scholarly journals Cost-Sensitive Approach to Improve the HTTP Traffic Detection Performance on Imbalanced Data

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Wenmin Li ◽  
Sanqi Sun ◽  
Shuo Zhang ◽  
Hua Zhang ◽  
Yijie Shi
Keyword(s):  
Feature Extraction ◽  
Cost Function ◽  
Network Traffic ◽  
Detection System ◽  
Imbalanced Data ◽  
Detection Performance ◽  
Learning Technology ◽  
The Cost ◽  
Decision Boundaries ◽  
Traffic Detection

Aim. The purpose of this study is how to better detect attack traffic in imbalance datasets. The deep learning technology has played an important role in detecting malicious network traffic in recent years. However, it suffers serious imbalance distribution of data if the traffic model skews towards the modeling in the benign direction, because only a small portion of traffic is malicious, while most network traffic is benign. That is the reason why the authors wrote this manuscript. Methods. We propose a cost-sensitive approach to improve the HTTP traffic detection performance with imbalanced data and also present a character-level abstract feature extraction approach that can provide features with clear decision boundaries in addition. Finally, we design a spark-based HTTP traffic detection system based on these two approaches. Results. The methods proposed in this paper work well in imbalanced datasets. Compared to other methods, the experiment results indicate that our system has F1-score in a high precision. Conclusion. For imbalanced HTTP traffic detection, we confirmed that the method of feature extraction and the cost function is very effective. In the future, we may focus on how to use the cost function to further improve detection performance.

scholarly journals Comparison of intrusion detection system based on feature extraction

2018 ◽  
Vol 7 (3.3) ◽  
pp. 536
Author(s):  
Pradeep Laxkar ◽  
Prasun Chakrabarti
Keyword(s):  
Feature Extraction ◽  
Big Data ◽  
Intrusion Detection ◽  
Network Traffic ◽  
Intrusion Detection System ◽  
Detection System ◽  
The Other ◽  
Traffic Classification ◽  
Redundant Feature ◽  
Network Traffic Classification

In network traffic classification redundant feature and irrelevant features in data create problems. All such types of features time-consuming make slow the process of classification and also affect a classifier to calculate accurate decisions such type of problem caused especially when we deal with big data. In this paper, we compare our proposed algorithm with the other IDS algorithm.  

scholarly journals Deep Feature Extraction via Sparse Autoencoder for Intrusion Detection System

2020 ◽  
Author(s):  
Cao Xiaopeng ◽  
Qu Hongyan
Keyword(s):  
Feature Extraction ◽  
Network Traffic ◽  
Time Series Data ◽  
Detection System ◽  
Series Data ◽  
High Dimensional ◽  
Novel Approach ◽  
Whale Optimization ◽  
Sparse Autoencoder ◽  
Low Dimensional

The massive network traffic and high-dimensional features affect detection performance. In order to improve the efficiency and performance of detection, whale optimization sparse autoencoder model (WO-SAE) is proposed. Firstly, sparse autoencoder performs unsupervised training on high-dimensional raw data and extracts low-dimensional features of network traffic. Secondly, the key parameters of sparse autoencoder are optimized automatically by whale optimization algorithm to achieve better feature extraction ability. Finally, gated recurrent unit is used to classify the time series data. The experimental results show that the proposed model is superior to existing detection algorithms in accuracy, precision, and recall. And the accuracy presents 98.69%. WO-SAE model is a novel approach that reduces the user’s reliance on deep learning expertise.

scholarly journals Malicious Network Traffic Detection Based on Deep Neural Networks and Association Analysis

Sensors ◽  
2020 ◽  
Vol 20 (5) ◽  
pp. 1452 ◽  
Author(s):  
Minghui Gao ◽  
Li Ma ◽  
Heng Liu ◽  
Zhijun Zhang ◽  
Zhiyan Ning ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Anomaly Detection ◽  
Association Analysis ◽  
Network Traffic ◽  
Deep Neural Network ◽  
Deep Neural Networks ◽  
Detection System ◽  
Detection Systems ◽  
Traffic Detection

