The rise of virtual exams has introduced new challenges in maintaining academic integrity, particularly in detecting collaborative cheating. In this study, we propose a novel framework that combines Graph Neural Networks (GNNs) with Federated Learning (FL) to detect collaborative cheating while preserving student privacy. Our approach models student interactions as a graph, where nodes represent students and edges represent potential cheating relationships. By training the GNN in a federated manner, we ensure that sensitive data remains on students' devices, addressing ethical and legal concerns.
| Mualliflar | Abdullayev, Jasurbek, Ergashev , Otabek |
|---|---|
| Jurnal | Al-Farg'oniy avlodlari |
| Nashr sanasi | 2025-03-23 |
| Son | 1 |
| Betlar | 106-111 |
| Til | Rus |
Графовые нейронные сети (GNN), федеративное обучение (FL), обнаружение группового мошенничества, защита конфиденциальности в ИИ, прокторинг виртуальных экзаменов, генерация синтетических данных, этичный ИИ в образовании, Graph Neural Networks (GNNs), Federated Learning (FL), Collaborative Cheating Detection, Privacy-Preserving AI, Virtual Exam Proctoring, Ethical AI in Education, Synthetic Data Generation, Graf neyron tarmoqlari (GNT), Federativ O‘qitish (FO’), Guruhiy aldashni aniqlash, Maxfiylikni saqlovchi sun’iy intellekt, Virtual imtihon nazorati, Ta’limda xulqiy va huquqiy sun’iy intellekt
Abstract: This article discusses the development of an automated system for selecting fabrics, taking into account the technical capabilities of equipment for the production of patterned fabrics. The methodology for…
Improving the energy efficiency of Internet of Things (IoT) devices is of great importance. This article proposes an adaptive algorithm that automatically regulates the data transmission interval in accordance with the…
Annotation: This article presents an expert system developed based on fuzzy logic for evaluating the performance of university teachers. Unlike traditional evaluation methods, the system performs assessments by taking…
This article discusses the design of a dynamic document template generation system based on existing documents and rule-based mechanisms. Methods of automating document workflows and generating flexible document…
As the complexity of ICS increases, security concerns also grow. Anomalies may indicate failures or cyberattacks, making early detection crucial. Traditional statistical and threshold-based methods are ineffective in…
The article describes the factors that determine energy losses and the efficiency of high-power transformers, as well as an analysis of the reasons for their occurrence. Based on the analysis, the efficiency of the…
This article provides an in-depth analysis of advanced algorithms for automatic object detection and classification in video imagery. The rapid advancements in artificial intelligence and deep learning have…
В данной статьи приведены результаты идентификации радионуклидного состава - 234U, 235U, 238U и коэффициентов радиоактивного равновесия - Крр между - 234U/238U в химконцентратах урана. В лабораторных условиях…
This article provides an overview of automated clinic systems and their main types. The importance of automation in healthcare lies in improving service quality, streamlining processes, and saving time. The article…
В данной статьи приведены результаты изучение значений суммарной - Σα и суммарной - Σβ -активности природных вод отобранных из ураноносных регионов. В лабораторных условиях изучены методов определение значений суммарной…