Sergey V. Kovalchuk

Yulong He, Nikita Verbin, Sergey Kovalchuk

Software repositories provide a detailed record of software evolution by capturing developer interactions through code-related activities such as pull requests and modifications. To better understand the underlying dynamics of codebase evolution, we introduce a novel approach that integrates semantic code embeddings with opinion dynamics theory, offering a quantitative framework to analyze collaborative development processes. Our approach begins by encoding code snippets into high-dimensional vector representations using state-of-the-art code embedding models, preserving both syntactic and semantic features. These embeddings are then processed using Principal Component Analysis (PCA) for dimensionality reduction, with data normalized to ensure comparability. We model temporal evolution using the Expressed-Private Opinion (EPO) model to derive trust matrices and track opinion trajectories across development cycles. These opinion trajectories reflect the underlying dynamics of consensus formation, influence propagation, and evolving alignment (or divergence) within developer communities -- revealing implicit collaboration patterns and knowledge-sharing mechanisms that are otherwise difficult to observe. By bridging software engineering and computational social science, our method provides a principled way to quantify software evolution, offering new insights into developer influence, consensus formation, and project sustainability. We evaluate our approach on data from three prominent open-source GitHub repositories, demonstrating its ability to reveal interpretable behavioral trends and variations in developer interactions. The results highlight the utility of our framework in improving open-source project maintenance through data-driven analysis of collaboration dynamics.

Yulong He, Ivan Smirnov, Dmitry Fedrushkov et al.

Effective evaluation of large language models (LLMs) remains a critical bottleneck, as conventional direct scoring often yields inconsistent and opaque judgments. In this work, we adapt the Analytic Hierarchy Process (AHP) to LLM-based evaluation and, more importantly, propose a confidence-aware Fuzzy AHP (FAHP) extension that models epistemic uncertainty via triangular fuzzy numbers modulated by LLM-generated confidence scores. Systematically validated on JudgeBench, our structured approach decomposes assessments into explicit criteria and incorporates uncertainty-aware aggregation, producing more calibrated judgments. Extensive experiments demonstrate that both crisp and fuzzy AHP consistently outperform direct scoring across model scales and dataset splits, with FAHP showing superior stability in uncertain comparison scenarios. Building on these insights, we propose \textbf{DualJudge}, a hybrid framework inspired by Dual-Process Theory that adaptively fuses holistic direct scores with structured AHP outputs via consistency-aware weighting. DualJudge achieves state-of-the-art performance, underscoring the complementary strengths of intuitive and deliberative evaluation paradigms. These results establish uncertainty-aware structured reasoning as a principled pathway toward more reliable LLM assessment. Code is available at https://github.com/hreyulog/AHP_llm_judge.

Liubov O. Elkhovskaya, Alexander D. Kshenin, Marina A. Balakhontceva et al.

Within Process mining, discovery techniques had made it possible to construct business process models automatically from event logs. However, results often do not achieve the balance between model complexity and its fitting accuracy, so there is a need for manual model adjusting. The paper presents an approach to process mining providing semi-automatic support to model optimization based on the combined assessment of the model complexity and fitness. To balance between the two ingredients, a model simplification approach is proposed, which essentially abstracts the raw model at the desired granularity. Additionally, we introduce a concept of meta-states, a cycle collapsing in the model, which can potentially simplify the model and interpret it. We aim to demonstrate the capabilities of the technological solution using three datasets from different applications in the healthcare domain. They are remote monitoring process for patients with arterial hypertension and workflows of healthcare workers during the COVID-19 pandemic. A case study also investigates the use of various complexity measures and different ways of solution application providing insights on better practices in improving interpretability and complexity/fitness balance in process models.

Alexey Gorbatovski, Sergey Kovalchuk

Within this study, we propose a new approach for natural language processing using Bayesian networks to predict and analyze the context and how this approach can be applied to the Community Question Answering domain. We discuss how Bayesian networks can detect semantic relationships and dependencies between entities, and this is connected to different score-based approaches of structure-learning. We compared the Bayesian networks with different score metrics, such as the BIC, BDeu, K2 and Chow-Liu trees. Our proposed approach out-performs the baseline model at the precision metric. We also discuss the influence of penalty terms on the structure of Bayesian networks and how they can be used to analyze the relationships between entities. In addition, we examine the visualization of directed acyclic graphs to analyze semantic relationships. The article further identifies issues with detecting certain semantic classes that are separated in the structure of directed acyclic graphs. Finally, we evaluate potential improvements for the Bayesian network approach.

Yulong He, Artem Ermakov, Sergey Kovalchuk et al.

