• Journal of Communications and Networks
• /
• v.18 no.5
• /
• pp.818-825
• /
• 2016

Network management and anomaly detection are challenges in high-speed networks due to the high volume of packets that has to be analysed. Flow-based analysis is a scalable method which reduces the high volume of network traffic by dividing it into flows. As sampling methods are extensively used in flow generators such as NetFlow, the impact of sampling on the performance of flow-based analysis needs to be investigated. Monitoring using sampled traffic is a well-studied research area, however, the impact of sampling on flow-based anomaly detection is a poorly researched area. This paper investigates flow sampling methods and shows that these methods have negative impact on flow-based anomaly detection. Therefore, we propose an efficient probabilistic flow sampling method that can preserve flow traffic distribution. The proposed sampling method takes into account two flow features: Destination IP address and octet. The destination IP addresses are sampled based on the number of received bytes. Our method provides efficient sampled traffic which has the required traffic features for both flow-based anomaly detection and monitoring. The proposed sampling method is evaluated using a number of generated flow-based datasets. The results show improvement in preserved malicious flows.

• Structural Engineering and Mechanics
• /
• v.82 no.3
• /
• pp.271-282
• /
• 2022

The active learning surrogate model based on adaptive sampling strategy is increasingly popular in reliability analysis. However, most of the existing sampling strategies adopt the trial and error method to determine the size of the Monte Carlo (MC) candidate sample pool which satisfies the requirement of variation coefficient of failure probability. It will lead to a reduction in the calculation efficiency of reliability analysis. To avoid this defect, a new method for determining the optimal size of the MC candidate sample pool is proposed, and a new structural reliability analysis method combining polynomial chaos-based Kriging model (PC-Kriging) with adaptive sampling region is also proposed (PCK-ASR). Firstly, based on the lower limit of the confidence interval, a new method for estimating the optimal size of the MC candidate sample pool is proposed. Secondly, based on the upper limit of the confidence interval, an adaptive sampling region strategy similar to the radial centralized sampling method is developed. Then, the k-means++ clustering technique and the learning function LIF are used to complete the adaptive design of experiments (DoE). Finally, the effectiveness and accuracy of the PCK-ASR method are verified by three numerical examples and one practical engineering example.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.23 no.4
• /
• pp.341-347
• /
• 2013

Objectives: The aim of this study is to investigate the differences in airborne TDI concentrations based on the filter collection method and liquid collection method and to compare airborne TDIs concentrations by sampling method change when using the filter collection method in the spray-painting process. Methods: For the sample measurement, the filter collection method(OSHA#42) and liquid collection method(NIOSH#5522) were used; for the sampling method, the full-period single sampling and full-period consecutive sampling methods were used. The samples were collected in spray-painting and drying process locations. Results: In all samples collected from the spray-painting and drying process locations through the filter collection and liquid collection methods, greater amounts of 2,6-TDI than 2,4-TDI were detected. When the TDI collection concentrations based on the sampling methods were compared, the concentrations of 2,4-TDI and 2,6-TDI collected by the consecutive sampling method were higher than the concentrations of 2,4-TDI and 2,6-TDI collected by the single sampling method for both the filter collection method and liquid collection method used in the spray-painting process. These differences were statistically significant. Conclusions: When TDI collection concentrations based on the sample measurement methods were compared, the concentration of 2,4-TDI and 2,6-TDI collected through the liquid collection method were higher than the concentrations of 2,4-TDI and 2,6-TDI collected by the filter collection method, and the differences were statistically significant. In the drying process, no difference was shown in the collection concentrations of 2,4-TDI and 2,6-TDI with the two measurement methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.4
• /
• pp.1110-1127
• /
• 2022

To improve the training efficiency and generalization performance of a support vector machine (SVM) in a large-scale set, an optimal SVM learning method based on adaptive sparse sampling and the granularity shift factor is presented. The proposed method combines sampling optimization with learner optimization. First, an adaptive sparse sampling method based on the potential function density clustering is designed to adaptively obtain sparse sampling samples, which can achieve a reduction in the training sample set and effectively approximate the spatial structure distribution of the original sample set. A granularity shift factor method is then constructed to optimize the SVM decision hyperplane, which fully considers the neighborhood information of each granularity region in the sparse sampling set. Experiments on an artificial dataset and three benchmark datasets show that the proposed method can achieve a relatively higher training efficiency, as well as ensure a good generalization performance of the learner. Finally, the effectiveness of the proposed method is verified.

