A Wavelet Transform Based Support Vector Machine Ensemble Algorithm and Its Application in Network Intrusion Detection

Xuesen Cai,Fanhua Yu 보안공학연구지원센터 2015 International Journal of Security and Its Applicat Vol.9 No.4

Traditional network intrusion detection algorithms are time consuming due to the existence of redundant attributes. In order to improve the efficiency of network intrusion detection, in this paper, we propose a wavelet transform based support vector machine ensemble algorithm. Firstly, we use wavelet transform to remove the redundant attributes from the original dataset. Then we train a support vector machine ensemble on the simplified dataset. As the wavelet transform in this algorithm can effectively remove the redundant attributes, the proposed algorithm is with high efficiency. Simulation experiments on KDD CUP 99 data set show that the proposed algorithm has good intrusion detection performance.

Foreign Detection based on Wavelet Transform Algorithm with Image Analysis Mechanism in The Inner Wall of The Tube

Jinlong Zhu,Fanhua Yu,Mingyu Sun,Dong Zhao,Qingtian Geng 한국정보처리학회 2019 Journal of information processing systems Vol.15 No.1

A method for detecting foreign substances in mould based on scatter grams was presented to protect mouldsautomatically during moulding production. This paper proposes a wavelet transform foreign detection methodbased on Monte Carlo analysis mechanism to identify foreign objects in the tube. We use the Monte Carlomethod to evaluate the image, and obtain the width of the confidence interval by the deviation statistical grayhistogram to divide the image type. In order to stabilize the performance of the algorithm, the high-frequencyimage and the low-frequency image are respectively drawn. By analyzing the position distribution of the pixelgray in the two images, the suspected foreign object region is obtained. The experiments demonstrate theeffectiveness of our approach by evaluating the labeled data.

Foreign Detection Based on Wavelet Transform Algorithm with Image Analysis Mechanism in the Inner Wall of the Tube

Zhu, Jinlong,Yu, Fanhua,Sun, Mingyu,Zhao, Dong,Geng, Qingtian Korea Information Processing Society 2019 Journal of information processing systems Vol.15 No.1

A method for detecting foreign substances in mould based on scatter grams was presented to protect moulds automatically during moulding production. This paper proposes a wavelet transform foreign detection method based on Monte Carlo analysis mechanism to identify foreign objects in the tube. We use the Monte Carlo method to evaluate the image, and obtain the width of the confidence interval by the deviation statistical gray histogram to divide the image type. In order to stabilize the performance of the algorithm, the high-frequency image and the low-frequency image are respectively drawn. By analyzing the position distribution of the pixel gray in the two images, the suspected foreign object region is obtained. The experiments demonstrate the effectiveness of our approach by evaluating the labeled data.

Classroom Roll-Call System Based on ResNet Networks

Jinlong Zhu,Fanhua Yu,Guangjie Liu,Mingyu Sun,Dong Zhao,Qingtian Geng,Jinbo Su 한국정보처리학회 2020 Journal of information processing systems Vol.16 No.5

A convolution neural networks (CNNs) has demonstrated outstanding performance compared to otheralgorithms in the field of face recognition. Regarding the over-fitting problem of CNN, researchers haveproposed a residual network to ease the training for recognition accuracy improvement. In this study, a novelface recognition model based on game theory for call-over in the classroom was proposed. In the proposedscheme, an image with multiple faces was used as input, and the residual network identified each face with aconfidence score to form a list of student identities. Face tracking of the same identity or low confidence weredetermined to be the optimisation objective, with the game participants set formed from the student identitylist. Game theory optimises the authentication strategy according to the confidence value and identity set toimprove recognition accuracy. We observed that there exists an optimal mapping relation between face andidentity to avoid multiple faces associated with one identity in the proposed scheme and that the proposedgame-based scheme can reduce the error rate, as compared to the existing schemes with deeper neural network.

Directional crossover slime mould algorithm with adaptive Lévy diversity for the optimal design of real-world problems

Qi Ailiang,Zhao Dong,Yu Fanhua,Liu Guangjie,Heidari Ali Asghar,Chen Huiling,Algarni Abeer D.,Elmannai Hela,Gui Wenyong 한국CDE학회 2022 Journal of computational design and engineering Vol.9 No.6

The slime mould algorithm (SMA) has become a classical algorithm applied in many fields since it was presented. Nevertheless, when faced with complex tasks, the algorithm converges slowly and tends to fall into the local optimum. So, there is still room for improvement in the performance of SMA. This work proposes a novel SMA variant (SDSMA), combining the adaptive Lévy diversity mechanism and directional crossover mechanism. Firstly, the adaptive Lévy diversity mechanism can improve population diversity. Then, the directional crossover mechanism can enhance the balance of exploration and exploitation, thus helping SDSMA to increase the convergence speed and accuracy. SDSMA is compared with SMA variants, original algorithms, improved algorithms, improved-SMAs, and others on the benchmark function set to verify its performance. Meanwhile, the Wilcoxon signed-rank test, the Friedman test, and other analytical methods are considered to analyze the experimental results. The analysis results show that SDSMA with two strategies significantly improves the performance of SMA. Meanwhile, the computational cost of SDSMA is smaller than that of SMA on benchmark function. Finally, the proposed algorithm is applied to three real-world engineering design problems. The experiments prove that SDSMA is an effective aid tool for computationally complex practical tasks.