ArkTS is a core programming language in the OpenHarmony ecosystem, yet research on ArkTS code intelligence is hindered by the lack of public datasets and evaluation benchmarks. This paper presents a large-scale ArkTS dataset constructed from open-source repositories, targeting code retrieval and code evaluation tasks. We design a single-search task, where natural language comments are used to retrieve corresponding ArkTS functions. ArkTS repositories are crawled from GitHub and Gitee, and comment-function pairs are extracted using tree-sitter-arkts, followed by cross-platform deduplication and statistical analysis of ArkTS function types. We further evaluate existing open-source code embedding models on the single-search task and perform fine-tuning using both ArkTS and TypeScript training datasets, resulting in a high-performing model for ArkTS code understanding. This work establishes the first systematic benchmark for ArkTS code retrieval. Both the dataset and our fine-tuned model are available at https://huggingface.co/hreyulog/embedinggemma_arkts and https://huggingface.co/datasets/hreyulog/arkts-code-docstring .

Oleg V. Kubryak, Sergey V. Kovalchuk, Nadezhda G. Bagdasaryan

The article proposes a universal dual-axis intelligent systems assessment scale. The scale considers the properties of intelligent systems within the environmental context, which develops over time. In contrast to the frequent consideration of the 'mind' of artificial intelligent systems on a scale from 'weak' to 'strong', we highlight the modulating influences of anticipatory ability on their 'brute force'. In addition, the complexity, the 'weight' of the cognitive task and the ability to critically assess it beforehand determine the actual set of cognitive tools, the use of which provides the best result in these conditions. In fact, the presence of 'common sense' options is what connects the ability to solve a problem with the correct use of such an ability itself. The degree of 'correctness' and 'adequacy' is determined by the combination of a suitable solution with the temporal characteristics of the event, phenomenon, object or subject under study.

Lijing Luo, Yiben Luo, Alexey Gorbatovski et al.

The remarkable progress of reinforcement learning (RL) is intrinsically tied to the environments used to train and evaluate artificial agents. Moving beyond traditional qualitative reviews, this work presents a large-scale, data-driven empirical investigation into the evolution of RL environments. By programmatically processing a massive corpus of academic literature and rigorously distilling over 2,000 core publications, we propose a quantitative methodology to map the transition from isolated physical simulations to generalist, language-driven foundation agents. Implementing a novel, multi-dimensional taxonomy, we systematically analyze benchmarks against diverse application domains and requisite cognitive capabilities. Our automated semantic and statistical analysis reveals a profound, data-verified paradigm shift: the bifurcation of the field into a "Semantic Prior" ecosystem dominated by Large Language Models (LLMs) and a "Domain-Specific Generalization" ecosystem. Furthermore, we characterize the "cognitive fingerprints" of these distinct domains to uncover the underlying mechanisms of cross-task synergy, multi-domain interference, and zero-shot generalization. Ultimately, this study offers a rigorous, quantitative roadmap for designing the next generation of Embodied Semantic Simulators, bridging the gap between continuous physical control and high-level logical reasoning.

Chao Li, Olga Petruchik, Elizaveta Grishanina et al.

This paper presents a Multi-Agent Norm Perception and Induction Learning Model aimed at facilitating the integration of autonomous agent systems into distributed healthcare environments through dynamic interaction processes. The nature of the medical norm system and its sharing channels necessitates distinct approaches for Multi-Agent Systems to learn two types of norms. Building on this foundation, the model enables agents to simultaneously learn descriptive norms, which capture collective tendencies, and prescriptive norms, which dictate ideal behaviors. Through parameterized mixed probability density models and practice-enhanced Markov games, the multi-agent system perceives descriptive norms in dynamic interactions and captures emergent prescriptive norms. We conducted experiments using a dataset from a neurological medical center spanning from 2016 to 2020.

Alexey Gorbatovski, Sergey Kovalchuk

In this study, we investigate the enhancement of the GPT Neo 125M performance in Community Question Answering (CQA) with a focus on programming, through the integration of Reinforcement Learning from Human Feedback (RLHF) and the utilization of scores from Stack Overflow. Two distinct reward model training strategies are employed for fine-tuning with Proximal Policy Optimization (PPO). Notably, the improvements in performance achieved through this method are comparable to those of GPT Neo 2.7B parameter variant. Additionally, an auxiliary scoring mechanism is introduced, which demonstrates the limitations of conventional linguistic metrics in evaluating responses in the programming domain. Through accurate analysis, this paper looks at the divergence between traditional linguistic metrics and our human-preferences-based reward model, underscoring the imperative for domain-specific evaluation methods. By elucidating the complexities involved in applying RLHF to programming CQA and accentuating the significance of context-aware evaluation, this study contributes to the ongoing efforts in refining Large Language Models through focused human feedback.

Tesfamariam M. Abuhay, Adane Mamuye, Stewart Robinson et al.

Patient flow analysis can be studied from a clinical and or operational perspective using simulation. Traditional statistical methods such as stochastic distribution methods have been used to construct patient flow simulation submodels such as patient inflow, Length of Stay (LoS), Cost of Treatment (CoT) and Clinical Pathway (CP) models. However, patient inflow demonstrates seasonality, trend and variation over time. LoS, CoT and CP are significantly determined by attributes of patients and clinical and laboratory test results. For this reason, patient flow simulation models constructed using traditional statistical methods are criticized for ignoring heterogeneity and their contribution to personalized and value based healthcare. On the other hand, machine learning methods have proven to be efficient to study and predict admission rate, LoS, CoT, and CP. This paper, hence, describes why coupling machine learning with patient flow simulation is important and proposes a conceptual architecture that shows how to integrate machine learning with patient flow simulation.

