Перегляд за Автор "Pavlova, Olga"

Зараз показуємо 1 - 7 з 7
  • Ескіз
    Документ
    An ensemble machine learning approach for Twitter sentiment analysis
    (CEUR-WS, 2022-07-17) Radiuk, Pavlo; Pavlova, Olga; Hrypynska, Nadiia
    The presented study addresses the issue of classifying emotional expressions based on small texts (tweets) extracted from the social network Twitter. In this paper, we propose a novel approach to preprocessing tweets to fit them more effectively into the classification model. Moreover, we suggest utilizing two types of features, namely unigrams and bigrams, to expand the feature vector. The classification task of emotional expressions was performed according to several machine learning algorithms: raw random forest, gradient boosting random forest, support vector machine, multilayer perceptron, recurrent neural network, and convolutional neural network. The feature vector elements are presented as sparse and dense subvectors. As a result of computational experiments, it was found that the “appearance” in the reflection of the sparse vector provided higher performance than the “regularity.” The experiments also showed that deep learning approaches performed better than traditional machine learning techniques. Consequently, the best recurrent neural network achieved an accuracy of 83.0% on the test dataset, while the best convolutional neural network reached 83.34%. At the same time, it was discovered that the convolutional model with the support vector machine classifier showed better performance than the single convolutional neural network. Overall, the proposed ensemble method based on receiving the most votes according to the five best models’ predictions has reached an absolute accuracy of 85.71%, proving its practical usefulness.
  • Ескіз
    Документ
    Convolutional neural network for parking slots detection
    (CEUR-WS, 2022-06-17) Radiuk, Pavlo; Pavlova, Olga; El Bouhissi, Houda; Avsiyevych, Volodymyr; Kovalenko, Volodymyr
    With the rapid growth of transport number on our streets, the need for finding a vacant parking spot today could most of the time be problematic, but even more in the coming future. Smart parking solutions have proved their usefulness for the localization of unoccupied parking spots. Nowadays, surveillance cameras can provide more advanced solutions for smart cities by finding vacant parking spots and providing cars safety in the public parking area. Based on the analysis, Google Cloud Vision technology has been selected to develop a cyber-physical system for smart parking based on computer vision technology. Moreover, a new model based on the fine-tuned convolutional neural network has been developed to detect empty and occupied slots in the parking lot images collected from the KhNUParking dataset. Based on the achieved results, the performance of parking slots’ detections can be simplified, and its accuracy improved. The Google Cloud Vision technology as parking slots detector and a pre-trained convolutional neural network as a feature extractor and a classifier were selected to develop a cyber-physical system for smart parking. As a result of the computational investigation, the proposed fine-tuned CNN managed to process 66 parking slots in roughly 0.14 seconds on a single GPU with an accuracy of 85.4%, demonstrating decent performance and practical value. Overall, all considered approaches contain strengths and weaknesses and might be applied to the task of parking slots detection depending on the number of images, CCTV angle, and weather conditions.
  • Ескіз
    Документ
    Financial Risk and Customs Control in Humanitarian Water Logistics: A Machine Learning Approach.
    (CEUR, 2026-02-07) Dumanska, Ilona; Pavlova, Olga; Rabcan, Jan; Melnyk, Alona; Kharun, Olena
    This study explores the application of machine learning to mitigate financial and regulatory risks in humanitarian water logistics. Through the WaterWayfinder mobile platform, aid coordinators in Ukraine’s Kherson and Zaporizhzhia regions achieved measurable gains in operational efficiency and compliance. AI-driven route optimization reduced delivery times by up to 32% and fuel costs by 22%, while predictive modeling improved resource allocation and reduced exposure to high-cost disruptions. The system’s customs control module enabled pre-clearance planning and real-time regulatory updates, shortening border processing times by an average of 2.5 hours per shipment. Despite connectivity and data challenges, WaterWayfinder demonstrated resilience and adaptability in conflict-affected environments. Its modular architecture, offline capabilities, and integration with geospatial intelligence position it for broader deployment across crisis zones. The findings highlight WaterWayfinder’s potential as a scalable, data-driven framework for intelligent humanitarian logistics, aligning with global efforts to enhance transparency, agility, and cross-border coordination in aid delivery.
  • Ескіз
    Документ
    Information System for Logistical Support of Volunteer Tasks: Basics and Functionality.