Anomaly detection systems can accurately identify malicious network traffic, providing network security. With the development of internet technology, network attacks are becoming more and more sourced and complicated, making it difficult for traditional anomaly detection systems to effectively analyze and identify abnormal traffic. At present, deep neural network (DNN) technology achieved great results in terms of anomaly detection, and it can achieve automatic detection. However, there still exists misclassified traffic in the prediction results of deep neural networks, resulting in redundant alarm information. This paper designs a two-level anomaly detection system based on deep neural network and association analysis. We made a comprehensive evaluation of experiments using DNNs and other neural networks based on publicly available datasets. Through the experiments, we chose DNN-4 as an important part of our system, which has high precision and accuracy in identifying malicious traffic. The Apriori algorithm can mine rules between various discretized features and normal labels, which can be used to filter the classified traffic and reduce the false positive rate. Finally, we designed an intrusion detection system based on DNN-4 and association rules. We conducted experiments on the public training set NSL-KDD, which is considered as a modified dataset for the KDDCup 1999. The results show that our detection system has great precision in malicious traffic detection, and it achieves the effect of reducing the number of false alarms.

Smart teaching mode based on particle swarm image recognition and human-computer interaction deep learning

2020 ◽  
Vol 39 (4) ◽  
pp. 5699-5711
Author(s):  
Shirong Long ◽  
Xuekong Zhao
Keyword(s):  
Feature Extraction ◽  
Particle Swarm Optimization ◽  
Deep Learning ◽  
Real Time ◽  
Image Recognition ◽  
Particle Swarm ◽  
Learning Technology ◽  
Search Performance ◽  
Swarm Optimization ◽  
Teaching Mode

The smart teaching mode overcomes the shortcomings of traditional teaching online and offline, but there are certain deficiencies in the real-time feature extraction of teachers and students. In view of this, this study uses the particle swarm image recognition and deep learning technology to process the intelligent classroom video teaching image and extracts the classroom task features in real time and sends them to the teacher. In order to overcome the shortcomings of the premature convergence of the standard particle swarm optimization algorithm, an improved strategy for multiple particle swarm optimization algorithms is proposed. In order to improve the premature problem in the search performance algorithm of PSO algorithm, this paper combines the algorithm with the useful attributes of other algorithms to improve the particle diversity in the algorithm, enhance the global search ability of the particle, and achieve effective feature extraction. The research indicates that the method proposed in this paper has certain practical effects and can provide theoretical reference for subsequent related research.

scholarly journals DNA tetrahedron-mediated immune-sandwich assay for rapid and sensitive detection of PSA through a microfluidic electrochemical detection system

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Dezhi Feng ◽  
Jing Su ◽  
Yi Xu ◽  
Guifang He ◽  
Chenguang Wang ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Reaction Time ◽  
Electrochemical Detection ◽  
Detection System ◽  
Detection Performance ◽  
Pdms Layer ◽  
Microfluidic Chips ◽  
Serum Samples ◽  
Gold Nanoflowers ◽  
Dna Tetrahedron

AbstractProstate-specific antigen (PSA) is the most widely used biomarker for the early diagnosis of prostate cancer. Existing methods for PSA detection are burdened with some limitations and require improvement. Herein, we developed a novel microfluidic–electrochemical (μFEC) detection system for PSA detection. First, we constructed an electrochemical biosensor based on screen-printed electrodes (SPEs) with modification of gold nanoflowers (Au NFs) and DNA tetrahedron structural probes (TSPs), which showed great detection performance. Second, we fabricated microfluidic chips by DNA TSP-Au NF-modified SPEs and a PDMS layer with designed dense meandering microchannels. Finally, the μFEC detection system was achieved based on microfluidic chips integrated with the liquid automatic conveying unit and electrochemical detection platform. The μFEC system we developed acquired great detection performance for PSA detection in PBS solution. For PSA assays in spiked serum samples of the μFEC system, we obtained a linear dynamic range of 1–100 ng/mL with a limit of detection of 0.2 ng/mL and a total reaction time <25 min. Real serum samples of prostate cancer patients presented a strong correlation between the “gold-standard” chemiluminescence assays and the μFEC system. In terms of operation procedure, cost, and reaction time, our method was superior to the current methods for PSA detection and shows great potential for practical clinical application in the future.

