• KIPS Transactions on Computer and Communication Systems
• /
• v.5 no.12
• /
• pp.471-480
• /
• 2016

Recently, the damage of cyber attack toward infra-system, national defence and security system is gradually increasing. In this situation, military recognizes the importance of cyber warfare, and they establish a cyber system in preparation, regardless of the existence of threaten. Thus, the study of Intrusion Detection System(IDS) that plays an important role in network defence system is required. IDS is divided into misuse and anomaly detection methods. Recent studies attempt to combine those two methods to maximize advantagesand to minimize disadvantages both of misuse and anomaly. The combination is called Hybrid IDS. Previous studies would not be inappropriate for near real-time network environments because they have computational complexity problems. It leads to the need of the study considering the structure of IDS that have high detection rate and low computational cost. In this paper, we proposed a Hybrid IDS which combines C4.5 decision tree(misuse detection method) and Weighted K-means algorithm (anomaly detection method) hierarchically. It can detect malicious network packets effectively with low complexity by applying mutual information and genetic algorithm based efficient feature selection technique. Also we construct upgraded the the hierarchical structure of IDS reusing feature weights in anomaly detection section. It is validated that proposed Hybrid IDS ensures high detection accuracy (98.68%) and performance at experiment section.

• Information Systems Review
• /
• v.25 no.1
• /
• pp.129-143
• /
• 2023

In the era of big data, the field of artificial intelligence is showing remarkable growth, and in particular, the image classification learning methods by deep learning are becoming an important area. Various studies have been actively conducted to further improve the performance of CNNs, which have been widely used in image classification, among which a representative method is the Convolutional Recurrent Neural Network (CRNN) algorithm. The CRNN algorithm consists of a combination of CNN for image classification and RNNs for recognizing time series elements. However, since the inputs used in the RNN area of CRNN are the flatten values extracted by applying the convolution and pooling technique to the image, pixel values in the same phase in the image appear in different order. And this makes it difficult to properly learn the sequence of arrangements in the image intended by the RNN. Therefore, this study aims to improve image classification performance by proposing a novel hybrid method of CNN and RNN applying the concepts of encoder and decoder. In this study, the effectiveness of the new hybrid method was verified through various experiments. This study has academic implications in that it broadens the applicability of encoder and decoder concepts, and the proposed method has advantages in terms of model learning time and infrastructure construction costs as it does not significantly increase complexity compared to conventional hybrid methods. In addition, this study has practical implications in that it presents the possibility of improving the quality of services provided in various fields that require accurate image classification.

• Korean Journal of Agricultural Science
• /
• v.14 no.1
• /
• pp.98-133
• /
• 1987

The mechanical and rheological properties of agricultural materials are important for engineering design and analysis of their mechanical harvesting, handling, transporting and processing systems. Agricultural materials, which composed of structural members and fluids do not react in a purely elastic manner, and their response when subjected to stress and strain is a combination of elastic and viscous behavior so called viscoelastic behavior. Many researchers have conducted studies on the mechanical and rheological properties of the various agricultural products, but a few researcher has studied those properties of rice plant, and also those data are available only for foreign varieties of rice plant. This study are conducted to experimentally determine the mechanical and the rheological properties such as axial compressive strength, tensile strength, bending and shear strength, stress relaxation and creep behavior of rice stems, and grain detachment strength. The rheological models for the rice stem were developed from the test data. The shearing characteristics were examined at some different levels of portion, cross-sectional area, moisture content of rice stem and shearing angle. The results obtained from this study were summarized as follows 1. The mechanical properties of the stems of the J aponica types were greater than those of the Indica ${\times}$ Japonica hybrid in compression, tension, bendingand shearing. 2. The mean value of the compressive force was 80.5 N in the Japonica types and 55.5 N in the Indica ${\times}$ Japonica hybrid which was about 70 percent to that of the Japonica types, and then the value increased progressively at the lower portion of the stems generally. 3. The average tensile force was about 226.6 N in the Japonica types and 123.6 N in the Indica ${\times}$ Japonica hybrid which was about 55 percent to that of the Japonica types. 4. The bending moment was $0.19N{\cdot}m$ in the Japonica types and $0.13N{\cdot}m$ in the Indica ${\times}$ Japonica hybrid which was 68 percent to that of the Japonica types and the bending strength was 7.7 MPa in the Japonica types and 6.5 MPa in the Indica ${\times}$ Japonica hybrid respectively. 5. The shearing force was 141.1 N in Jinju, the Japonica type and 101.4 N in Taebaeg, the Indica ${\times}$ Japonica hybrid which was 72 percent to that of Jinju, and the shearing strength of Taebaeg was 63 percent to that of Jinju. 6. The shearing force and the shearing energy along the stem portion in Jinju increased progressively together at the lower portions, meanwhile in Taebaeg the shearing force showed the maximum value at the intermediate portion and the shearing energy was the greatest at the portion of 21 cm from the ground level, and also the shearing strength and the shearing energy per unit cross-sectional area of the stem were the greater values at the intermediate portion than at any other portions. 7. The shearing force and the shearing energy increased with increase of the cross-sectional area of the rice stem and with decrease of the shearing angie from $90^{\circ}$ to $50^{\circ}$. 8. The shearing forces showed the minimum values of 110 N at Jinju and of 60 N at Taebaeg, the shearing energy at the moisture content decreased about 15 percent point from initial moisture content showed value of 50 mJ in Jinju and of 30 mJ in Taebaeg, respectively. 9. The stress relaxation behavior could be described by the generalized Maxwell model and also the compression creep behavior by Burger's model, respectively in the rice stem. 10. With increase of loading rate, the stress relaxation intensity increased, meanwhile the relaxation time and residual stress decreased. 11. In the compression creep test, the logarithmic creep occured at the stress less than 2.0 MPa and the steady-state creep at the stress larger than 2.0 MPa. 12. The stress level had not a significant effect on the relaxation time, while the relaxation intensity and residual stress increased with increase of the stress level. 13. In the compression creep test of the rice stem, the instantaneous elastic modulus of Burger's model showed the range of 60 to 80 MPa and the viscosities of the free dashpot were very large numerical value which was well explained that the rice stem was viscoelastic material. 14. The tensile detachment forces were about 1.7 to 2.3 N in the Japonica types while about 1.0 to 1.3 N in Indica ${\times}$ Japonica hybrid corresponding to 58 percent of Japonica types, and the bending detachment forces were about 0.6 to 1.1 N corresponding to 30 to 50 percent of the tensile detachment forces, and the bending detachment of the Indica ${\times}$ Japonica hybrid was 0.1 to 0.3 N which was 7 to 21 percent of Japonica types. 15. The detachment force of the lower portion was little bigger than that of the upper portion in a penicle and was not significantly affected by the harvesting period from September 28 to October 20. 16. The tensile and bending detachment forces decreased with decrease of the moisture content from 23 to 13 percent (w.b.) by the natural drying, and the decreasing rate of detachment forces along the moisture content was the greater in the bending detachment force than the tensile detachment force.

