AFAndrey M. Fedorov

Andrey M. Fedorov

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Andrey M. Fedorov has a total of 16 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • AUTOMATED HYPERTEXT DOCUMENTATION OF LEGACY SYSTEMS BASED ON STATIC ANALYSIS OF PROGRAM CODE REPOSITORIES AND GENERATIVE ARTIFICIAL INTELLIGENCE

    by Andrey M. Fedorov, Igor O. Datyev, Mikhail O. Ilyasov, Ivan G. Vishnyakov

    Automating software development and maintenance processes is a key factor in increasing the efficiency and speed of information system creation. As projects become increasingly complex and documentation quality requirements increase, tools that automate source code analysis and the generation of structured descriptions are becoming increasingly important. This article presents a solution for automated documentation of legacy code based on the integration of version control systems and a large language model. The developed pipeline sequentially processes source modules, extracts metadata for software entities (classes, methods, properties), generates descriptions using large language model when they are missing, and automatically publishes the results as wiki documentation in a collaborative development environment. The proposed approach reduces the labor costs of creating and maintaining documentation, ensures its constant synchronization with the code, and improves the quality of project documentation through artificial intelligent descriptions. The solution is designed for projects in various programming languages that require automatic documentation of legacy code, integration of descriptions into collaborative development systems, and accessibility of documentation via a web interface.

  • TECHNOLOGY FOR DEVELOPING INTELLIGENT INFORMATION SYSTEMS BASED ON THE COMMUNICATIVE CAPABILITIES OF LARGE LANGUAGE MODELS

    by Andrey M. Fedorov, Igor O. Datyev, Mikhail O. Ilyasov, Ivan G. Vishnyakov, Mark O. Bazegskiy, Daniil S. Figurkin, Christina D. Lyubimova

    Automating information technology development processes is a pressing issue. Current development trends require solutions to quickly adapt to new conditions, but traditional practices of classical development and modernization of existing solutions have a number of known shortcomings that limit the productivity and competitiveness of projects. This study aims to develop a technology for the rapid creation of information systems through the component integration of existing developments and intelligent assistants and agents. The proposed primary tool for such integration is the standard MCP protocol, which expands the communication capabilities of artificial intelligence technologies, models, and implementation tools. The proposed solution significantly improves the timeframe for conceptual design and development of applied and research information systems, while also increasing their intellectualization and mutual integration. The resulting technology is of interest to researchers and developers in the fields of systems analysis, management, and information processing.


  • Open Access

    The Method for Integrating Large Language Models into Algorithms for Focused Monitoring of Open Social Media Data

    by Andrey Fedorov, Igor Datyev, Ivan Vishnyakov

    The relevance of the study is determined by the importance and complexity of performing rapid summarization of a vast array of user-generated content on social networks. It is proposed to reduce the complexity of the problem by using robotic algorithms and their automated intelligent focusing on specific platforms, data availability, and data volumes. The paper examines the ability of large language models (LLMs) to generate high-quality, coherent, and context-sensitive annotations (summaries) that are suitable for the dynamic nature of unstructured, noisy social network data. The features of the RAG LLM technology for summarizing social network publications are presented. The main disadvantage of language models is instability and the difficulty of tracking the results to confirm factual accuracy. The authors propose a hybrid method for summarizing social media posts over a given period of time. The method involves a complex and variable combination of classical methods for extracting data from their repositories, as well as the abstractive and generative capabilities of large language models. Large language models are used to vectorize the analyzed data. The application of clustering algorithms to the obtained vector representations made it possible to increase the stability and quality of the results. Within the RAG technology, the capabilities of large language models are expanded by means of intelligent search in the MongoDB database used to store the original data. The paper presents three pipelines, each of which is a variant of the method implementation, and has advantages and disadvantages in various application conditions. The metrics used to evaluate the pipelines are given, and a comparative analysis is performed. Overall, the method allows us to reduce the confabulations of a large language model and obtain annotations of publications for different time periods in real-time. The proposed method is used in practice in the open social media data monitoring system developed by the authors.


  • USING RAG TECHNOLOGY TO DESIGN AN INTELLIGENT INFORMATION SYSTEM FOR SUPPORT EXPLORATORY SEARCH

    by Andrey Oleynik, Andrey Fedorov, Igor Datyev, Alexander Zuenko, Aleksey Shestakov, Ivan Vishnyakov

    This paper presents the integration of a large language model (LLM) into an intelligent information and analytical system for exploratory search support within specific texts. An overview of the capabilities of modern large language models is provided. Special attention is paid to the Retrieval Augmented Generation (RAG) technology used for efficient formation and control of search contexts. The design and development of the system's business logic functions and user interface elements are detailed. Experimental results applying the developed system to texts with specific characteristics are presented. Conclusions are drawn regarding the system's applicability for automating ontology population. Conclusions are drawn regarding the system's applicability for automating the process of populating ontologies, as well as filling other knowledge repositories built using automated monitoring of open big data and intelligent data processing.


  • Information and analytical system for regional development management support based on open big data of social media: development concept and implementation practice

    by A. M. Fedorov, Putilov Institute for Informatics and Mathematical Modeling of the Kola Science Centre of the Russian Academy of Sciences, I. O. Datyev, M. G. Shishaev, S. S. Fedotov, I. G. Vishnyakov

    The paper considers the possibilities of using big open data of online social networks for information and analytical support of regional development. The concepts of social objects and processes are introduced, which are associated with the socio-economic sphere of the region and reflected in the data presented in social networks. An overview of modern tools and technologies for analyzing data of social networks for solving state, municipal and business problems is given.The potential for expanding the functionality of the considered systems based on the use of formative artificial intelligence is noted. The experience of forming the necessary system infrastructure and developing components of a software framework for working with big data of social networks used for information and analytical support of regional development, is presented on the example of the Murmansk region.


  • Online Social Networks Analysis for Digitalization Evaluation

    by Andrey M. Fedorov, Igor O. Datyev, Andrey L. Shchur, Andrey G. Oleynik