• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.157-173
• /
• 2011

As the Internet use explodes recently, the malicious attacks and hacking for a system connected to network occur frequently. This means the fatal damage can be caused by these intrusions in the government agency, public office, and company operating various systems. For such reasons, there are growing interests and demand about the intrusion detection systems (IDS)-the security systems for detecting, identifying and responding to unauthorized or abnormal activities appropriately. The intrusion detection models that have been applied in conventional IDS are generally designed by modeling the experts' implicit knowledge on the network intrusions or the hackers' abnormal behaviors. These kinds of intrusion detection models perform well under the normal situations. However, they show poor performance when they meet a new or unknown pattern of the network attacks. For this reason, several recent studies try to adopt various artificial intelligence techniques, which can proactively respond to the unknown threats. Especially, artificial neural networks (ANNs) have popularly been applied in the prior studies because of its superior prediction accuracy. However, ANNs have some intrinsic limitations such as the risk of overfitting, the requirement of the large sample size, and the lack of understanding the prediction process (i.e. black box theory). As a result, the most recent studies on IDS have started to adopt support vector machine (SVM), the classification technique that is more stable and powerful compared to ANNs. SVM is known as a relatively high predictive power and generalization capability. Under this background, this study proposes a novel intelligent intrusion detection model that uses SVM as the classification model in order to improve the predictive ability of IDS. Also, our model is designed to consider the asymmetric error cost by optimizing the classification threshold. Generally, there are two common forms of errors in intrusion detection. The first error type is the False-Positive Error (FPE). In the case of FPE, the wrong judgment on it may result in the unnecessary fixation. The second error type is the False-Negative Error (FNE) that mainly misjudges the malware of the program as normal. Compared to FPE, FNE is more fatal. Thus, when considering total cost of misclassification in IDS, it is more reasonable to assign heavier weights on FNE rather than FPE. Therefore, we designed our proposed intrusion detection model to optimize the classification threshold in order to minimize the total misclassification cost. In this case, conventional SVM cannot be applied because it is designed to generate discrete output (i.e. a class). To resolve this problem, we used the revised SVM technique proposed by Platt(2000), which is able to generate the probability estimate. To validate the practical applicability of our model, we applied it to the real-world dataset for network intrusion detection. The experimental dataset was collected from the IDS sensor of an official institution in Korea from January to June 2010. We collected 15,000 log data in total, and selected 1,000 samples from them by using random sampling method. In addition, the SVM model was compared with the logistic regression (LOGIT), decision trees (DT), and ANN to confirm the superiority of the proposed model. LOGIT and DT was experimented using PASW Statistics v18.0, and ANN was experimented using Neuroshell 4.0. For SVM, LIBSVM v2.90-a freeware for training SVM classifier-was used. Empirical results showed that our proposed model based on SVM outperformed all the other comparative models in detecting network intrusions from the accuracy perspective. They also showed that our model reduced the total misclassification cost compared to the ANN-based intrusion detection model. As a result, it is expected that the intrusion detection model proposed in this paper would not only enhance the performance of IDS, but also lead to better management of FNE.

• The Journal of Society for e-Business Studies
• /
• v.25 no.2
• /
• pp.15-27
• /
• 2020

This study starts with the idea that the process of creating a Gantt Chart for schedule planning is similar to Tetris game with only a straight line. In Tetris games, the X axis is M machines and the Y axis is time. It is assumed that all types of orders can be worked without separation in all machines, but if the types of orders are different, setup cost will be incurred without delay. In this study, the game described above was named Gantris and the game environment was implemented. The AI-scheduling table through in-depth reinforcement learning compares the real-time scheduling table with the human-made game schedule. In the comparative study, the learning environment was studied in single order list learning environment and random order list learning environment. The two systems to be compared in this study are four machines (Machine)-two types of system (4M2T) and ten machines-six types of system (10M6T). As a performance indicator of the generated schedule, a weighted sum of setup cost, makespan and idle time in processing 100 orders were scheduled. As a result of the comparative study, in 4M2T system, regardless of the learning environment, the learned system generated schedule plan with better performance index than the experimenter. In the case of 10M6T system, the AI system generated a schedule of better performance indicators than the experimenter in a single learning environment, but showed a bad performance index than the experimenter in random learning environment. However, in comparing the number of job changes, the learning system showed better results than those of the 4M2T and 10M6T, showing excellent scheduling performance.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.29 no.1
• /
• pp.76-83
• /
• 1984