Opposition-based ant colony optimization with all-dimension neighborhood search for engineering design

Zhao Dong,Liu Lei,Yu Fanhua,Heidari Ali Asghar,Wang Maofa,Chen Huiling,무함마드칸 한국CDE학회 2022 Journal of computational design and engineering Vol.9 No.3

The ant colony optimization algorithm is a classical swarm intelligence algorithm, but it cannot be used for continuous class optimization problems. A continuous ant colony optimization algorithm (ACOR) is proposed to overcome this difficulty. Still, some problems exist, such as quickly falling into local optimum, slow convergence speed, and low convergence accuracy. To solve these problems, this paper proposes a modified version of ACOR called ADNOLACO. There is an opposition-based learning mechanism introduced into ACOR to effectively improve the convergence speed of ACOR. All-dimension neighborhood mechanism is also introduced into ACOR to further enhance the ability of ACOR to avoid getting trapped in the local optimum. To strongly demonstrate these core advantages of ADNOLACO, with the 30 benchmark functions of IEEE CEC2017 as the basis, a detailed analysis of ADNOLACO and ACOR is not only qualitatively performed, but also a comparison experiment is conducted between ADNOLACO and its peers. The results fully proved that ADNOLACO has accelerated the convergence speed and improved the convergence accuracy. The ability to find a balance between local and globally optimal solutions is improved. Also, to show that ADNOLACO has some practical value in real applications, it deals with four engineering problems. The simulation results also illustrate that ADNOLACO can improve the accuracy of the computational results. Therefore, it can be demonstrated that the proposed ADNOLACO is a promising and excellent algorithm based on the results.

Directional mutation and crossover for immature performance of whale algorithm with application to engineering optimization

Qi Ailiang,Zhao Dong,Yu Fanhua,Heidari Ali Asghar,Chen Huiling,Xiao Lei 한국CDE학회 2022 Journal of computational design and engineering Vol.9 No.2

In recent years, a range of novel and pseudonovel optimization algorithms has been proposed for solving engineering problems. Swarm intelligence optimization algorithms (SIAs) have become popular methods, and the whale optimization algorithm (WOA) is one of the highly discussed SIAs. However, regardless of novelty concerns about this method, the basic WOA is a weak method compared to top differential evolutions and particle swarm variants, and it suffers from the problem of poor initial population quality and slow convergence speed. Accordingly, in this paper, to increase the diversity of WOA versions and enhance the performance of WOA, a new WOA variant, named LXMWOA, is proposed, and based on the Lévy initialization strategy, the directional crossover mechanism, and the directional mutation mechanism. Specifically, the introduction of the Lévy initialization strategy allows initial populations to be dynamically distributed in the search space and enhances the global search capability of the WOA. Meanwhile, the directional crossover mechanism and the directional mutation mechanism can improve the local exploitation capability of the WOA. To evaluate its performance, using a series of functions and three models of engineering optimization problems, the LXMWOA was compared with a broad array of competitive optimizers. The experimental results demonstrate that the LXMWOA is significantly superior to its exploration and exploitation capability peers. Therefore, the proposed LXMWOA has great potential to be used for solving engineering problems.

A horizontal and vertical crossover cuckoo search: optimizing performance for the engineering problems

Su Hang,Zhao Dong,Yu Fanhua,Heidari Ali Asghar,Xu Zhangze,Alotaibi Fahd S,Mafarja Majdi,Chen Huiling 한국CDE학회 2023 Journal of computational design and engineering Vol.10 No.1

As science and technology advance, more engineering-type problems emerge. Technology development has likewise led to an increase in the complexity of optimization problems, and the need for new optimization techniques has increased. The swarm intelligence optimization algorithm is popular among researchers as a flexible, gradient-independent optimization method. The cuckoo search (CS) algorithm in the population intelligence algorithm has been widely used in various fields as a classical optimization algorithm. However, the current CS algorithm can no longer satisfy the performance requirements of the algorithm for current optimization problems. Therefore, in this paper, an improved CS algorithm based on a crossover optimizer (CC) and decentralized foraging (F) strategy is proposed to improve the search ability and the ability to jump out of the local optimum of the CS algorithm (CCFCS). Then, in order to verify the performance of the algorithm, this paper demonstrates the performance of CCFCS from six perspectives: core parameter setting, balance analysis of search and exploitation, the impact of introduced strategies, the impact of population dimension, and comparison with classical algorithms and similar improved algorithms. Finally, the optimization effect of CCFCS on real engineering problems is tested by five classic cases of engineering optimization. According to the experimental results, CCFCS has faster convergence and higher solution quality in the algorithm performance test and maintains the same excellent performance in engineering applications.