• Journal of Korea Multimedia Society
• /
• v.6 no.1
• /
• pp.136-147
• /
• 2003

To achieve an optimal performance for a given problem, we need an integrative process of the parameter optimization via learning and the structure optimization via model selection. In this paper, we propose an efficient optimization method for improving generalization performance by considering the property of each sub-method and by combining them with common theoretical properties. First, weight parameters are optimized by natural gradient teaming with adaptive regularization, which uses a diverse error function. Second, the network structure is optimized by eliminating unnecessary parameters with natural pruning. Through iterating these processes, candidate models are constructed and evaluated based on the Bayesian Information Criterion so that an optimal one is finally selected. Through computational experiments on benchmark problems, we confirm the weight parameter and structure optimization performance of the proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.3
• /
• pp.1633-1649
• /
• 2017

In remote sensing image processing, the traditional fusion algorithm is based on the Intensity-Hue-Saturation (IHS) transformation. This method does not take into account the texture or spectrum information, spatial resolution and statistical information of the photos adequately, which leads to spectrum distortion of the image. Although traditional solutions in such application combine manifold methods, the fusion procedure is rather complicated and not suitable for practical operation. In this paper, an improved IHS transformation fusion algorithm based on the local variance weighting scheme is proposed for remote sensing images. In our proposal, firstly, the local variance of the SPOT (which comes from French "Systeme Probatoire d'Observation dela Tarre" and means "earth observing system") image is calculated by using different sliding windows. The optimal window size is then selected with the images being normalized with the optimal window local variance. Secondly, the power exponent is chosen as the mapping function, and the local variance is used to obtain the weight of the I component and match SPOT images. Then we obtain the I' component with the weight, the I component and the matched SPOT images. Finally, the final fusion image is obtained by the inverse Intensity-Hue-Saturation transformation of the I', H and S components. The proposed algorithm has been tested and compared with some other image fusion methods well known in the literature. Simulation result indicates that the proposed algorithm could obtain a superior fused image based on quantitative fusion evaluation indices.

• International Journal of Computer Science & Network Security
• /
• v.21 no.6
• /
• pp.319-328
• /
• 2021

Parallel administration of numerous drugs increases Drug-Drug Interaction (DDI) because one drug might affect the activity of other drugs. DDI causes negative or positive impacts on therapeutic output. So there is a need to discover DDI to enhance the safety of consuming drugs. Though there are several DDI system exist to predict an interaction but nowadays it becomes impossible to maintain with a large number of biomedical texts which is getting increased rapidly. Mostly the existing DDI system address classification issues, and especially rely on handcrafted features, and some features which are based on particular domain tools. The objective of this paper to predict DDI in a way to avoid adverse effects caused by the consumed drugs, to predict similarities among the drug, Drug pair similarity calculation is performed. The best optimal weight is obtained with the support of KHA. LSTM function with weight obtained from KHA and makes bets prediction of DDI. Our methodology depends on (LSTM-KHA) for the detection of DDI. Similarities among the drugs are measured with the help of drug pair similarity calculation. KHA is used to find the best optimal weight which is used by LSTM to predict DDI. The experimental result was conducted on three kinds of dataset DS1 (CYP), DS2 (NCYP), and DS3 taken from the DrugBank database. To evaluate the performance of proposed work in terms of performance metrics like accuracy, recall, precision, F-measures, AUPR, AUC, and AUROC. Experimental results express that the proposed method outperforms other existing methods for predicting DDI. LSTMKHA produces reasonable performance metrics when compared to the existing DDI prediction model.

• Journal of Korean Society of Steel Construction
• /
• v.16 no.5 s.72
• /
• pp.683-694
• /
• 2004

Advanced analysis and optimal design of semi-rigid space steel frames were presented. The advanced analysis can predict the combined nonlinear effects of connection, geometry, and material on the behavior and strength of semi-rigid frames. The Kishi-Chen power model was used to describe the nonlinear behavior of semi-rigid connections. Geometric nonlinearity was determined using stability functions. Material nonlinearity was determined using the Column Research Council (CRC) tangent modulus and the parabolic function. The direct search method proposed by Choi and Kim was used as optimization technique. One by one, the member with the largest unit value evaluated using the LRFD interaction equation were placed adjacent to a larger member selected from the database. The objective function was assumed to be the weight of steel frame, while the constraint functions were load-carrying capacities, deflections, inter-story drifts, and the ductility requirements. The member sizes determined using the proposed method were compared to those derived from the conventional LRFD method.

• Journal of Korean Society of Steel Construction
• /
• v.15 no.6 s.67
• /
• pp.661-672
• /
• 2003

The advanced analysis and optimal design of semi-rigid frame were presented. Advanced analysis can predict the combined nonlinear effects of connection, geometry, and material on the behavior and strength of semi-rigid frames. The Kishi-Chen power model was used to describe the nonlinear behavior of semi-rigid connections. Geometric nonlinearity was determined using stability functions. On the other hand, material nonlinearity was determined using the Column Research Council (CRC) tangent modulus and parabolic function. The direct search method proposed by Choi and Kim was used as optimization technique. The member with the largest unit value evaluated using the LRFD interaction equation was replaced one by one with an adjacent larger member selected from the database. The objective function was assumed as the weight of steel frame, with the constraint functions accounting for load-carrying capacities, deflections. inter-story drifts, and ductility requirement. Member sizes determined by the proposed method were compared with those derived using the conventional LRFD method.