This experiment was performed to investigate the flowering habit of sesame (Sesamum indicum L.). Sesame varieties tested could be classified into 8 different plant types by their morphological traits such as capsule shape, capsule setting habit and branching types among sesame gene pool of Crop Experiment Station, ORD. The first flower was appeared at the lowest node on main stem. Flowers were appeared progressively toward the tip of the main stem and also toward the tips of branches. The interval of flowering for a node was about one day, but 3 to 8 days for the flowers on the tips. Side flowers started at 4 to 5 nodes lower than those of center flower at the same day. Flowers were beared 2 by 1 node on the middle part of flower setting node (7-9) in mono capsule setting habit in spite of its normal is 1 by 1 node on the other nodes. Flowers were beared opposite direction on each node of stem and flowering toward the tip of main stem composed of cross shape between nodes and spiral, reverse of clockwise direction. We called this habit as cross spiral flowering order and cross spiral phyllotaxis. The first flower on branches was appeared when center flower on the 5th node of main stem began to flower. The branches produced at higher nodes on main stem showed larger flowering periods and more number of flowers than that at lower parts. BTB (Branch, Tricapsule, Bicarpels, 4 Loculi) type showed three capsule setting habits and same flowering period both on main stem and branches while BTQ (Branch, Tricapsule, Quadricarpels, 8 Loculi) type showed three capsule setting habit on main stem and mono-capsule setting habit on branches. In BTQ type, the period of flowering was much shorter on branches than on main stem. Branching type was considered more promising than non branching type for the breeding of early maturing high yielding variety because branching type has the advantage of bearing a lot of flowers in comparatively short flowering period.

• Korean Journal of Veterinary Research
• /
• v.36 no.3
• /
• pp.577-598
• /
• 1996