Sergey V. Kovalchuk, Georgy D. Kopanitsa, Ilia V. Derevitskii et al.

The paper presents an approach for building consistent and applicable clinical decision support systems (CDSSs) using a data-driven predictive model aimed at resolving the problem of low applicability and scalability of CDSSs in real-world applications. The approach is based on a threestage application of domain-specific and data-driven supportive procedures that are to be integrated into clinical business processes with higher trust and explainability of the prediction results and recommendations. Within the considered three stages, the regulatory policy, data-driven modes, and interpretation procedures are integrated to enable natural domain-specific interaction with decisionmakers with sequential narrowing of the intelligent decision support focus. The proposed methodology enables a higher level of automation, scalability, and semantic interpretability of CDSSs. The approach was implemented in software solutions and tested within a case study in T2DM prediction, enabling us to improve known clinical scales (such as FINDRISK) while keeping the problem-specific reasoning interface similar to existing applications. Such inheritance, together with the three-staged approach, provide higher compatibility of the solution and leads to trust, valid, and explainable application of data-driven solutions in real-world cases.

Ksenia Balabaeva, Anastasia Funkner, Sergey Kovalchuk

The main goal of this paper is to develop a spell checker module for clinical text in Russian. The described approach combines string distance measure algorithms with technics of machine learning embedding methods. Our overall precision is 0.86, lexical precision - 0.975 and error precision is 0.74. We develop spell checker as a part of medical text mining tool regarding the problems of misspelling, negation, experiencer and temporality detection.

Anastasia Funkner, Ksenia Balabaeva, Sergey Kovalchuk

Developing predictive modeling in medicine requires additional features from unstructured clinical texts. In Russia, there are no instruments for natural language processing to cope with problems of medical records. This paper is devoted to a module of negation detection. The corpus-free machine learning method is based on gradient boosting classifier is used to detect whether a disease is denied, not mentioned or presented in the text. The detector classifies negations for five diseases and shows average F-score from 0.81 to 0.93. The benefits of negation detection have been demonstrated by predicting the presence of surgery for patients with the acute coronary syndrome.

Anastasia A. Funkner, Aleksey N. Yakovlev, Sergey V. Kovalchuk

The paper proposes and investigates an approach for surrogate-assisted performance prediction of data-driven knowledge discovery algorithms. The approach is based on the identification of surrogate models for prediction of the target algorithm's quality and performance. The proposed approach was implemented and investigated as applied to an evolutionary algorithm for discovering clusters of interpretable clinical pathways in electronic health records of patients with acute coronary syndrome. Several clustering metrics and execution time were used as the target quality and performance metrics respectively. An analytical software prototype based on the proposed approach for the prediction of algorithm characteristics and feature analysis was developed to provide a more interpretable prediction of the target algorithm's performance and quality that can be further used for parameter tuning.

Tesfamariam M. Abuhay, Sergey V. Kovalchuk, Klavdiya O. Bochenina et al.

This paper presents results of topic modeling and network models of topics using the International Conference on Computational Science corpus, which contains domain-specific (computational science) papers over sixteen years (a total of 5695 papers). We discuss topical structures of International Conference on Computational Science, how these topics evolve over time in response to the topicality of various problems, technologies and methods, and how all these topics relate to one another. This analysis illustrates multidisciplinary research and collaborations among scientific communities, by constructing static and dynamic networks from the topic modeling results and the keywords of authors. The results of this study give insights about the past and future trends of core discussion topics in computational science. We used the Non-negative Matrix Factorization topic modeling algorithm to discover topics and labeled and grouped results hierarchically.

Pavel A. Smirnov, Sergey V. Kovalchuk, Alexander V. Boukhanovsky

The existing approaches for scientific workflows composition face the problems of domain knowledge integration. By this paper we summarize the results, which have been elaborated and implemented during the 2-year research concerning to Virtual Simulation Objects (VSO) concept and technology development. The contribution of this paper consists of formal models of the VSO internal structures and user-assistance logic, which may be obtained as a result of the reasoning over knowledge base. (This paper was rejected to appear in "Recent Advances in Knowledge Based Technologies and Applications" by Hindawi in 2014, but was stolen and published by Ke Han under the name "The Study on Workflows Composition Based on Virtual Simulation Objects Technology")

Sergey V. Kovalchuk, Pavel A. Smirnov, Konstantin V. Knyazkov et al.

Presented paper describes the development of comprehensive approach for knowledge processing within e-Sceince tasks. Considering the task solving within a simulation-driven approach a set of knowledge-based procedures for task definition and composite application processing can be identified. This procedures could be supported by the use of domain-specific knowledge being formalized and used for automation purpose. Within this work the developed conceptual and technological knowledge-based toolbox for complex multidisciplinary task solv-ing support is proposed. Using CLAVIRE cloud computing environment as a core platform a set of interconnected expressive technologies were developed.