivity. Theorems are formulated and proved on the main connectedness of the graph of the derivative and on the primitive graph of the main connected graphs. The theoretical and practical significance of the result is to simpli...

Algebraic Bayesian networks belong to the class of machine-learning probabilistic graphical models. One of the main tasks during researching machine learning models is the optimization of their time of work. This paper presents approaches to parallelizing algorithms for maintaining local consistency in algebraic Bayesian networks as one of the ways...

In this research we present the hybrid model of finding the most critical distribution trajectories of multipath Social engineering attacks, passing through which by the malefactor on a global basis has the topmost degree of probability and will bring the greatest loss to the company. The solution of search problem concerning the most critical traj...

The article is aimed at generalizing the concepts of a derivative graph and a primitive graph for graphs with trunk connectivity. Theorems are formulated and proved on the main connectedness of the graph of the derivative and on the graph of the antiderivative main connected graphs. The theoretical and practical significance lies in the study of st...

The purpose of this article is to propose an approach to denoting the parameters of the model for assessing the probability of success of a multi-pass social engineering attack of an attacker on a user. These parameters characterize the evaluation of the probability of propagation of social engineering attacks from user to user in one type of inter...

The article is aimed at demonstrating the practical application of graph theory as a subsection of the theoretical foundations of computer science in solving one of the interdisciplinary problems — describing the structure of the cadmium sulfide molecule using methods and indices of chemoinformatics. The article presents the results of calculations...

The material is devoted to solving an individual task of the research aimed at automating the construction of estimates of protection of users and, indirectly, critical documents from social engineering attacks. This issue is closely related to soft social computing. Also the estimates of protection of users are the basis of building a expert syste...

The task of model learning arises in algebraic Bayesian networks as one of the probabilistic graphical models. Several approaches to machine learning of algebraic Bayesian networks are known. This research is dedicated to the algorithm of machine learning of algebraic Bayesian network represented by a knowledge pattern on missing data. Besides this...

The article describes the approach to solving the problem of comparing user profiles of different social networks and identifying those that belong to one person. An appropriate method is proposed based on a comparison of the social environment and the values of account profile attributes in two different social networks. The results of applying va...

This paper presents approaches to parallelization of algorithms for maintaining external and internal consistency in algebraic Bayesian networks as one of the representatives of probabilistic graphical models. The algorithms modified based on these approaches are described and presented in the form of schemes.

In this article we propose to consider the trajectories of social engineering attacks, which are the most critical from the point of view of the expected damage to the organization, and not from the point of view of the probability of success of the defeat of the user and, indirectly, critical documents to which he has access. The article proposes...

Bayesian Belief Networks (BBN) provide a comprehensible framework for representing complex systems that allows including expert knowledge and statistical data simultaneously. We explored BBN models for estimating risky behavior rate and compared several network structures, both expert-based and data-based. To learn and evaluate models we used gener...

Introduction: Social engineering attacks can be divided into two types: direct (one-way) and multi-pass ones, passing through a chain of users. Normally, there are several propagation paths for a multi-pass social engineering attack between two users. Estimates of the probabilities of an attack to spread along different trajectories will differ. Pu...

An approach to the sensitivity analysis of local a posterior inference equations in algebraic Bayesian networks is proposed in the paper. Performed a sensitivity analysis of first a posterior inference task for stochastic and deterministic evidences propagated into the knowledge pattern with scalar estimates. For each of the considered cases the ne...

Introduction: Studies in sociology, psychology, epidemiology, marketing or information security often face the issue of estimating the behavior rate (either individually or at the level of a population). Direct methods of behavior rate estimation are sometimes not available; hence, it is important to develop indirect methods. Earlier studies propos...

An approach to the sensitivity analysis of local a posteriori inference equations in algebraic Bayesian networks is proposed in this paper. Some basic definitions and formulations are briefly given and the development of the matrix-vector a posteriori inference approach is considered. Some cases of the propagation of deterministic and stochastic ev...

Subject of Research. Algebraic Bayesian networks are referred to a class of probabilistic graphical models that are a representation of knowledge bases with uncertainty. The distinguishing feature of ABN is the availability of global structures. Among them there are primary and secondary structures that are directly used in various kinds of probabi...

Questions of information security for today’s business, somehow using information technologies are among the most important. Probabilistic relational model complex “information system—critical documents—the staff—the attacker” are being concidered, as well as an algorithm for estimating the probability of users of information systems security by so...