This study was conducted to identify the effects of caffeine or combinations of caffeine and iron or vitamin E on the lipid and protein components or blood chemistry levels of the serum as well as the total homogenate, mitochondrial and microsomal fraction of the rat(Sprague-Dawley, female) liver. Chronic test were conducted to determine those effects. The chronic test was conducted by dividing rats into 5 groups according to the type of drugs and dosages administrated as follows; the control(group A), and group B was given 25mg/kg caffeine orally once daily for 30 days, group C was given 50mg/kg caffeine orally once daily for 30 days, group D was given 25mg/kg caffeine and orally ferric chloride once daily for 30 days and group E was given 25mg/kg caffeine and 25mg/kg vitamin E once daily for 30 days. The concentrations of glucose, urea nitrogen, uric acid, creatinine, total protein, albumin, A/G ratio, triglyceride, total cholesterol, HDL-cholesterol, free fatty acid, phospholipid as well as the activities of alanine aminotransferase(ALT), aspartate aminotransferase(AST) and alkaline phosphatase(ALP) were measured in the serum of each experimental groups. The concentrations of the carbonyl group and malondiaidehyde(MDA) and the patterns of the SDS-PAGE(Sodium Dodecyl Sulfate - Polyacrylamide Gel Electrophoresis) and fatty acid compositions in free fatty acids and phospholipids were analyzed to determine the oxidative damages and metabolic changes on the lipid and protein components in the serum, and total homogenate, mitochondrial and microsomal fractions of the rat liver. The results obtained from this study were summarized as follows; 1. Body weights of groups B, C, D and E were significantly decreased(p < 0.01) in comparison with that of the control in the chronic test. 2. The concentrations of serum glucose in groups B(124.5mg/dl), C(130.1mg/dl), D(122.1mg/dl), E(119.3mg/dl) were significantly higher(p < 0.01) in comparison to that of the control(101.5mg/dl). But, there were no significant differences in the concentrations of urea nitrogen, uric acid, creatinine, total protein, albumin and A/G ratio in comparison to that of the control. 3. The concentrations of total cholesterol and HDL-cholesterol in serum of groups B(69.6, 53.4mg/dl), C(73.0, 56.3mg/dl), D(68.9, 51.1mg/dl) and E(68.2, 51.3mg/dl) were significantly higher(p < 0.01) in comparison to that of the control(52.6, 38.8mg/dl). On the other hand, the concentrations of triglyceride in serum of groups B(45.0mg/dl), C(40.4mg/dl), D(33.8mg/dl) and E(47.2mg/dl) were significantly lower(p < 0.01) in comparison to that of the control(66.2mg/dl). There were no significant differences in the activities of ALT, AST and ALP in comparison to that of the control. 4. The concentrations of free fatty acid and phospholipid in serum of groups B(45.7, 154.4mg/dl), C(50.0, 167.2mg/dl), D(52.5, 148.4mg/dl) and E(41.1, 159.2mg/dl) were higher(p < 0.01) in comparison to that of the control(35.2, 125.3mg/dl). And the concentrations of the carbonyl group and malondialdehyde in serum of group D(1.82, 0.52nM/mg protein) were significantly higher(p < 0.01) in comparison to the control(1.53nM/mg protein). 5. The concentrations of carbonyl group in total homogenate, mitochondrial and microsomal fraction of group D(1.45, 0.94, 1.67nM/mg protein) were significantly higher (p < 0.01) in comparison to the control(1.16, 0.66, 1.27nM/mg protein). And the concentrations of malondialdehyde in the total homogenate, mitochondrial and microsomal fraction of group D(6.70, 6.10, 1.36nM/mg protein) were significantly higher(p < 0.01) in comparison to the control(5.17, 3.64, 0.68nM/mg protein). 6. As the analytical results of the fatty acid compositions of free fatty acid in serum, the proportions of stearic acid and arachidonic acid of groups B(16.52, 12.62%), C(17.52, 15.18%), D(19.73, 13.47%) and E(17.62, 13.28%) were significantly higher(p < 0.01) in comparison to the control(14.75, 7.88%), but the proportions of oleic acid and linoleic acid of groups B(12.97, 32.59%), C(10.88, 31.23%), D(12.37, 30.66%) and E(11.95, 32.41%) were significantly lower(p < 0.01) in comparison to the control(16.44, 35.12%). Otherwise, as the results of the fatty acid compositions of phospholipid in serum, the proportions of stearic acid and arachidonic acid of groups B(39.37, 16.39%), C(40.63, 17.83%), D(42.73, 15.39%) and E(39.16, 15.70%) were significantly higher(p < 0.01) in comparison to the control(37.74, 14.24%), but the proportions of oleic acid and linoleic acid of groups B(4.03, 14.38%), C(3.54, 12.38%), D(4.52, 11.68%) and E(4.29, 13.64%) were significantly lower(p < 0.01) in comparison to the control(5.53, 16.14%). 7. As the analytical results of the fatty acid compositions of free fatty acid in total homogenate, mitochondrial and microsomal fraction of liver, the proportions of oleic acid of groups B(7.8**, 8.73**, 6.88%) and C(6.89**, 7.75**, 6.58%) were lower(**:p < 0.01) in comparison to the control(8.67, 10.08, 7.81%), but the proportions of arachidonic acid of group C(22.62, 19.79, 23.71%) were significantly higher(p < 0.01) in comparison to the control(20.93, 18.47, 22.24%). And the proportions of palmitic acid of group D(25.95**, 26.16, 26.34**%) were significantly higher(**:p < 0.01) in comparison to the control(24.43, 25.42, 23.34%). In addition, the proportions of linoleic acid of group D(23.43, 25.02, 23.95%) were also significantly higher(p < 0.01) in comparison to the control(22.17, 23.75, 21.26%). The proportions of stearic acid of group D(19.87, 19.76**%) in mitochondrial and microsomal fraction were lower(**:p < 0.01) in comparison to the control(21.01, 24.18%), and the proportions of stearic acid of group E(16.71*, 19.65**%) in mitochondrial and microsomal fraction were significantly lower(**:p < 0.01, *:p < 0.05) in comparison to the control(21.01, 24.18%), and the proportions of linoleic acid of group E(25.04, 29.20, 26.48%) in total homogenate, mitochondria and microsome were significantly higher(p < 0.01) in comparison to the control(22.17, 23.75, 21.26%). 8. As the results of the fatty acid compositions of phospholipid in total homogenate, mitochondrial and microsomal fraction of liver, the proportions of palmitic acid of group D(17.58**, 18.78*, 18.23%**) were significantly higher(**:p < 0.01, *:p < 0.05) in comparison to the control(16.28, 17.22, 16.38%), and the proportions of stearic acid of group D(36.41, 37.23, 39.53%) were also significantly higher(p < 0.01) in comparison to the control(34.18, 34.16, 36.04%). But the proportions of oleic acid(3.41*, 3.11**, 3.12**%) and linoleic acid (18.03**, 15.79**, 14.74**%) of group D were significantly lower(**:p < 0.01, *:p < 0.05) in comparison to the control(oleic : 3.63, 3.72, 3.79%, linoleic : 20.03, 18.71, 18.48%). 9. In order to determine the oxidative damages to the protein in serum, mitochondrial and microsomal fraction of the rat liver, the patterns of the SDS-PAGE were identified, but the results of SDS-PAGE were not significantly different between the control and experimental groups.

