Enhancing IoT Security in Russian Language Teaching: A Improved BPNN and Blockchain-Based Approach for Privacy and Access Control

Abstract

Russian language instruction emerges as a pivotal course in tertiary education, necessitating novel approaches to maintain instructional quality and efficacy. This study introduces a novel approach to Russian language teaching that combines the robustness of Machine Learning with the security framework of Blockchain technology and is tailored to the unique needs of the Internet of Things (IoT) environment. At its core, the study creates an advanced back-propagation deep neural network enriched with a deep noise-reducing auto-encoder and a support vector machine to improve privacy and access control in IoT-based educational platforms. The proposed model employs a polynomial kernel function and a one-error penalty factor in a single hidden layer, resulting in a system that is not only efficient in handling small-scale data samples but also adept at processing larger data volumes, a common scenario in IoT settings. This design effectively overcomes the problems of overfitting and slow convergence that are common in traditional models. Furthermore, the incorporation of blockchain technology ensures a decentralized and secure data handling framework, reinforcing the privacy and access control aspects that are critical in the digital education domain. The combination of these technologies yields a more rational, scientifically based evaluation system, propelling the standardization and enhancement of Russian language instruction forward. This method not only improves language teaching quality, but it also paves the way for more secure, scalable, and efficient IoT applications in educational settings.