?
Модификации EM-алгоритма для вероятностного тематического моделирования
Машинное обучение и анализ данных. 2013. Т. 1. № 6. С. 657–686.
Vorontsov K. V., Potapenko A.
Probabilistic topic models discover a low-dimensional interpretable representation of text corpora by estimating a multinomial distribution over topics for each document and a multinomial distribution over terms for each topic. A uni ed family of expectation-maximization (EM) like algorithms with smoothing, sampling, sparsing, and robustness heuristics that can be used in any combinations is considered. The known models PLSA (probabilistic latent semantic analysis), LDA (latent Dirichlet allocation), SWB (special words with background), as well as new ones can be considered as special cases of the presented broad family of models. A new simple robust algorithm suitable for sparse models that do not require to estimate and store a big matrix of noise parameters is proposed. The present authors nd experimentally optimal combinations of heuristics with sparsing strategies and discover that sparse robust model without Dirichlet smoothing performs very well and gives more than 99% of zeros in multinomial distributions without loss of perplexity.
Shabanov N., Kuricheva O., Kurbatova J. et al., / Series Working Papers SSRN "Department of Economics Ca’ Foscari University of Venice". 2026.
The carbon balance of an ecosystem is the difference between Gross Primary Productivity (GPP) and Ecosystem Respiration (Reco) as expressed by Net Ecosystem Exchange (NEE). While remote sensing retrievals of GPP have reached maturity, Reco estimation remains underexplored and ultimately cast bias on NEE. Here we present an end-to-end multi-scale analysis of the mechanism of ...
Added: August 21, 2026
Ponomarenko A., / Series Computer Science "arxiv.org". 2025.
This paper addresses the challenge of merging hierarchical navigable small world (HNSW) graphs, a critical operation for distributed systems, incremental indexing, and database compaction. We propose three algorithms for this task: Naive Graph Merge (NGM), Intra Graph Traversal Merge (IGTM), and Cross Graph Traversal Merge (CGTM). These algorithms differ in their approach to vertex selection ...
Added: July 30, 2026
Piontkovski D., / Series arXiv "math". 2026.
A noncommutative projective variety is defined, following Artin and Zhang, by a graded coherent algebra 𝐴. The category of coherent sheaves is then the quotient qgr(𝐴) of the category of finitely presented graded modules by the subcategory of torsion modules. We consider the categorical and polynomial entropies of the Serre twist, that is, of the ...
Added: June 23, 2026
Piontkovski D., / Series arXiv "math". 2025.
If a symmetric multilinear algebra is weakly nil, then it is Engel. This result may be regarded as an infinite-dimensional analogue of the well-known Jacobian theorem, which states that if a polynomial mapping has a polynomial inverse, then its Jacobian matrix is invertible. This refines a theorem of Gerstenhaber and partially answers a question posed ...
Added: June 23, 2026
Shipilov F., Barnyakov A., Ivanov A. et al., / Series Physics "arxiv.org". 2026.
A fast simulation of the detector response is a vital task in high-energy physics (HEP). Traditional Monte-Carlo methods form the backbone of modern particle physics simulation software but are computationally expensive. We present a machine-learning-based approach to fast simulation of the Focusing Aerogel Ring Imaging Cherenkov (FARICH) detector response. Given a particle track and momentum, ...
Added: May 19, 2026
Derkacheva A., Sakirkina M., Kraev G. et al., /. 2026.
Comprehensive data on natural hazards and their consequences are crucial for effective for risk assessment, adaptation planning, and emergency response. However, many countries face challenges with fragmented, inconsistent, and inaccessible data, particularly regarding local-scale events. To address this data gap in Russia, we developed an end-to-end processing pipeline that scrapes news from various online sources, ...
Added: April 28, 2026
Pilé I., Deng Y., Shchur L., / Series arXiv "math". 2026. No. 2604.10254.
We investigate the spatial overlap of successive spin configurations in Markov chain Monte Carlo simulations using the local Metropolis algorithm and the Svendsen-Wang and Wolff cluster algorithms. We examine the dynamics of these algorithms for two models in different universality classes: the Ising model and the Potts model with three components. The overlap of two ...
