https://repository.hneu.edu.ua/handle/123456789/178 2026-09-10T23:05:31Z https://repository.hneu.edu.ua/handle/123456789/41645 Назва: Автоматизація огляду відгуків споживачів на базі методів природної мовної обробки Автори: Скорін Ю. І. Короткий огляд (реферат): У статті розглянуто створення та всебічне визначення дієвості комплексної системи автоматизованого огляду відгуків споживачів на базі новітніх підходів NLP, що гарантує формування практичних рекомендацій стосовно впровадження результатів у бізнес-середовище. Результати дослідження можна використати для побудови програмних комплексів, які дозволять компаніям краще обробляти запити власних клієнтів, швидко відгукуватися на негаразди, покращувати якість своїх виробів і послуг. 2026-01-01T00:00:00Z https://repository.hneu.edu.ua/handle/123456789/41471 Назва: Implementing Artificial Intelligence for Scrum Teams Management Автори: Znakhur S.; Znakhur L. Короткий огляд (реферат): The paper addresses the optimization of Scrum team management in application development, where traditional frameworks suffer from human cognitive biases, estimation errors, device fragmentation, and complex app store release lifecycles. The authors propose a novel Hybrid AI-Agile Framework that integrates predictive machine learning models, natural language processing, and multi-agent Large Language Model (LLM) architectures to automate backlog refinement, sprint planning, risk assessment, dynamic resource allocation, and retrospective feedback. The operational mechanics feature specialized virtual role agents (Product Owner, Developer Agent, Technical Code Reviewer, and Scrum Master Facilitator), supported by mathematical models for real-time risk scoring and linear-programming resource allocation, and execute locally via an Ollama server to ensure enterprise data privacy. The proposed system establishes a reproducible paradigm for integrating cognitive automation with human-in-the-loop oversight to reduce IT project failure rates in information systems engineering. 2026-01-01T00:00:00Z https://repository.hneu.edu.ua/handle/123456789/41444 Назва: Neural network optimisation: quantisation techniques, practical benchmarking, and deployment recommendations Автори: Знахур С. Короткий огляд (реферат): The paper addresses the deployment of deep convolutional neural networks under the memory and energy limitations typical of mobile hardware, focusing on quantisation as a practical optimisation technique. The author performs a systematic empirical comparison of five convolutional architectures (MobileNetV2, ResNet-50, DenseNet-121, EfficientNet-B3, EfficientNet-B4), each converted with TensorFlow Lite Converter into three numerical-precision formats — FP32 baseline, FP16 and INT8 — and benchmarked on a physical iPhone 13 device (Apple A15 Bionic) through a purpose-built React Native 0.82 application using the react-native-fast-tflite library. Classification accuracy is assessed on a 500-image ImageNet-v2 subset. The results demonstrate that FP16 conversion halves model size while keeping Top-1 accuracy loss below 0.3% for every architecture, whereas INT8 reduces file size by 71–74% and improves inference latency by up to 2.8 times for mobile-friendly networks, but yields no consistent latency gain for EfficientNet-B3/B4 because part of their computation graph falls back to CPU execution within the TensorFlow Lite Metal delegate. Three factors governing quantisation effectiveness are identified: the structural compatibility of an architecture with the hardware delegate, the representativeness of the calibration dataset, and the delegate's operator coverage. Based on these findings, the author formulates differentiated, evidence-based deployment recommendations for mobile AI developers, organised by architecture family and use-case requirements, and proposes a reusable conversion-and-benchmarking methodology applicable beyond the five architectures studied. 2026-01-01T00:00:00Z https://repository.hneu.edu.ua/handle/123456789/41441 Назва: Mobile information system based on optimized deep neural networks for resource-constrained platforms Автори: Знахур Л. Короткий огляд (реферат): The paper is devoted to designing a mobile information system built around optimised deep neural networks intended for resource-constrained platforms. The author provides a comprehensive analysis of model-optimisation techniques — numerical quantisation, structural pruning and knowledge distillation — with a primary focus on post-training quantisation (PTQ) and quantisation-aware training (QAT), and formulates the underlying scale-factor and zero-point mapping equations used for INT8 conversion. The practical part experimentally evaluates five convolutional architectures — MobileNetV2, ResNet-50, DenseNet-121, EfficientNet-B3, EfficientNet-B4 — converted with the TensorFlow Lite framework and benchmarked with a single-threaded React Native CLI application on a physical Apple iPhone (A14 Bionic). Four metrics are analysed for every model and precision format: disk footprint, inference latency, classification accuracy, and resource efficiency. The results confirm that FP16 conversion halves storage requirements with a negligible accuracy loss of up to 0.1%, while full INT8 static quantisation reduces file size by up to 75% and accelerates inference by up to 62.1%, at the cost of accuracy degradation ranging from 0.6% for MobileNetV2 to 2.4% for EfficientNet-B4. This variation is linked to the structural properties of each network: depthwise separable convolutions and linear bottlenecks in MobileNetV2, together with the residual shortcuts of ResNet-50, provide strong resilience to fixed-point rounding errors, whereas the dense feature concatenation of DenseNet-121 and the tightly balanced compound scaling of the EfficientNet family make them considerably more sensitive to INT8 conversion. Part of the observed latency behaviour is attributed to the fallback of unsupported operators from INT8 kernels to FP32 CPU execution inside the TensorFlow Lite interpreter. Qualitative thermal observations, supported by published hardware benchmarks, additionally indicate 35–50% lower energy consumption for INT8 models compared with FP32 baselines. The findings support recommending quantisation-aware training for architecture families most sensitive to static INT8 quantisation, so as to preserve classification accuracy while retaining the compression and efficiency benefits required for mobile deployment. 2026-01-01T00:00:00Z