The present research deals with the possibility of social network users value orientations reconstruction based on the information contained on the their personal pages. In the first part of the research 126 participants were included. They were students with the average age of 22, 39 % men and 61 % women. During the second part of the research mor...

Great attention of specialists information security given to the protection of software and hardware components of the information system, while users of the information system has been neglected and may violate the confidentiality of corporate data. The article considers the addition of the complex “information system—personnel—critical documents”...

An improved algorithm for the synthesis of the secondary structure of algebraic Bayesian networks represented by a minimal join graph is proposed in the paper. The algorithm differs from the previously offered one so that it relies on the incremental principle, uses specially selected edges and, finally, eliminates redundant edges by a greedy algor...

The paper considers algorithms for global structures generation in algebraic Bayesian networks. A decremental algorithm for constructing a secondary structure after deleting vertex from the adjacency graph is proposed supplemented by a listing of the algorithm code and by the proof of its correctness. The results of the statistical tests for decrem...

The paper considers a C# software package that implements algorithms of the local probabilistic-logical inference in algebraic Bayesian networks and the synthesis of their secondary structure. The package supports local network consistency verification and maintenance, priori and posteriori inference operations. In addition, two assembly algorithms...

The paper focuses on features of Internet support of professional activities in modern university educators and of learning activities in students through the organization of a new web-based space for communication between instructors and learners. It describes the authors’ experience in creating such space basing on social networking within the co...

Prior to 2010, medical care for people living with HIV/AIDS was provided at an outpatient facility near the center of St. Petersburg. Since then, HIV specialty clinics have been established in more outlying regions of the city. The study examined the effect of this decentralization of HIV care on patients' satisfaction with care in clinics of St. P...

A posteriori inference is one of three kinds of inference that underlie the processing of knowledge patterns with probabilistic uncertainty in intelligent decision making systems using algebraic Bayesian networks (ABNs). In this paper, the key terms and formulations of theorems describing local a posteriori inference in algebraic Bayesian networks...

Posteriori inference is one of the three kinds of probabilistic-logic inferences in the probabilistic graphical models theory and the base for processing of knowledge patterns with probabilistic uncertainty using Bayesian networks. The paper deals with a task of local posteriori inference description in algebraic Bayesian networks that represent a...

We consider a task of local posteriori inference description by means of matrix-vector equations in algebraical Bayesian networks that represent a class of probabilistic graphical models. Such equations were generally presented in previous publications, however containing normalizing factors that were provided with algorithmic descriptions of their...

We offer a task statement on estimating HIV-acquisition risk. Our method is based on using data concerning several last episodes of risky behavior. We describe a mathematical model of acquiring infectious diseases that allows to correctly set up a mathematical task of considering last episodes data. We offer several approaches to solving this task.

Two-stage scheme for minimal join graph subsets synthesis that involves the construction of three sets (stereoseparators, their possessions and necessary edges) over given subalphabet set with subsequent construction a set of a certain kind of sinews for each stereoseparator. Systematized algorithms, implementing both stages, are systematized and t...

The paper introduces a new line of research — intelligent search of predictors of social events (Predictor Mining); this direction lies in the intersection of Data Mining and Social Computing areas of research. The proposed approach includes an automated composition of a predictors set for the social event and then interpretation of the predictors...

The purpose of this paper is twofold. On the one hand, we will provide model for the meme transmission from its bearer to a member of his virtual friends set and meme spreading in social networks in the context of socio-humanitarian views on the nature of the internet memes. On the other hand, we will highlight issues of generation, circulation and...

In the field of information security it is necessary to develop scientific and proved and mathematical methods and the models reflecting specifics of subject domain for ensuring activity of experts, allowing to automate the analysis of information systems user’s security from socio-engineering attacks. The purpose of this paper is consideration of...

Due to the inability of some algorithms for global logical and probabilistic inference on cyclic secondary structure of algebraic Bayesian network (ABN) and a relatively large time complexity of the algorithm for constructing such a structure it’s appropriate to present a criterion. Which allows checking if the ABN is cycling before building the se...

The problem of algebraic Bayesian network's (ABN) secondary structure construction based on known primary structure was outlined in the ABN's theory. Logic-probabilistic inference can be performed only when secondary structure is a join graph. The algorithm for randomized synthesis of minimal join graph is formulated in the paper. The theorem was p...