• Journal of Acupuncture Research
• /
• v.29 no.3
• /
• pp.9-18
• /
• 2012

Objectives : This research was performed to investigate the effects of Rhei Radix et Rhizoma pharmacopuncture at Bi-Su($BL_{20}$) on weight change, food efficiency, concentration of serum lipid, liver function of rats fed high fat diet for 8 weeks. Methods : Subject groups were diveded into normal diet group(Normal), high fat diet group(Control), high fat diet and saline in $BL_{20}$ treated group(Saline), high fat diet and Rhei Radix et Rhizoma pharmacopuncture in $BL_{20}$ treated group(PT-1, 0.008mg/ml/g), high fat diet and Rhei Radix et Rhizoma pharmacopuncture in $BL_{20}$ treated group(PT-2, 0.021mg/ml/g), and high fat diet and Rhei Radix et Rhizoma pharmacopuncture in $BL_{20}$ treated group(PT-3, 0.042mg/ml/g). Pharmacopuncture was bilaterally treated at BL20 every 4 days, totally 14 times in 56 days. We observed the changes of body weight, food efficiency, ALT, AST, r-GTP, total cholesterol, triglyceride, HDL, LDL-cholesterol of each group. Results : 1. Body weight significantly decreased, as compared with the control group, in PT-1 from 2nd week to 8th week, in PT-2 from 4th week to 7th week. 2. Food efficiency decreased, as compared with the control group, in PT-1(0.05${\leq}$p<0.06). 3. HDL-cholesterol increased, as compared with the control group, in PT-1, PT-3, but not significantly. 4. In the changes of ALT, as compared with the control group, only in PT-1 group, the level of serum ALT decreased(0.05${\leq}$p<0.06). 5. As for the changes of r-GTP, as compared with control group, in every subject group, the levels of serum r-GTP significantly decreased. Conclusions : According to the above results, Rhei Radix et Rhizoma pharmacopuncture at $BL_{20}$ can act on body weight, food efficiency and liver function. It also does not seem to be injurious on liver function. To use Rhei Radix et Rhizoma pharmacopuncture for obesity, further studies on its underlying mechanism and optimal dosage might be needed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.4
• /
• pp.767-778
• /
• 2003

This paper presents a method for generating optimal fuzzy rules using a genetic algorithm. Fuzzy rules are generated from the training data in the first stage. In this stage, fuzzy c-Means clustering method and cluster validity are used to determine the structure and initial parameters of the fuzzy inference system. A cluster validity is used to determine the number of clusters, which can be the number of fuzzy rules. Once the structure is figured out in the first stage, parameters relating the fuzzy rules are optimized in the second stage. Weights and variance parameters are tuned using genetic algorithms. Variance parameters are also managed with left and right for asymmetrical Gaussian membership function. The method ensures convergence toward a global minimum by using genetic algorithms in weight and variance spaces.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.23 no.11
• /
• pp.1453-1463
• /
• 1999

An electromagnetic flowmeter, based on a magnetic induction principle; provides an obstructionless flowmeter that essentially averages the velocity distribution over the pipe cross-sectional area. To predict the installation effects, the flowmeter installed near $90^{\circ}$ elbow is simulated by using a commercial code FLUENT(ver. 4.48) for the laminar flow field and a code developed through this study for magnetic field. The installation effects of the flowmeter are estimated by varying a number of the dependent parameters such as the radius of the elbow(Rc=1D, 1.5D, 2D, 3D), the location, Reynolds number and the direction of electrodes plane(${\varphi}$). It was found that all these factors affect the performance of the electromagnetic flowmeter significantly. The longer installation distance from the elbow is not always optimal to minimize the error, and also there exists an optimal location to install the EMF for a minimum error. Especially the flow signal with the electrodes plane direction of ${\varphi}=45^{\circ}$ is shown to yield smallest measurement error regardless of the Reynolds number and the curvature of elbow.

• Information and Communications Magazine
• /
• v.4 no.4
• /
• pp.505-524
• /
• 1987

Nowaday all the countries of the world have studied the various problems caused in operating their own ports efficiently. Ship delay in the port is attributal to the inefficient operation in the navigation aids, the cargo handling, the storage and transfer facilities, and to the inefficient allocation of gangs or to a bad service for ships. Among these elements the allocation of gangs is the predominating factor in minimizing ship's turn round time. At present, in the case of Pusan Port. the labour union and stevedoring companies allocate gangs in every hatches of ships by a rule of thumb, just placing emphasis on minimizing ship's turn round time, without applying the principle of allocation during the cargo handling. Owing to this the efficiency of the cargo handling could not be expected to be maximized and this unsystematic operation result in supplying human resources of much unnecessary surplus gangs. Therefore in this paper the optimal size and allocation of gangs for minimizing the ship's turn round time is studied and formularized. For the determination of the priority for allocation the evaluation function, namely $F=PHi^{n}{\times}(W+H)$, can be obtained; where, PHI : Principal Hatch Index W : Total Cargo Weight represented in Gang-Shifts H : Total Number of Ship's hatches and also for the optimal size of gangs the average number of gang allocated per shift (Ng), namely Ng=W/PHI, is used. The proposed algorithm is applied to Pusan Port and its validity is verified.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.10a
• /
• pp.754-758
• /
• 2002

According to the enlargement of production facilities and structures, the requirements of high-capacity load cells are increased for monitoring the process conditions in many fields. Generally, however, the accuracy of the column-type high-capacity load cells is not enough due to the geometric nonlinearity. It is supposed to result from the fact that the whole spring element is under high-level stress for the uniform strain field. In this paper, a new shape of spring element is developed which utilizes the stress concentration. As a design criterion, an object function which quantifies the degree of nonlinearity is defined and optimized by use of response surface modeling. As a result, the weight of the spring element is reduced shout 50% in comparison to the conventional shape. The bonding positions of stain gages are found. which show theoretically zero geometrical nonlinearity, while the ratio of overload protection is reduced from 130% to 125% Also it is shown that the response surface method is very efficient in the optimization approach by use of FEM.