    (CEUR Workshop Proceedings, 2023) Dumanska Ilona,; Pavlova, Olga; El, Bouhissi Houda
    The paper examines the theoretical and analytical principles of creating an information system for logistical support of volunteer tasks (IS LSVT) to ensure the achievement of humanitarian goals. The parameters of the external environment of IS LSVT functioning are identified on the basis of a system approach. The dominant role of time as a safety factor over profit in the logistic tasks of the IS LSVT functional model is determined. The conceptual principles of building the IS LSVT functional model were determined, taking into account the following parameters: the volunteer task, data on the territory of military operations, indicators of resource provision and criteria for the optimal load of road transport and routing. The elements of the IS LSVT structure in the conditions of military operations, which generates the necessary input information for building the task of resource optimization, have been established. Numerical experiments were carried out on IS LSVT approbation based on the method of resource minimization, during which the influence of the total time factor on the optimization of volunteer tasks based on the time spent on decisionmaking and cargo delivery under the conditions of curfew and the presence of an air alert was determined.
  • Ескіз
    Документ
    Information system for public places and institutions visualization with opportunities of inclusive access and optimal routing
    (Khmelnytskyi National University, 2022-04-14) Pavlova, Olga; Radiuk, Pavlo; Kravchuk, Sofia; Kulbachnyi, Vladyslav
    Inclusive access has been considered essential and relevant for decades. However, this issue has been in demand in the past years, both in Europe and Ukraine. One of the popular means of providing inclusive access within the city is information systems that are friendly to people with disabilities. The theoretical basis of such systems is the smart city concept, which has been briskly developed recently. It contains the principles of accessibility of public places, institutions, and establishments for people with special needs. In this work, it is analyzed the well-known algorithms for building optimal routes and available information services and mobile applications that solve the problem of visualizing public places and institutions with inclusive access and paving optimal routes to them.
  • Ескіз
    Документ
    Information Technology for Logistics Infrastructure Based on Digital Visualization and WEB-Cartography Under the Conditions of Military Conflicts
    (CEUR Workshop Proceedings, 2023) Dumanska, Ilona; Pavlova, Olga; Bouhissi, El Houda
    Current trends and obstacles to the development of logistics infrastructure based on IT solutions in the conditions of military conflict were considered. Basic components of digital logistics infrastructure based on Industry 4.0 technologies were indicated. Digital Visualization and the inclusion of WCAG 2.0 include analytical tools, simulation models in the management of logistics chains. However, Ukraine currently does not have a ready-made solution that would take into account all aspects of managing logistics chains in the conditions of a military conflict. The concept of information technology for logistics infrastructure based on digital visualization and WEB-cartography in conditions of military conflict is proposed. The economic effect of the synergy of Digital Visualization and WEBcartography for the proposed information technology for logistics infrastructure is highlighted and the organizational model of their implementation is presented.
  • Ескіз
    Документ
    Агентно-орієнтована інформаційна технологія оцінювання початкових етапів життєвого циклу програмного забезпечення на основі онтологічного підходу партнерів
    (Хмельницький національний університет, 2020) Павлова, Ольга Олександрівна; Pavlova, Olga
    У дисертаційній роботі розв’язана актуальна науково-прикладна задача автоматизації оцінювання початкових етапів життєвого циклу програмного забезпечення шляхом розроблення агентно-орієнтованої інформаційної технології (АОІТ) на основі онтологічного підходу, яка забезпечує: автоматизацію трудомісткого, рутинного та схильного до помилок завдання розбору специфікацій вимог та майже миттєве його виконання; підказку, де потрібна повторна робота над специфікацією вимог (користувач може переглядати відсутні атрибути та бачити області специфікації, яким потрібна додаткова увага, а також які вимоги потребують переробки); забезпечення навчання для нових розробників специфікацій, системних інженерів та менеджерів проєктів (використання цієї AOIT допомагає їм побачити помилки, які вони можуть допускати, та допомагає їм розпізнати ці помилки в роботі інших); допомогу в розробці вимог високої якості; допомогу у виправленні та усуненні помилок у вимогах там, де вони виникають – на ранніх етапах життєвого циклу програмного забезпечення – до того, як вони стануть дорожчими для виправлення; надання інструменту для вибору більш якісних специфікацій програмних вимог; безкоштовний доступ через Інтернет в будь-який час без будь-якої реєстрації.