• Journal of the Mineralogical Society of Korea
• /
• v.29 no.1
• /
• pp.1-10
• /
• 2016

We studied the heavy minerals in 46 surface sediments collected from the Central Yellow Sea Mud (CYSM) to characterize the type, abundance, mineralogical properties and distribution pattern using the stereo-microscopy, field-Emission scanning electron microscopy (FE SEM) and chemical analysis through the energy dispersive spectrometer (EDS). Heavy mineral assemblages are primarily composed of epidote group, amphibole group, garnet group, zircon, rutile and sphene in descending order. Epidote group and amphibole group minerals account for more than 50% of total heavy minerals. The minerals in epidote group, amphibole group and garnet group in studied area are epidote, edenite and almandine, respectively. When we divided the CYSM into two regions by $124^{\circ}E$, the eastern region contain higher contents of epidote and (zircon + rutile), which are more resistant to weathering but lower of amphibole, which is less resistant to weathering than the western region. Based on this results, it is possible to estimate that the eastern region sediments are transported for a long distance while western region sediments are transported for a short distance from the source area. In the future, the additional study on the heavy minerals in river sediments flowing into the Yellow Sea and much more samples for marine sediments must be carried out to interpret exactly the provenance and sedimentation process.

• Korean Chemical Engineering Research
• /
• v.55 no.5
• /
• pp.660-667
• /
• 2017

Equilibrium data of the mixture is essential in the design and operation of separation equipment such as distillation or extraction in chemical processes. These equilibrium data can be obtained through experiments or by calculations using the known binary parameters and the thermodynamic models. Generally, to obtain these parameters, phase equilibrium experimental data such as gas-liquid and liquid-liquid are used. In this study, the excess enthalpy of the mixture was measured using the flow type microcalorimeter which is a simpler method than phase equilibria experiments, and the parameters of various theories were obtained by using this data. In order to investigate the relationship between carbon chain length, enthalpy and binary parameters in the alkane system, excess enthalpies for the n-hexane + alkane (n-pentane, n-heptane, n-octane and n-dodecane) were measured at 298.15 K and the banary interaction parameters of Wilson, NRTL, and UNIQUAC were obtained from the experimental data. In addition, we wanted to obtain basic information on the interaction and association phenomena of the system including electrolyte applicable to various fields by using the excess enthalpy experimental data and the existing theory. First, we investigated the excess enthalpy for the NaOH / Water / Ethanol system as a basic experiment and examined the applicability using the electrolyte-NRTL (eNRTL) theory.

• Journal of the Korean Chemical Society
• /
• v.47 no.5
• /
• pp.479-486
• /
• 2003

In this study, poly(DL-lactide-co-glycolide) nanoparticles loaded with water-insoluble vitamins such as vitamin A (Retinol) and vitamin E acetate were prepared by the emulsification diffusion method. Polymer solution was prepared by the two water-miscible organic solvent, such as ethanol and acetone. Because of its biocompatible property, polyethyleneglycol-polypropyleneglycol diblock copolymer was used as surfactant and stabilizer. The influence of some preparative variables on the nanoparticle formation and on the loading efficiency of active agents, such as the type and concentration of stabilizing agent, the stirring methods, the water/oil phase ratio and the polymer concentration were investigated in order to control and optimize the process. After preparation of nanoparticles loaded with active agent, particle size and distribution were evaluated by the light scattering particle analyzer. The loading efficiency of active agents was evaluated by the UV-visible spectroscopy. As the results, particle size were 50-200 nm and dispersibility was monodisperse. The optimum loading efficiency of active agents was observed 50-60%. It was found that the appropriate of selections of binary solvent mixtures and polymeric concentrations in both organic and aqueous phases could provide good yield and favorable physical properties of PLGA nanoparticles.

