?
Fast Nearest-Neighbor Classifier based on Sequential Analysis of Principal Components
Ch. 7. P. 73–80.
Sokolova A., Savchenko A.
In this paper we improve the speed of the nearest neighbor classifiers of a set of points based on sequential analysis of high-dimensional feature vectors. Each input object is associated with a sequence of principal component scores of aggregated features extracted by deep neural network. The number of components in each element of this sequence is dynamically chosen based on explained proportion of total variance for the training set.We propose to process the next element with higher explained variance only if the decision for the current element is unreliable. This reliability is estimated by matching of the ratio of the minimum distance and all other distances with a certain threshold. Experimental study for face recognition with the Labeled Faces in the Wild and YouTube Faces datasets demonstrates the decrease of running time up to 10 times when compared to conventional instance-based learning.
Keywords: face recognition principal component analysis Sequential analysishigh-dimensional features deep neural networknearest neighbor classifier
Publication based on the results of:
In book
Vol. 11832. , Cham: Springer, 2019.
Liliya А. Rodionova, Elena D. Kopnova, Kamalendinova S. et al., Population and Economics 2026 No. 10(2) P. 92–114
This article proposes an approach to forecasting fertility using Google Trends search query statistics. Unlike existing studies on fertility forecasting, this paper introduces a method for aggregating thematic search queries using machine learning techniques, reducing the dimensionality of search queries through principal component analysis, and incorporating a lag structure of search queries ranging from 1 ...
Added: April 7, 2026
Sedashov E., Vitaly Kabernik, Stukal D., , in: 2025 18th International Conference on Management of Large-Scale System Development (MLSD).: IEEE, 2025. P. 1–4.
We offer a novel approach to the measurement of national power. The approach consists of two related parts: raw data processing and a measurement model. Suggested approach improves existing practices by allowing dynamic changes in the weights of national power components. It also returns credible intervals around point estimates, paving the way for statistical comparisons ...
Added: November 21, 2025
Konstantin Polyakov, Marina Polyakova, Polyakov A., Procedia Computer Science 2025 Vol. 266 P. 205–212
This study is devoted to the analysis of the relationship between the solution of two different tasks facing the management of a commercial organization - ensuring the efficiency and sustainability of the business. An analysis of publications on this topic shows the existence of contradictory conclusions. The authors hypothesize that there is a strong statistical ...
Added: October 8, 2025
Aladyshkina A. S., Leonova L., Ушакова Е. А., Псковский регионологический журнал 2025 Т. 21 № 2 С. 35–51
В статье представлен анализ дифференциации регионов в контексте устойчивого развития. В качестве объекта исследования использовались субъекты Российской Федерации. На основе данных официальной статистики составлен авторский перечень показателей, отражающих экономическое, социальное и экологическое развитие региона с учетом оценки эффективности управления. Для выявления дифференциации территорий по выбранным характеристикам был выбран кластерный анализ с использованием метода главных компонент. ...
Added: September 4, 2025
Savchenko A., Maslov D., Makarov I., , in: ECAI 2024. 27th European Conference on Artificial Intelligence, October 19 – 24 October 2024, Santiago de Compostela, Spain – Including 13th Conference on Prestigious Applications of Intelligent Systems (PAIS 2024).: IOS Press, 2024. P. 4439–4442.
Added: February 15, 2025
Arkhipova M., Sirotin V., , in: 2024 17th International Conference on Management of Large-Scale System Development (MLSD).: IEEE, 2024. Ch. 12 P. 1–4.
Added: October 14, 2024
Makarov I., Вопросы государственного и муниципального управления 2024 № 3 С. 81–109
In the circumstances of digital transformation of the state and society, the indicative approach to analyzing and governing digitalization finds active usage. However, the characteristics of relevant data cause methodological problems, including statistical duplication. The article considers various methods for constructing composite digitalization indices based on secondary aggregation of principal components in order to overcome ...
Added: October 11, 2024
Konstantin Polyakov, Marina Polyakova, , in: Procedia Computer Science, Volume 242: 11th International Conference on Information Technology and Quantitative Management (ITQM 2024).: ScienceDirect, 2024. P. 1125–1132.
Added: September 11, 2024
Polyakov K., Polyakova M. V., Экономический журнал Высшей школы экономики 2023 Т. 27 № 3 С. 390–411
This study is devoted to the analysis of the statistical relationship between the effectiveness of the bank and its specialization. Efficiency in the study is considered as a metric that evaluates the quality of management of an organization. Specialization in this case refers to the bank's concentration on certain types of banking products. The specifics ...
Added: October 30, 2023
Melville A. Y., Мальгин А. В., Mironyuk M. et al., Полис. Политические исследования 2023 № 5 С. 153–171
In recent decades, the expanding volume, diversity and coverage of data have created new or have transformed existing areas of research. They have also turned data into a key element of politics today. In this context, the status of empirical research that became the political science mainstream at the turn of the 20th - 21st ...
