编号 2 (2022)
Intelligent Systems and Robots
Intelligent System for Predicting the Feasibility of Using Computed Tomography
摘要
The article describes principles of creating an intelligent system using JSM-method of automated research support (JSM-method ARS) to predict the necessity for computed tomography application. The procedures of JSM-research (one of the JSM-method ARS stages) designed to increase the reliability of the regularities obtained in the system are described. The obtained regularities and their expert ratings are given.



System, Evolutionary, Cognitive Modeling
System-Object Determinant Analysis. Partitive Classification Using the Formal-Semantic Normative System
摘要
The paper considers the construction of a partitive classification when conducting a systemobject determinant analysis using a formal-semantic normative system. The formal semantic alphabet and the rules for its use are described in terms of descriptive logic. The developed algorithms provide computer support for the decomposition of a complex system in graphical-analytical modeling. An illustrative example is given.



Analysis of Textual and Graphical Information
Methods for Cross-Lingual Retrieval of Similar Documents in Legal Domain Based on Machine Learning
摘要
The need of studying the international experience to improve legislation cause the need of information retrieval systems to be good in multilingual legal domain. One of the possible solutions is thematically similar document retrieval. However, there is an important task to transfer between languages to let the user put a document on the one language and get the search result on another one. The paper describes different approaches to solve this problem: from classical mediator-based methods to modern procedures of distributive semantics. As a test collection, we have used the UN digital library. The combination of the extended translation model and BM25 ranking function demonstrates the best results.



Representation of Syntactic Structures with Coordinating Conjunctions
摘要
The paper discusses sentences with coordinating conjunctions and homonymy where it is hard or impossible to build feasible syntactic structures using well-known models – dependency-based parse trees, constituency-based parse trees, and syntactic groups model. We suggest an approach to represent syntactic structures of sentences with conjunctions. We present features which distinguish our approach from the models under investigation. The paper shows multiple ways of visualization of syntactic structures.



Decision Analysis
Visual Decision Support for Curriculum Development Using the UGVA Method
摘要
The article deals with an approach to comparing, evaluating, and improving curricula. A model of parameterization of the curriculum with respect of key professional skills. For data concentration and analysis, we used the Unified Graphic Visualization of Activity (UGVA) method. We describe the results of an analysis of 36 curricula for the Russian academic major “system analysis and control”. The images created using the described method allowed us to summarize curricula evaluation data, identify the differences in teaching students and best practices. Comparing the selected curriculum with others in UGVA notation, we developed recommendations on changing the curriculum structure regarding courses developing key professional skills. General recommendations on the use of the method in decision-making are given.



Machine Learning, Neural Networks
Method for DeepFake Detection Using Convolutional Neural Networks
摘要
The article proposed the face anti-digital-spoofing countermeasures method for improving the protection of the facial biometric system. The DeepFake detection method is based on the convolutional neural networks, trained on a large dataset that contains different fake types with different qualities. This has resulted in at least 99% of detection quality. The suggested method can be used to increase the protection of facial biometric systems by reducing the risk of unauthorized access.



Conferences
XIX National Conference on Artificial Intelligence
摘要
The traditional Nineteenth National Conference on Artificial Intelligence with international participation (CAI-2021) was held in Taganrog, Russia, on October 11-16, 2021. The conference was coorganized by the Russian Association of Artificial Intelligence, Federal Research Center “Computer Science and Control” of the Russian Academy of Sciences, Moscow Physical and Technical Institute (National Research University), Southern Federal University. Conference co-chairs are Academician of the RAS S.N. Vasiliev (Institute of Control Sciences RAS, Moscow), Academician of the RAS I.A. Kalyaev (Southern Federal University, Rostov-on-Don), Academician of the RAS I.A. Sokolov (FRC “Computer Science and Control” RAS, Moscow). Various areas of artificial intelligence were presented in plenary reports and at section meetings.


