Detection of malware using an artificial neural network based on adaptive resonant theory

2021 ◽  
pp. 69-82
Author(s):  
D. G. Bukhanov ◽  
◽  
V. M. Polyakov ◽  
M. A. Redkina ◽  
◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Significant Loss ◽  
Malicious Code ◽  
Resonance Theory ◽  
Similarity Parameter ◽  
Adaptive Resonance ◽  
Artificial Neural ◽  
And Behavior ◽  
Classification Time

The process of detecting malicious code by anti-virus systems is considered. The main part of this process is the procedure for analyzing a file or process. Artificial neural networks based on the adaptive-resonance theory are proposed to use as a method of analysis. The graph2vec vectorization algorithm is used to represent the analyzed program codes in numerical format. Despite the fact that the use of this vectorization method ignores the semantic relationships between the sequence of executable commands, it allows to reduce the analysis time without significant loss of accuracy. The use of an artificial neural network ART-2m with a hierarchical memory structure made it possible to reduce the classification time for a malicious file. Reducing the classification time allows to set more memory levels and increase the similarity parameter, which leads to an improved classification quality. Experiments show that with this approach to detecting malicious software, similar files can be recognized by both size and behavior.

PASSPHRASE AUTHENTICATION BASED ON TYPING STYLE THROUGH AN ART 2 NEURAL NETWORK

2002 ◽  
Vol 02 (02) ◽  
pp. 131-152 ◽  
Author(s):  
JASON BECHTEL ◽  
GURSEL SERPEN ◽  
MARCUS BROWN
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Authentic Sample ◽  
Adaptive Resonance Theory ◽  
Resonance Theory ◽  
Adaptive Resonance ◽  
Network Algorithm ◽  
Neural Network Algorithm ◽  
Artificial Neural ◽  
Artificial Neural Network Algorithm

This study proposes the use of an artificial neural network algorithm to perform passphrase authentication based on the typing style of a user. The only hardware required is a keyboard. Prior studies have demonstrated the feasibility of this approach and its limitations, one of which was the need for collection of impostor samples for training the artificial neural network based classifier algorithm. This requirement is rather impractical for most application domains. The proposed study eliminates the need to collect impostor samples by employing an unsupervised and self-organizing artificial neural network algorithm, the Adaptive Resonance Theory 2 neural network, and therefore pushes the passphrase authentication technology one step closer to the realm of practical implementation. The preliminary study performed demonstrates that it is possible to train an Adaptive Resonance Theory 2 neural network using only authentic sample data and still provide a relatively low impostor pass rate. Given the minimal cost and easy in-field trainability of the proposed passphrase authentication system, the developed system can greatly enhance the security of computing environments with wide acceptance.

Analysis of Structural Integrity Using an ARTMAP-Fuzzy Artificial Neural Network

2013 ◽  
Vol 838-841 ◽  
pp. 3287-3290 ◽  
Author(s):  
Adriano dos Santos e Souza ◽  
Fábio Roberto Chavarette ◽  
Fernando Parra dos Anjos Lima ◽  
Mara Lúcia Martins Lopes ◽  
Simone Silva Frutuoso de Souza
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Structural Integrity ◽  
Base Line ◽  
Structural Failure ◽  
Resonance Theory ◽  
Adaptive Resonance ◽  
Artificial Neural ◽  
Identification And Characterization

This paper presents the application of artificial neural networks in the analysis of the structural integrity of a building. The main objective is to apply an artificial neural network based on adaptive resonance theory, called ARTMAP-Fuzzy neural network and apply it to the identification and characterization of structural failure. This methodology can help professionals in the inspection of structures, to identify and characterize flaws in order to conduct preventative maintenance to ensure the integrity of the structure and decision-making. In order to validate the methodology was modeled a building of two walk, and from this model were simulated various situations (base-line condition and improper conditions), resulting in a database of signs, which were used as input data for ARTMAP-Fuzzy network. The results show efficiency, robustness and accuracy.

scholarly journals LUNG DISORDERS DETECTION BASED ON IRISES IMAGE USING COMPUTATIONAL INTELLIGENT ART

2016 ◽  
Vol 8 (2) ◽  
Author(s):  
Anna Triwijayanti K. ◽  
Hadi Suwastio ◽  
Rini Damayanti
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Artificial Neural Network ◽  
Network Architecture ◽  
Learning Algorithm ◽  
Back Propagation ◽  
Digital Camera ◽  
Resonance Theory ◽  
Adaptive Resonance ◽  
Artificial Neural