Added: April 20, 2026
Gabdullin N., Androsov I., / Series Computer Science "arxiv.org". 2026.
Label prediction in neural networks (NNs) has O(n) complexity proportional to the number of classes. This holds true for classification using fully connected layers and cosine similarity with some set of class prototypes. In this paper we show that if NN latent space (LS) geometry is known and possesses specific properties, label prediction complexity can ...
Added: April 2, 2026
Sorokin K., Beketov M., Онучин А. et al., / arxiv.org. Серия cs.SI "Social and Information Networks ". 2025.
Community detection in complex networks is a fundamental problem, open to new approaches in various scientific settings. We introduce a novel community detection method, based on Ricci flow on graphs. Our technique iteratively updates edge weights (their metric lengths) according to their (combinatorial) Foster version of Ricci curvature computed from effective resistance distance between the ...
Added: January 15, 2026
Petrovanov I., Sergeev A., / Series Computer Science "arxiv.org". 2025. No. 2512.18332.
Transport coding reduces message delay in packet-switched networks by introducing controlled redundancy at the transport layer: original packets are encoded into coded packets, and the message is reconstructed after the first successful deliveries, effectively shifting latency from the maximum packet delay to the -th order statistic. We present a concise, reproducible discrete-event implementation of transport coding in OMNeT++, including ...
Added: December 24, 2025
Hessian-based lightweight neural network for brain vessel segmentation on a minimal training dataset
Меньшиков И. А., Бернадотт А. К., Elvimov N. S., / Series arXie "Statistical mechanics". 2025.
Accurate segmentation of blood vessels in brain magnetic resonance angiography (MRA) is essential for successful surgical procedures, such as aneurysm repair or bypass surgery. Currently, annotation is primarily performed through manual segmentation or classical methods, such as the Frangi filter, which often lack sufficient accuracy. Neural networks have emerged as powerful tools for medical image ...
Added: December 1, 2025
Gorshkov S., Ilyushin E., Chernysheva A. et al., International Journal of Open Information Technologies 2021 Vol. 9 No. 5 P. 12–17
Topic modeling is one of the most widely used methods in text analysis. It can be used to select topics as well as to find the topics distributed in each document from the corpus. In this article, we present a method for clustering communities in the social network VKontakte (the most popular Russian social network) ...
Added: December 25, 2024
Malik M. S., Nawaz A., Jamjoom M. M. et al., Intelligent Data Analysis 2024 Vol. 28 No. 4 P. 1045–1065
Online product reviews (OPR) are a commonly used medium for consumers to communicate their experiences with products during online shopping. Previous studies have investigated the helpfulness of OPRs using frequency-based, linguistic, meta-data, readability, and reviewer attributes. In this study, we explored the impact of robust contextual word embeddings, topic, and language models in predicting the ...
Added: February 26, 2024
Милкова М. А., Экономическая наука современной России 2020 № 1 С. 143–157
Nowadays the process of information accumulation is so rapid that the concept of the usual iterative search requires revision. Being in the world of oversaturated information in order to comprehensively cover and analyze the problem under study, it is necessary to make high demands on the search methods. An innovative approach to search should flexibly ...
Added: June 29, 2023
Badryzlova Y., Nikiforova A., Lyashevskaya O., , in: Analysis of Images, Social Networks and Texts: 9th International Conference, AIST 2020, Skolkovo, Moscow, Russia, October 15–16, 2020, Revised Selected PapersVol. 12602.: Springer, 2021. P. 69–81.
The paper examines the efficiency of topic models as features for computational identification and conceptual analysis of linguistic metaphor on Russian data. We train topic models using three algorithms (LDA and ARTM – sparse and dense) and evaluate their quality. We compute topic vectors for sentences of a metaphor-annotated Russian corpus and train several classifiers ...
Added: October 7, 2020
Topic Modelling of the Russian Corpus of Pikabu Posts: Author-Topic Distribution and Topic Labelling
Mitrofanova O., Sampetova V., Mamaev I. et al., , in: Information Society: Education, Science, Culture, and Modern Technologies. (Proceedings of XXIII International Joint Conference «Internet and Modern Society», IMS-2020, St. Petersburg, 17 - 20 June 2020).: CEUR Workshop Proceedings, 2020. P. 101–116.