This paper studies associations between the last alcohol consumption (that can be used as an indicator of a risky behavior rate) and socio-demographical and psychological characteristics of respondents. We construct the appropriate linear regression model with the interval between the last alcohol consumption and an interview as a response variable...

Algebraic Bayesian Network (ABN) is one of the logical and probabilistic graphical models of bases of knowledge patterns uncertainty. Algorithms for global logical and probabilistic inference in ABN can be applied only under the condition of acyclicity of its secondary structure — join graph. The existing method for join graph transformation into j...

The paper offers a method for respondents' behavior modeling based on data about last episodes adjacent to interview and proposes improved techniques of modeling and processing of initial data uncertainty based on hybrid probabilistic and fuzzy approaches. This paper provides analytical (including asymptotic) analysis of these estimates and numeric...

Recalculation of an existing truth probability estimates given the probability of incoming evidence constutes the second problem of posterior inference in algebraic Bayesian networks. We consider the analysis of non-linear optimization problem arising from the atomic stochastic evidence propagation in the knowledge pattern with interval estimations...

The prototype of the program complex, used for demonstration of basic possibility for estimation the protection of personnel of information system from socioengineering attacks on the base of generalized approach, focused on analysis of trees of attacks, is described. The representation of information system and its personnel in the specified progr...

Algebraic Bayesian network (ABN) tertiary structure is required to both a random minimal join graph and the minimal join graph set. In addition, it is required to find the best or an optimal secondary structure over the given ABN primary structure. The goal of the work is to create a clear definition of the ABN tertiary structure and associated obj...

Algebraic Bayesian networks (ABN) are probabilistic-logical graphical models of knowledge systems with uncertainty. ABN probabilistic logical inference algorithms pro-cessing considerably depends on its secondary structure, which is usually represented as a join graph. In particular, the graphs cycles prevent the possibility of the mentioned algori...

There exist algorithms for determining the possibility of building an acyclic secondary structure of the network for its primary structure, and hence the possibility for effective carrying out of the posterior inference in the theory of algebraic Bayesian networks. Their presence allowed the introduction of a global posteriori algorithm for probabi...

We consider a problem of socially significant behavior rate estimates in terms of probabilistic graphical models. Such formal description of the problem allows applying powerful methods and developed algorithms of the theory of Bayesian belief networks. We can use existing software to make computational simulations and apply the model to solve prac...

The condition for algebraic Bayesian networks (ABN) global logical-probabilistic inference algorithms performance is the absence of cycles in its secondary structure. The primary structure, on which an acyclic secondary structure can be synthesized is called an acyclic one. The goal of the work is to propose an algorithm to detect primary structure...

Complex corporate information systems are widely distributed in the modern world. Development, support and protection of similar systems require a significant amount of time and resources, besides only highly skilled experts can be engaged in the management of such systems. Information stored in such information systems has enormous value for the o...

Article is devoted to the results of sociological research, held by the SPIIRAN laboratory's employees. This research was devoted to the revealing interrelations between psychological features, vulnerabilities and possible actions of information system's users, who are under the threat of socio-engineering attacks. Structuring the user's vulnerabil...

Algebraic Bayesian networks (ABN) belong to a class of logical and probabilistic graphical models of systems of knowledge with uncertainty. ABN allows to use interval probability estimates to represent uncertainty in knowledge. One of the most important conditions for ABN performance capability is the absence of cycles in its secondary structure. T...

We describe a technique that improves the beta-prime models fitted for the last episodes of risky behavior. Regression models show interconnections between rate parameters and respondents’ demographic and psychological trades. We examine these models using such criteria as jackknife and test of overdispersion. Also we develop a method for uncertain...

The paper is devoted to the theoretical consideration of psychological traits underlying user’s vulnerability under the threat of socio-engineering attack, their properties and features. For the convenience of the subsequent studying of socio-engineering attacks and their prevention a detailed classification of psychological traits on various param...

We develop a technique for quantitative estimates of respondents’ behavior that uses respondents’ answers about the time interval since the last episode. The paper provides the block of questions and formalized set of answers to be used in a questionnaire as well as the mathematical approach for data processing and making the estimates. The respond...

The prototype of the program complex, used for demonstration of basic possibility for estimation the protection of personnel of informative system from socioengineering attack on the base of generalized approach, focused on analyze of trees of attacks, is described. The representation of informative system and its personnel in the specified program...