• Journal of Intelligence and Information Systems
• /
• v.19 no.2
• /
• pp.55-71
• /
• 2013

The prediction model is the main factor affecting the performance of a knowledge-based system for bankruptcy prediction. Earlier studies on prediction modeling have focused on the building of a single best model using statistical and artificial intelligence techniques. However, since the mid-1980s, integration of multiple techniques (hybrid techniques) and, by extension, combinations of the outputs of several models (ensemble techniques) have, according to the experimental results, generally outperformed individual models. An ensemble is a technique that constructs a set of multiple models, combines their outputs, and produces one final prediction. The way in which the outputs of ensemble members are combined is one of the important issues affecting prediction accuracy. A variety of combination schemes have been proposed in order to improve prediction performance in ensembles. Each combination scheme has advantages and limitations, and can be influenced by domain and circumstance. Accordingly, decisions on the most appropriate combination scheme in a given domain and contingency are very difficult. This paper proposes a confidence-based selection approach as part of an ensemble bankruptcy-prediction scheme that can measure unified confidence, even if ensemble members produce different types of continuous-valued outputs. The present experimental results show that when varying the number of models to combine, according to the creation type of ensemble members, the proposed combination method offers the best performance in the ensemble having the largest number of models, even when compared with the methods most often employed in bankruptcy prediction.

• Earthquakes and Structures
• /
• v.12 no.3
• /
• pp.321-332
• /
• 2017

The reverse fault is a dangerous geological hazard faced by buried steel pipelines. Permanent ground deformation along the fault trace will induce large compressive strain leading to buckling failure of the pipe. A hybrid pipe-shell element based numerical model programed by INP code supported by ABAQUS solver was proposed in this study to explore the strain performance of buried X80 steel pipeline under reverse fault displacement. Accuracy of the numerical model was validated by previous full scale experimental results. Based on this model, parametric analysis was conducted to study the effects of four main kinds of parameters, e.g., pipe parameters, fault parameters, load parameter and soil property parameters, on the strain demand. Based on 2340 peak strain results of various combinations of design parameters, a semi-empirical model for strain demand prediction of X80 pipeline at reverse fault crossings was proposed. In general, reverse faults encountered by pipelines are involved in 3D oblique reverse faults, which can be considered as a combination of reverse fault and strike-slip fault. So a compressive strain demand estimation procedure for X80 pipeline crossing oblique-reverse faults was proposed by combining the presented semi-empirical model and the previous one for compression strike-slip fault (Liu 2016). Accuracy and efficiency of this proposed method was validated by fifteen design cases faced by the Second West to East Gas pipeline. The proposed method can be directly applied to the strain based design of X80 steel pipeline crossing oblique-reverse faults, with much higher efficiency than common numerical models.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.11 no.1
• /
• pp.29-37
• /
• 2006

