We investigated the process of unsupervised generative learning and the structure of informative generative representations of images of handwritten digits (MNIST dataset). Learning models with the architecture of spa...
详细信息
A method for determining the probability of occurrence of a critical combination of events for the three-element minimum cross-sections of attributes of the characteristics of the software quality model of intelligent...
详细信息
The main methods of analysis the accuracy of the geolocation system were studied. The aspects of technology for obtaining geographical coordinates using Google geolocation system (GPS) were considered. The best method...
详细信息
DC traction motors are widely used in all branches of urban economy. They began to be used in industry about a hundred years ago with the advent of the first calculation methods. When designing a DC motor, it is impor...
详细信息
Most of the sensor devices in the Internet of Things systems are based on energy-efficient microcontrollers, the computing resources of which are limited, as well as the amount of available memory. Increasing the secu...
详细信息
ISBN:
(数字)9798350384499
ISBN:
(纸本)9798350384505
Most of the sensor devices in the Internet of Things systems are based on energy-efficient microcontrollers, the computing resources of which are limited, as well as the amount of available memory. Increasing the security of the use of such devices with the help of neural networks is an important and urgent problem. The article describes the possibility of using artificial neural networks in small microcontrollers with limited resources. The purpose of this work is to check the possibility of calculating neural networks based on integer arithmetic to reduce the time of calculating a neural network and eliminate data normalization operations, as well as to evaluate the feasibility of using such neural networks in the field of security of the Internet of Things in comparison with traditional methods, such as black lists and white lists. The following results were obtained: when switching to integer arithmetic, compared to floating point, the accuracy of the result calculations is within the permissible error of neural network training, that is, it has not changed. Execution time decreased by $30-96 \%$ , depending on the architecture of the microcontroller. The program size is reduced by $22-48 \%$ , also depending on the microcontroller architecture. Conclusions: the possibility and expediency of using neural networks optimized for microcontrollers with limited resources was proved. This will increase the security of Internet of Things systems, especially against device authentication threats and intrusion detection. Prospects for further research are determined.
This study aims to reveal the best deep learning models that are improved and optimized by predicting undesirable behavior patterns using a dataset consisting of artificial and real exam data of students taking online...
详细信息
ISBN:
(数字)9798331509934
ISBN:
(纸本)9798331509941
This study aims to reveal the best deep learning models that are improved and optimized by predicting undesirable behavior patterns using a dataset consisting of artificial and real exam data of students taking online distance education courses in an online environment through the distance education system. Using online exam data of 129 students, the researchers conducted analysis with two different scenarios to determine the best prediction performance through regression and classification models. The model we proposed was determined as a four-layer DNN with 80.4% test performance in detecting students who “cheated” from undesirable behavior patterns, which was performed with K-10, K-5 and K-3 cross-validation. The results prove that students' online distance education exam data can be easily applied to the DNN model. The models presented in the study provide a roadmap for educational institutions to evaluate their online examination practices and develop more effective strategies for academic honesty.
This work presents a practical approach of designing a functionally safe ECU for automotive application by implementing the Model Based Development (MBD) methodology. Functional safety (FuSa) in automotive can be achi...
This work presents a practical approach of designing a functionally safe ECU for automotive application by implementing the Model Based Development (MBD) methodology. Functional safety (FuSa) in automotive can be achieved by strictly following the processes of electrical and electronic systems development as given in the ISO 26262. The scope of FuSa encompasses the risks emerging from systematic failures and random hardware failures. One approach to mitigate these failures is the Model Based Development (MBD) on which the ISO 26262 provides guidance. This approach enables to model an ECU and accelerates an IP creation by automatic RTL code generation for rapid prototyping on an FPGA level. In this work the MBD is implemented in accordance to the ISO 26262 to design and test a hardware model of an embedded ECU. The embedded Design under Test (DUT) is simulated and its output verified in Matlab/Simulink. Upon successful results, the RTL codes of the DUT are generated for a target FPGA. The RTL codes are simulated and the functionality of the DUT is verified at the FPGA level.
In this work, the effectiveness of using classical machine learning methods and modern deep neural network models for intrusion detection in computer networks has been investigated. The purpose of this work is to deve...
详细信息
ISBN:
(数字)9798350384499
ISBN:
(纸本)9798350384505
In this work, the effectiveness of using classical machine learning methods and modern deep neural network models for intrusion detection in computer networks has been investigated. The purpose of this work is to develop a model for detecting intrusions into computer networks based on the Transformer model using tabular input data. In this work, the UNSW-NB15 dataset is used as the source data. This dataset contains information about normal network behaviour as well as during synthetic intrusions. Models for intrusion detection in computer networks based on machine learning methods were investigated: Decision Tree, KNN, Logistic Regression, SVM, Gradient Boosting, Random Forest. A method of converting tabular data into images was developed, which made it possible to build intrusion detection models based on Vision Transformer and Vision Transformer for small-size datasets on modern Transformer architecture. The research results showed that developed model based on Vision Transformer and Vision Transformer for small-size datasets improves the quality of identification, and eliminates the need for a preprocessing step such as dimensionality reduction.
The concept of acceptable systems with a limited level of systemic pathologies and dysfunctions is considered. The main provisions of cognitive modeling are given and the tasks of single- and multi-criteria optimizati...
详细信息
The work is devoted to solving the current scientific and technical problem of constructing a diagnostic decision support system in medicine based on a heterogeneous ensemble classifier model that implements two appro...
详细信息
ISBN:
(数字)9798350384499
ISBN:
(纸本)9798350384505
The work is devoted to solving the current scientific and technical problem of constructing a diagnostic decision support system in medicine based on a heterogeneous ensemble classifier model that implements two approaches to formulating a diagnostic conclusion: a probabilistic one based on the analysis of the training sample, and expert information on the structure of symptom complexes. The choice of prototype matching method as a probabilistic component is justified. Formalization of expert information on the structure of symptom complexes was carried out by representing symptom complexes of diseases with numerical intervals of linguistic variables. Options for taking into account expert assessments about the structure of symptom complexes in an ensemble classifier are considered. Test verification of the developed classifier was done on real medical data and confirmed the effectiveness of its work.
暂无评论