Added: September 29, 2023
Savchenko A., Savchenko L., Makarov I., IEEE Access 2023 Vol. 11 P. 65977–65990
This paper addresses the face recognition task for offline mobile applications. Using AutoML techniques, a novel technological framework is proposed to develop a fast neural network-based facial feature extractor for a concrete device. First, the Once-for-All SuperNet is trained on a large facial dataset. Each device is characterized by its lookup table, which contains the ...
Added: August 28, 2023
Sokolova A., Savchenko A., , in: 2023 IX International Conference on Information Technology and Nanotechnology (ITNT).: IEEE, 2023. P. 1–5.
In this paper, we explore the possibility to improve efficiency of face recognition using information about anomaly input images. Indeed, modern publicly-available datasets typically contain images of mostly middle-aged and Caucasian people, which cause most algorithms to fail on photos of older people or children, rarer ethnicities, poor-quality images, etc. Detection of such anomaly data ...
Added: June 13, 2023
Sokolova A., Savchenko A., , in: 2022 International Joint Conference on Neural Networks (IJCNN).: Institute of Electrical and Electronics Engineers Inc., 2022.
One of the main issues in face identification is to create a real-time application with high accuracy. Images are presented by high-dimensional feature vectors that are produced by convolutional neural networks. In order to effectively process such vectors, the hierarchical algorithm was proposed in this paper that applies sequential analysis to search the nearest neighbors ...
Added: May 29, 2023
Churaev E., Savchenko A., Software Impacts 2023 Vol. 16 Article 100507
Nowadays, many meetings, lessons, conferences, and presentations are organized online, where it is complicated to communicate with an audience and control their engagement and emotions. In this article, we present a novel C++ application that is led to help estimate facial identities and expressions. It captures a screen with a window of an arbitrary online ...
Added: May 18, 2023
Barabashev A. G., Zarochintcev S., Makarov I., Государственное управление. Электронный вестник 2022 № 95 С. 192–208
The article presents a mechanism for searching for unexpected, suddenly realized and unavoidable country risks of high-tech development (“black swans” of high-tech development) using an indicative approach to risk analysis; introduces the concept of stepped risk corridors of high-tech development as conventions that establish acceptable risk boundaries. An acceptable corridor of high-tech risks has also ...
Added: November 23, 2022
Drapkin I. M., Чукавина К. В., Грозных Р. И., Экономика и математические методы 2021 Т. 57 № 1 С. 33–42
The paper analyses the influence of institutional development level on bilateral foreign direct investment (FDI) flows. The theory suggests that more developed institutions in both home and host countries should increase FDI flows between countries. At the same time existing empirical research often have controversial results. We use the database on bilateral FDI flows for ...
Added: November 3, 2022
Malik M. S., Soft Computing 2020 Vol. 24 No. 18 P. 13913–13928
The number of reviews is rapidly increasing which creates information overload problem. To solve this problem, many websites introduced a feedback mechanism to vote for a review (helpful or not). The attracted votes reflect the review helpfulness. This study addresses the review helpfulness prediction problem and investigated the impact of review, reviewer and product features. ...
Added: November 1, 2022
Alexandrov D., Программная инженерия 2022 Vol. 13 No. 7 P. 331–343
Trends in computer vision and pattern recognition and capabilities of modern computers contributed to a consid- erable amount of research of these areas application in facial recognition systems. The purpose of this paper is to investigate the most significant methods of face recognition. In the first two sections of current paper, the methods of face ...
Added: October 31, 2022
Suleimanova A., Zangieva I., Tomsk State University Journal of Philosophy, Sociology and Political Science 2022 No. 69 P. 152–160
It is a common practice among social scientists to use “factor analysis” and “principal components analysis” interchangeably, even though PCA is not a factor extraction method, but a dimension reduction technique. Most of the recent studies with factor analysis rely solely on PCA or fail to specify which factor extraction method was used. Supposedly, it ...
Added: October 27, 2022
Manna S., Ghildiyal S., Bhimani K. R., IEEE Access 2020 Article 1
Face recognition (FR) and verification is the immeasurable technology to encounter any criminal activities nowadays. With the remarkable applications extending from criminal ID, security, and observation to amusement sites. This system (recognition of faces) is exceptionally helpful in banks, air terminals, and different associations for screening customers. In deep learning, convolutional neural networks (CNN) have ...
Added: October 14, 2022
Polyakov K., Polyakova M. V., Zhukova L., Экономический журнал Высшей школы экономики 2022 Т. 26 № 3 С. 450–474
The profitability and sustainability of the bank in the long term is largely determined by
the quality of its management. Recognition of the importance of this factor is, for example, its inclusion
as one of the components in the CAMELS rating system, which has not lost its popularity
so far. At the same time, a direct quantitative assessment ...
Added: October 13, 2022
Маршалко Г. Б., Никифорова Л. О., Automatic Control and Computer Sciences 2019 No. 8 P. 980–986
We present a spoofing attack on biometric identification system, which uses Eigenfaces algorithm for image classification. In this attack we try to modify an adversarial image in such a way that it is recognized as a target image while preserving visual similarity to the initial image. Like other similar attacks, we use a variant of ...
Added: October 7, 2022