Iridology as a way of revealing human organs and tissues conditions is done by iridologist by taking the image of both irises of the patients. This can be done by using a digital camera and observe each iris on the LCD display or connect the camera to a computer or a television set and observe it through the display. Research on computerized iridology has been performed before by using artificial neural network of back propagation, which is a kind of supervised learning algorithm, as the classifier [13]. Such system should be able to retain its stability while still being plastic enough to adapt to arbitrarily input patterns. Adaptive Resonance Theory (ART), another kind of artificial neural network which uses unsupervised learning algorithm, has some important traits, such as real-time learning, self-stabilizing memory in response to arbitrarily many input patterns, and fast adaptive search for best match of input-to-stored patterns [9]. That way, ART architecture is expected to be the best stable and adaptable solution in changing environment of pattern recognition. In this research, the lung disorders detection is simply designed through the steps of segmentation, extraction of color variations, transformation of lung and pleura representation area in iris image to binary form as the input of ART 1, and pattern recognition by ART 1 neural network architecture. With 32 samples and 4 nodes of output layer of ART1, the system is able to determine the existences of the four stadiums of lung disorders (acute, subacute, chronic and degenerative) in relatively short time process (approximately 1.8 to 3.2 seconds) with the accuracy of stadium recognition 91.40625% by applying the vigilance parameter value of 0.4.Keywords: iridology, lung, pleura, segmentation, ART 1 neural network

scholarly journals Smart Doctor: Performance of Supervised ART-I Artificial Neural Network for Breast Cancer Diagnoses

2020 ◽  
pp. 2385-2394
Author(s):  
Kamal R. AL-Rawi ◽  
Saifaldeen K. AL-Rawi
Keyword(s):  
Breast Cancer ◽  
Neural Network ◽  
Artificial Neural Network ◽  
Breast Cancer Diagnosis ◽  
Breast Cancer Dataset ◽  
Resonance Theory ◽  
Fuzzy Classifier ◽  
Cancer Dataset ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

Wisconsin Breast Cancer Dataset (WBCD) was employed to show the performance of the Adaptive Resonance Theory (ART), specifically the supervised ART-I Artificial Neural Network (ANN), to build a breast cancer diagnosis smart system. It was fed with different learning parameters and sets. The best result was achieved when the model was trained with 50% of the data and tested with the remaining 50%. Classification accuracy was compared to other artificial intelligence algorithms, which included fuzzy classifier, MLP-ANN, and SVM. We achieved the highest accuracy with such low learning/testing ratio.

scholarly journals A Combination of Artificial Neural Network and Artificial Immune System for Virus Detection

2016 ◽  
Author(s):  
Mai Trong Khang ◽  
Vu Thanh Nguyen ◽  
Tuan Dinh Le
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
False Positive Rate ◽  
Virus Detection ◽  
Malicious Code ◽  
Immune Network ◽  
Artificial Immune ◽  
Positive Rate ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

In this paper, we propose an Artificial Neural Immune Network (ANIN) for virus detection. ANIN is a combination of Artificial Neural Network (ANN) and Artificial Immune Network (AiNet). In ANIN, each ANN is considered as a detector. A pool of initial detectors then undergoes a mature process, called AiNet, to improve its recognizing ability. Thus, more than one ANN objects can cooperate to detect malicious code. The experimental results show that ANIN can achieve a detection rate of 87.98% on average with an acceptable false positive rate.

scholarly journals Multi Variant Feature Similarity based Behavior Tracking in Video Surveillance using Ann

2019 ◽  
Vol 8 (3) ◽  
pp. 1419-1423
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Object Tracking ◽  
Behavior Analysis ◽  
Video Surveillance ◽  
Feature Model ◽  
Feature Similarity ◽  
Artificial Neural ◽  
Behavior Tracking ◽  
And Behavior

The problem of video surveillance has been well studied which has been adapted for several issues. The behavior of any human can be monitored through video surveillance. There are number of approaches available for the video surveillance and behavior analysis. The previous methods uses background models, object tracking for the problem of behavior analysis. The methods suffer with poor accuracy in behavior analysis. To improve the performance, a multi variant feature similarity model based behavior tracking in video surveillance is presented. The method involves in identifying interest points throughout the images of video. Second, the changing feature has been identified to measure the multi variant feature similarity by using multi variant feature model. Based on the MVFS, the object tracking is performed. The human tracking is performed in the same way and the multi variant features are trained with artificial neural network which has number of behavior classes. At the testing phase, the video has been removed with background features according to the multi feature model adapted. Once the object has been identified, then tracking and behavior analysis is performed by measuring MVFS with the features at different behavior classes. The artificial neural network has been used for the classification of behavior identified through video surveillance. The method would produce higher accuracy and improves the performance.

Comprehensive Method of Botnet Detection Using Machine Learning

2021 ◽  
Vol 12 (4) ◽  
pp. 0-0
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Malicious Code ◽  
Large Network ◽  
Botnet Detection ◽  
Comprehensive Method ◽  
Input Layer ◽  
Optimal Accuracy ◽  
Artificial Neural ◽  
Hidden Layer

The botnet interrupts network devices and keeps control of the connections with the command, which controls the programmer, and the programmer controls the malicious code injected in the machine for obtaining information about the machines. The attacker uses a botnet to commence dangerous attacks as DDoS, phishing, despoil of information, and spamming. The botnet establishes with a large network and several hosts belong to it. In the paper, the authors proposed the framework of botnet detection by using an Artificial Neural Network. The author research upgrading the extant system by comprising of cache memory to fast the process. Finally, for detection, the author used an analytical approach, which is known as an artificial neural network that contains three layers: the input layer, hidden layer, output layer, and all layers are connected to correlate and approximate the results. The experiment result determines that the classifier with 25 epochs gives optimal accuracy is 99.78 percent and shows the detection rate is 99.7 percent.

Sign in / Sign up

Export Citation Format

Share Document