The paper discusses development of a corpus of Russian posts with hash tags based on Pikabu social network. We developed a balanced and representative corpus as regards the impact of certain authors, the amount and size of their posts. Our study is aimed at the development of probabilistic topic models revealing the authors’ interests and ...
Added: September 30, 2020
Koltcov Sergei, Ignatenko V., Entropy 2020 Vol. 22 No. 5 P. 1–23
In practice, to build a machine learning model of big data, one needs to tune model parameters. The process of parameter tuning involves extremely time-consuming and computationally expensive grid search. However, the theory of statistical physics provides techniques allowing us to optimize this process. The paper shows that a function of the output of topic ...
Added: May 18, 2020
Koltsov S., Ignatenko V., Boukhers Z. et al., Entropy 2020 Vol. 22 No. 4 P. 1–13
Topic modeling is a popular technique for clustering large collections of text documents. A variety of different types of regularization is implemented in topic modeling. In this paper, we propose a novel approach for analyzing the influence of different regularization types on results of topic modeling. Based on Renyi entropy, this approach is inspired by ...
Added: April 1, 2020
Koltsov S., Ignatenko V., , in: Intelligent Computing: SAI 2020: Volume 1* 1. Vol. 1228.: Switzerland: Springer, 2020. P. 234–247.
Topic modeling is a widely used approach for clustering text documents, however, it possesses a set of parameters that must be determined by a user, for example, the number of topics. In this paper, we propose a novel approach for fast approximation of the optimal topic number that corresponds well to human judgment. Our method ...
Added: November 11, 2019
Алексейчук Н. Н., Sarkisyan V., Emelyanov A. et al., , in: Digital Transformation and Global Society. Fourth International Conference, DTGS 2019, St. Petersburg, Russia, June 19–21, 2019, Revised Selected Papers.: Springer, 2019. P. 68–81.
In this paper, we present a project on the analysis of an extensive corpus of strategic planning documents, devoted to various aspects of the development of Russian regions. The main purposes of the project are: 1) to extract different aspects of goal setting and planning, 2) to form an ontology of goals and criteria of ...
Added: October 30, 2019
Koltsov S., Ignatenko V., Koltsova O., Entropy 2019 Vol. 21 No. 7 P. 1–29
Topic modeling is a popular approach for clustering text documents. However, current tools have a number of unsolved problems such as instability and a lack of criteria for selecting the values of model parameters. In this work, we propose a method to solve partially the problems of optimizing model parameters, simultaneously accounting for semantic stability. ...
Added: July 5, 2019
Zelenkov Y., , in: Knowledge Management in Organizations. 14th International Conference, KMO 2019, Zamora, Spain, July 15–18, 2019, ProceedingsVol. 1027.: Switzerland: Springer, 2019. P. 324–335.
The intellectual structure of academic discipline can be viewed as a set of interacting topics evolving over time. Dynamics of those topics i.e. changes in their popularity and impact is the subject of special attention because it reflects a shift in actual researchers’ interest. This paper analyzes topics of knowledge management (KM) on the base ...
Added: June 14, 2019
Ignatenko V., Sergei Koltcov, Staab S. et al., Journal of Physics: Conference Series 2019 Vol. 1163 No. 1 P. 1–6
In the framework of this paper we apply multifractal formalism to the analysis of
statistical behaviour of topic models under variation of the number of topics. Fractal analysis
of topic models allows to show that self-similar fractal clusters exist in large textual collections.
We provide numerical results for 3 topic models (PLSA, ARTM, LDA Gibbs sampling) on
2 datasets, ...
Added: November 30, 2018
Molchanov D., Kharitonov V., Artem Sobolev et al., / Series arxiv.org "stat.ML". 2018.
We extend the existing framework of semi-implicit variational inference (SIVI) and introduce doubly semi-implicit variational inference (DSIVI), a way to perform variational inference and learning when both the approximate posterior and the prior distribution are semi-implicit. In other words, DSIVI performs inference in models where the prior and the posterior can be expressed as an ...
Added: October 21, 2018