Precise control of interior permanent magnet synchronous motor(IPMSM) over wide speed range is an engineering challenge. This paper considers the design and implementation of novel technique of high performance speed control for IPMSM using learning mechanism-fuzzy neural network(LM-FNN) and ANN(artificial neural network) control. The hybrid combination of neural network and fuzzy control will produce a powerful representation flexibility md numerical processing capability. Also, this paper proposes speed control of IPMSM using LM-FNN and estimation of speed using artificial neural network controller. The back propagation neural network technique is used to provide a real time adaptive estimation of the motor speed. 'The error between the desired state variable and the actual one is back-propagated to adjust the rotor speed, so that the actual state variable will coincide with the desired one. The back propagation mechanism is easy to derive and the estimated speed tracks precisely the actual motor speed. Analysis results to verify the effectiveness of the new hybrid intelligent control proposed in this paper.

• Radiation Oncology Journal
• /
• v.3 no.2
• /
• pp.153-157
• /
• 1985

Radiation effect on mammals, especially on hematologic changes, has been studied since discovery of x-ray. Various experimental animals were tried for radiobiological studies. 72hybrid mice with conventional faction $(15\times/week)$, large abdominal field $12\times3cm$, from symphysis pubic to xyphoid process) were used. Body weight was declined gradually by increasing irradiation doses, nadir was about $29.7\%$ in male; $30.4\%$ in female at 6000 rad irradiation group. Hemoglobin value was nearly normal throughout entire treatment. Significant dropping of WBC count was noted to $40-50\%$ of pretreatment values by only 1000rad irradiation. Change of differential count was interesting; : lymphocyte proportion showed gradual reduction, instead of gradual increasing of segmented neutrophil. Those proportion were reversed after 6000 rad irradiation. Urinary protein tests showed + - +++, showing no correlation with dosage. Application of our study in clinical combination therapy (radiation + chemotherapy) was discussed.

• Journal of Korea Multimedia Society
• /
• v.6 no.1
• /
• pp.48-57
• /
• 2003

This paper describes an approach for classifying myoelectric patterns using a multilayer perceptrons (MLP's) with genetic algorithm and hidden Markov models (HMM's) hybrid classifier. Genetic Algorithms play a role of selecting Multilayer Perceptron's optimized initial connection weights by its typical global search. The dynamic aspects of EMG are important for tasks such as continuous prosthetic control or various time length EMG signal recognition, which have not been successfully mastered by the most neural approaches. It is known that the hidden Markov model (HMM) is suitable for modeling temporal patterns. In contrast, the multilayer feedforward networks are suitable for static patterns. And, a lot of investigators have shown that the HMM's to be an excellent tool for handling the dynamical problems. Considering these facts, we suggest the combination of ANN and HMM algorithms that might lead to further improved EMG recognition systems.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.1
• /
• pp.793-800
• /
• 2015

This study is to analyze correlation between Daniel Libeskind's architecture works and the effects as a series of music works be learning in youth and opera "Aron and Mose". The results showed that First, Libeskind created convergence of invisible line by borrowing the composition of freely flowed scales in score from his architecture. Second, he composed geometrical shapes of contrapuntal reiteration based on double tune in music structure, in other words forms of polyphonic proportion. and he expressed the geometrically, freely line rhythm by planning composition of multidimensional spaces, "Hybrid", planning the contrast of material, form by results of Intertextually combination between Architecture and Music. Third, he tried to express the pain, fear, anxiety, etc. of the past spatially, and constructed "the spaces of absence" on his works through inspiration from Arnold Sch$\ddot{o}$nberg's works.

• Structural Monitoring and Maintenance
• /
• v.5 no.2
• /
• pp.205-229
• /
• 2018

Energy induced by minor earthquake and micro vibration cannot be dissipated by traditional buckling-restrained braces (BRBs). To solve this problem, a new type of hybrid passive control device, named as VE-BRB, which is configured by a BRB with high-damping viscoelastic (VE) layers, is developed and studied. Theoretical analysis, performance tests, numerical simulation and case analysis are conducted to study the seismic behavior of VE-BRBs. The results indicate that the combination of hysteretic and damping devices lead to a multi-phased nature and good performance. VE-BRB's working state can be divided into three phases: before yielding of the steel core, VE layers provide sufficient damping ratio to mitigate minor vibrations; after yielding of the steel core, the steel's hysteretic deformations provide supplemental dissipative capacity for structures; after rupture of the steel core, VE layers are still able to work normally and provide multiple security assurance for structures. The simulation results agreed well with the experimental results, validating the finite element analysis method, constitutive models and the identified parameters. The comparison of the time history analysis on a 6-story frame with VE-BRBs and BRBs verified the advantages of VE-BRB for seismic protection of structures compared with traditional BRB. In general, VE-BRB had the potential to provide better control effect on structural displacement and shear in all stages than BRB as expected.