scholarly journals Adaptive Detection of Direct-Sequence Spread-Spectrum Signals Based on Knowledge-Enhanced Compressive Measurements and Artificial Neural Networks

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2538
Author(s):  
Shuang Zhang ◽  
Feng Liu ◽  
Yuang Huang ◽  
Xuedong Meng
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Spread Spectrum ◽  
High Efficiency ◽  
Detection System ◽  
Direct Sequence Spread Spectrum ◽  
Detection Performance ◽  
Detection Methods ◽  
Direct Sequence ◽  
Artificial Neural

The direct-sequence spread-spectrum (DSSS) technique has been widely used in wireless secure communications. In this technique, the baseband signal is spread over a wider bandwidth using pseudo-random sequences to avoid interference or interception. In this paper, the authors propose methods to adaptively detect the DSSS signals based on knowledge-enhanced compressive measurements and artificial neural networks. Compared with the conventional non-compressive detection system, the compressive detection framework can achieve a reasonable balance between detection performance and sampling hardware cost. In contrast to the existing compressive sampling techniques, the proposed methods are shown to enable adaptive measurement kernel design with high efficiency. Through the theoretical analysis and the simulation results, the proposed adaptive compressive detection methods are also demonstrated to provide significantly enhanced detection performance efficiently, compared to their counterpart with the conventional random measurement kernels.

scholarly journals An Adaptive Optimization Method Based on Learning Rate Schedule for Neural Networks

2021 ◽  
Vol 11 (2) ◽  
pp. 850
Author(s):  
Dokkyun Yi ◽  
Sangmin Ji ◽  
Jieun Park
Keyword(s):  
Artificial Intelligence ◽  
Cost Function ◽  
Numerical Experiments ◽  
Global Minimum ◽  
Optimization Method ◽  
Learning Method ◽  
Adaptive Optimization ◽  
The Cost ◽  
Proof Of Convergence ◽  
Learning Data

Artificial intelligence (AI) is achieved by optimizing the cost function constructed from learning data. Changing the parameters in the cost function is an AI learning process (or AI learning for convenience). If AI learning is well performed, then the value of the cost function is the global minimum. In order to obtain the well-learned AI learning, the parameter should be no change in the value of the cost function at the global minimum. One useful optimization method is the momentum method; however, the momentum method has difficulty stopping the parameter when the value of the cost function satisfies the global minimum (non-stop problem). The proposed method is based on the momentum method. In order to solve the non-stop problem of the momentum method, we use the value of the cost function to our method. Therefore, as the learning method processes, the mechanism in our method reduces the amount of change in the parameter by the effect of the value of the cost function. We verified the method through proof of convergence and numerical experiments with existing methods to ensure that the learning works well.

Deep convolutional neural networks for human movement detection using wireless signals

2021 ◽  
pp. 1-10
Author(s):  
Chien-Cheng Leea ◽  
Zhongjian Gao ◽  
Xiu-Chi Huanga
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Convolutional Neural Network ◽  
Detection System ◽  
Deep Convolutional Neural Network ◽  
Human Detection ◽  
Two Dimensional ◽  
Dimensional Matrix ◽  
State Classification ◽  
Propagation Paths

This paper proposes a Wi-Fi-based indoor human detection system using a deep convolutional neural network. The system detects different human states in various situations, including different environments and propagation paths. The main improvements proposed by the system is that there is no cameras overhead and no sensors are mounted. This system captures useful amplitude information from the channel state information and converts this information into an image-like two-dimensional matrix. Next, the two-dimensional matrix is used as an input to a deep convolutional neural network (CNN) to distinguish human states. In this work, a deep residual network (ResNet) architecture is used to perform human state classification with hierarchical topological feature extraction. Several combinations of datasets for different environments and propagation paths are used in this study. ResNet’s powerful inference simplifies feature extraction and improves the accuracy of human state classification. The experimental results show that the fine-tuned ResNet-18 model has good performance in indoor human detection, including people not present, people still, and people moving. Compared with traditional machine learning using handcrafted features, this method is simple and effective.

Sign in / Sign up

Export Citation Format

Share Document