• Structural Engineering and Mechanics
• /
• v.92 no.1
• /
• pp.111-120
• /
• 2024

Sampling methods are powerful approaches to solving the problems of structural reliability analysis and estimating the failure probability of structures. In this paper, a new sampling method is proposed offering lower variance and lower computational cost for complex and high-dimensional problems. The method is called Optimal Latinized partially stratified sampling (OLPSS) as it is based upon the Latinized Partially Stratified Sampling (LPSS) which itself is based on merging Stratified Sampling (SS) and Latin Hypercube Sampling (LHS) algorithms. While LPSS has a low variance, it may suffer from a lack of good space-filling of its generated samples in some cases. In the OLPSS, this issue has been resolved by employing a new columnwise-pairwise exchange optimization procedure for sample generation. The efficiency of the OLPSS has been tested and reported under several benchmark mathematical functions and structural examples including structures with a large number of variables (e.g., a structure with 67 variables). The proposed method provides highly accurate estimates of the failure probability of structures with a significantly lower variance relative to the Monte Carlo simulations, Latin Hypercube, and standard LPSS.

• Structural Engineering and Mechanics
• /
• v.38 no.1
• /
• pp.125-140
• /
• 2011

This study presents an innovative method to estimate the reliability sensitivity based on the low-discrepancy sampling which is a new technique for structural reliability analysis. Two advantages are contributed to the method: one is that, by developing a general importance sampling procedure for reliability sensitivity analysis, the partial derivative of the failure probability with respect to the distribution parameter can be directly obtained with typically insignificant additional computations on the basis of structural reliability analysis; and the other is that, by combining various low-discrepancy sequences with the above importance sampling procedure, the proposed method is far more efficient than that based on the classical Monte Carlo method in estimating reliability sensitivity, especially for problems of small failure probability or problems that require a large number of costly finite element analyses. Examples involving both numerical and structural problems illustrate the application and effectiveness of the method developed, which indicate that the proposed method can provide accurate and computationally efficient estimates of reliability sensitivity.

• Structural Engineering and Mechanics
• /
• v.32 no.1
• /
• pp.55-69
• /
• 2009

A quasi ideal importance sampling simulation method combined in the conditional expectation is proposed for the structural reliability estimation. The quasi ideal importance sampling joint probability density function (p.d.f.) is so composed on the basis of the ideal importance sampling concept as to be proportional to the conditional failure probability multiplied by the p.d.f. of the sampling variables. The respective marginal p.d.f.s of the ideal importance sampling joint p.d.f. are determined numerically by the simulations and partly by the piecewise integrations. The quasi ideal importance sampling simulations combined in the conditional expectation are executed to estimate the failure probabilities of structures with multiple failure surfaces and it is shown that the proposed method gives accurate estimations efficiently.

• Communications for Statistical Applications and Methods
• /
• v.14 no.3
• /
• pp.667-677
• /
• 2007

In this paper, we propose a method for estimating the mean of a population which has a linear trend, when both n, the sample size, and k, the reciprocal of the sampling fraction, are odd numbers. The proposed method, not having the drawbacks of centered systematic sampling, centered modified sampling and centered balanced sampling, consists of selecting a sample by balanced systematic sampling and estimating the population mean by using interpolation. We compare the efficiency of the proposed method and existing methods under the criterion of the expected mean square error based on the infinite superpopulation model.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.33 no.12
• /
• pp.1464-1470
• /
• 2009

RBDO approach based on a sampling method with the Kriging metamodel and Constraint Boundary Sampling (CBS), which is sequential sampling method to generate metamodels is proposed. The major advantage of the proposed RBDO approach is that it does not require Most Probable failure Point (MPP) which is essential for First-Order Reliability Method (FORM)-based RBDO approach. The Monte Carlo Sampling (MCS), most well-known method of the sampling methods for the reliability analysis is used to assess the reliability of constraints. In addition, a Cumulative Distribution Function (CDF) of the constraints is approximated using Moving Least Square (MLS) method from empirical distribution function. It is possible to acquire a probability of failure and its analytic sensitivities by using an approximate function of the CDF for the constraints. Moreover, a concept of inactive design is adapted to improve a numerical efficiency of the proposed approach. Computational accuracy and efficiency of the proposed RBDO approach are demonstrated by numerical and engineering problems.

• Structural Engineering and Mechanics
• /
• v.68 no.5
• /
• pp.549-561
• /
• 2018

Sobol method is applied as a powerful variance decomposition technique in the field of global sensitivity analysis (GSA). The paper is devoted to increase convergence speed of the extracted Sobol indices using a new proposed sampling technique called genetic based Latine hypercube sampling (GBLHS). This technique is indeed an improved version of restricted Latine hypercube sampling (LHS) and the optimization algorithm is inspired from genetic algorithm in a new approach. The new approach is based on the optimization of minimax value of LHS arrays using manipulation of array indices as chromosomes in genetic algorithm. The improved Sobol method is implemented to perform factor prioritization and fixing of an uncertain comprehensive high speed rotor-bearing system. The finite element method is employed for rotor-bearing modeling by considering Eshleman-Eubanks assumption and interaction of axial force on the rotor whirling behavior. The performance of the GBLHS technique are compared with the Monte Carlo Simulation (MCS), LHS and Optimized LHS (Minimax. criteria). Comparison of the GBLHS with other techniques demonstrates its capability for increasing convergence speed of the sensitivity indices and improving computational time of the GSA.