• Proceedings of the Materials Research Society of Korea Conference
• /
• 2003.11a
• /
• pp.235-235
• /
• 2003

The development of a buffer layer is an important issue for the second -generation wire, YBCO coated metal wire. The buffer layer demands not only on the prohibition of the reaction between YBCO and metal substrate, but also the proper lattice match and conductivity for high critical current density (Jc) of YBCO superconductor, In order to satisfy these demands, we suggested CaRuO3 as a useful candidate having that the lattice mismatches with Ni (200) and with YBCO are 8.2% and 8.0%, respectively. The CaRuO3 thin films were deposited on Ni substrates using various methods, such as e-beam evaporation and DC and RF magnetron sputtering. These films were investigated using SEM, XRD, pole-figure and AES. In e-beam evaporation, the deposition temperature of CaRuO3 was the most important since both hi-axial texturing and NiO formation between Ni and CaRuO3 depended on it. Also, the oxygen flow rate had i[n effect on the growth of CaRuO3 on Ni substrates. The optimal conditions of crystal growth and film uniformity were 400$^{\circ}C$, 50 ㎃ and 7 ㎸ when oxygen flow rate was 70∼100sccm In RF magnetron sputtering, CaRuO3 was deposited on Ni substrates with various conditions and annealing temperatures. As a result, the conductivity of CaRuO3 thin films was dependent on CaRuO3 layer thickness and fabrication temperature. We suggested the multi-step deposition, such as two-step deposition with different temperature, to prohibit the NiO formation and to control the hi-axial texture.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.2
• /
• pp.962-969
• /
• 2013

This study used Computational Fluid Dynamics(CFD) to assess the properties of the air current inflow and the flow velocity distribution in the fume hood. In order to verify the effect of improvement, it was also predicted the characteristics of the flow pattern in case the hood face is structurally improved. The assessment of the capture velocity with the existing fume hood confirmed maximum 23 to 30% difference as compared to the root mean square (RMS). And the hood face showed great difference in flow velocity, with the flow velocity in the upper part is 58 to 68% faster than that in the lower part of the hood. So, as a result of the improvement of the hood designed to maintain a steady exhaust at the hood face (that is, installing a baffle on the inner wall and designing the slot type face), a difference of maximum 7% as compared to RMS appeared while maximum 12% differentiation in flow velocity through sections was predicted, showing mitigation of much of the difference in control velocity as compared to the previous structure.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.999-1004
• /
• 2005

An algorithm for a hybrid controller consists of a sliding mode control part and a fuzzy logic part which ar purposely for nonlinear systems. The sliding mode part of the solution is based on "eigenvalue/vector"-type controller is used as the backstepping approach for tracking errors. The fuzzy logic part is a Mamdani fuzzy model. This is designed by applying sliding mode control (SMC) method to the dynamic model. The main objective is to keep the update dynamics in a stable region by used SMC. After that the plant behavior is presented to train procedure of adaptive neuro-fuzzy inference systems (ANFIS). ANFIS architecture is determined and the relevant formulation for the approach is given. Using the error (e) and rate of error (de), occur due to the difference between the desired output value (yd) and the actual output value (y) of the system. A dynamic adaptation law is proposed and proved the particularly chosen form of the adaptation strategy. Subsequently VSC creates a sliding mode in the plant behavior while the parameters of the controller are also in a sliding mode (stable trainer). This study considers the ANFIS structure with first order Sugeno model containing nine rules. Bell shaped membership functions with product inference rule are used at the fuzzification level. Finally the Mamdani fuzzy logic which is depends on adaptive neuro-fuzzy inference systems structure designed. At the transferable stage from ANFIS to Mamdani fuzzy model is adjusted for the membership function of the input value (e, de) and the actual output value (y) of the system could be changed to trapezoidal and triangular functions through tuning the parameters of the membership functions and rules base. These help adjust the contributions of both fuzzy control and variable structure control to the entire control value. The application example, control of a mass-damper system is considered. The simulation has been done using MATLAB. Three cases of the controller will be considered: for backstepping sliding-mode controller, for hybrid controller, and for adaptive backstepping sliding-mode controller. A numerical example is simulated to verify the performances of the proposed control strategy, and the simulation results show that the controller designed is more effective than the adaptive backstepping sliding mode controller.