The paper offers an improved method for respondents' behavior parameters estimates based on data about several last episodes. We try to avoid an assumption (which often is not correct) that the next behavior episode happens at the same time as the moment of interview does. The paper outlines some disadvantages of previous method and proposes modeli...

Developed description of informative models of the component of complex “in-formative system – personnel”, which is under threat of socioengineering attack is being presented in this paper. Informative model of user, users group, controlling areas, information objects (system of documents), hardware-software maintenance and information system itsel...

Algebraic Bayesian networks (ABN) are probabilistic-logic graphic models of knowledge systems with uncertainty and gives an advantage to deal with interval probability estimates. Secondary structure usually represented as an join graph is essential for ABN work. The article analyses edges of various minimal join graph cliques to specify different c...

The paper generalizes the transformation of a directed cycle in Bayesian belief networks (BBN) with binary random variables into a knowledge patterns chain in algebraical Bayesian networks (ABN) for the case of multivariate random variables. Under the assumption that multivariate random variables are represented with binary random variables conjunc...

The paper describes an algorithm for synthesis of the secondary structure of an algebraic Bayesian network (ABN) by its first structure. The second structure of ABN is a joint graph with minimal number of edges. A proof of algorithm’s correctness is given.

The paper states the problem of social engineering attacks, outlines their classification, pre-sents a possible approach to estimation of an index of security of System from the point of view of the human factor. The suggested approach is based on adaptation of the methods applied to the analysis of security of hardware-software complex of computer...

In article the approach to an estimation of intensity of the behavior which mathematical model is Poisson the casual process modeling this behavior as a series of episodes is described. The intensity estimation is formed in the conditions of deficiency of the information, namely on the basis of data on a small number of last episodes, and these dat...

One of the key analytic problems in behavioral epidemiology research is related to determining most important factors among those that influence the behavior of interest. The present paper offers a brief review of the method designed to estimate relative importance of predictor variables described by J. W. Johnson. Examples are discussed that illus...

The goal of this work is to generalize structure analysis of minimal join graphs representing algebraic Bayesian network secondary structure on common join graphs representing the same structure. A term system spreading the existing one with common join graphs is designed. New join graph properties are researched. Two lemmas characterizing an homag...

The paper proposes a way to estimate risky behavior rates ratio in two groups. These groups can be either the same group but considered a priori and a posteriori a behavioral intervention, or two groups/two samples from different populations to be compared simultaneously. A similar estimation is proposed to cumulative risk ratio (odds ratio). The r...

Algebraic Bayesian Networks (ABN) is a probabilistic-logic model for knowledge patterns bases with uncertainty. A knowledge pattern mathematical model is a conjuncts ideal with their estimates of probability. The estimates can be scalar as well as interval. An ABN consists of a set of knowledge patterns; this set is referred to as the primary struc...

A conjunct ideal with probabilistic estimates of its elements is considered as a knowledge pattern (KP) mathematical model. A set of KPs can be assembled in an algebraical Bayesian network (ABN). We analyze local and global structures of ABNs in order to represent KPs and ABNs as objectoriented data structures in software applications. Our primary...

We consider a problem of risky behavior intensity estimation based on data received from respondent’s answers about their behavior. The method of maximum likelihood estimation is used for derivation of the estimate, and likelihood function describes the likelihood of realization of the particular system of answers. The likelihood function is derive...

One of the main problems of researches in the field of socio-engineering attacks is the development of analysis algorithms (assessment) of information system's users' protection estimated on the basis of computing complexity. According to preliminary estimates the application of probabilistic relational algorithm will essentially reduce computing c...

Probabilistic graphical models including hidden Markov models and Bayesian networks are widespread in process modeling in such fields as speech recognition, information theory, machine translation and molecular biology. The goal of this work is toresearch of mutual relations between hidden Markov models and algebraic Bayesian networks. An algorithm...

In the theory of algebraic Bayesian networks, there are four operations classified as a kind of local synthesis of consistent truth estimates: knowledge pattern consistency verification, knowledge pattern reconciliation, a posteriori inference, and knowledge pattern enclosing reconciliation. The paper presents a knowledge pattern model formalizatio...

Hidden Markov models (HMM) and algebraic Bayesian networks (ABN) are proba-bilistic graphical models and because of that they are quit similar. HMM has wide application while ABN are not so widespread, but its instruments allow to simulate and solve hidden Markov models problems. The goal of this work is to solve hidden Markov model first problem b...

## Network

Cited

Cited By