• Journal of the Korean Society of Hazard Mitigation
• /
• v.2 no.4 s.7
• /
• pp.123-129
• /
• 2002

In this study the results of Chungju reservoir inflow forecasting using 3 layered neural network model were analyzed in order to investigate the characteristics of neural network model for reservoir inflow forecasting. The proper neuron numbers of input and hidden layer were proposed after examining the variations of forecasted values according to neuron number and training epoch changes, and the probability of underestimation was judged by deliberating the variation characteristics of forecasting according to the differences between training and forecasting peak inflow magnitudes. In addition, necessary minimum training data size for precise forecasting was proposed. As a result, We confirmed the probability that excessive neuron number and training epoch cause over-fitting and judged that applying $8{\sim}10$ neurons, $1500{\sim}3000$ training epochs might be suitable in the case of Chungju reservoir inflow forecasting. When the peak inflow of training data set was larger than the forecasted one, it was confirmed that the forecasted values could be underestimated. And when the comparative short period training data was applied to neural networks, relatively inaccurate forecasting outputs were resulted and applying more than 600 training data was recommended for more precise forecasting in Chungju reservoir.

• Journal of the Korean GEO-environmental Society
• /
• v.3 no.4
• /
• pp.11-17
• /
• 2002

The purpose of this study is the prediction of pollutant loads into Seomjin river watershed using neural networks model. The pollutant loads into river watershed depend upon the water quantity of inflow from the upstream as well as the water quality of the inflow into the river. For the estimation of pollutants into river, a neural networks model which has the features of multi-layered structure and parallel multi-connections is used. The used water quality parameters are BOD, COD and SS into Seomjin river. The results of calibration are satisfactory, and proved the availability of a proposed neural networks model to estimate short-term water quality pollutants into river system.

• Journal of the Korean Geotechnical Society
• /
• v.19 no.6
• /
• pp.343-350
• /
• 2003

The range of applications of artificial neural networks(Am) in many branches of geotechnical engineering is growing rapidly. This study was undertaken to develop an analysis model representing ultrasonically enhanced soil flushing by the use of ANN. Input data for the model-development were obtained by laboratory study, and used for training and verification. Analyses involved various ranges of momentum, loaming rate, activation function, hidden layer, and nodes. Results of the analyses were used to obtain the optimum conditions for establishing and verifying the model. The coefficient of correlation between the measured and the predicted data using the developed model was relatively high. It shows potential application of ANN to ultrasonically enhanced soil flushing which is not easy to build up a mathematical model.

• Herbal Formula Science
• /
• v.17 no.2
• /
• pp.99-110
• /
• 2009

목 적 : 한약 또는 한약제제의 아토피성 피부염에 대한 치료효과를 조사하고 향후 연구방향을 제시하기 위하여 고찰연구를 시행하였다. 방 법 : PubMed에 한약과 아토피성 피부염과 관련된 검색어의 조합을 넣어 포함기준에 맞는 무작위배정 대조군 임상연구만 포함하였다. 연구설계, 치료방법, 대조군, 평가지표, 결과, 부작용 관련 정보를 미리 정해놓은 자료 추출 형식에 맞추어 추출하고 방법론적 질 평가는 옥스포드 질 평가 척도와 그룹 할당 은닉(allocation concealment) 여부를 평가하였다. 연구들이 임상적 및 통계적으로 상이하여 메타분석은 이루어지지 않고 기술적 고찰만 실시하였다. 결 과 : 모두 8편의 연구가 고찰기준을 만족시켰다. 다양한 복합한약제제와 한약이 포함된 외용제가 평가되었는데 8편 가운데 5편에서 아토피성 피부염의 증상을 호전시키는 것으로 나타났다. 방법론적 질은 대체로 양호한 것으로 나타났으며 일부 효과적인 것으로 나타난 한약복합제제에서 간손상 등의 부작용도 보고되었다. 결 론 : 한약 또는 한약제제를 이용한 치료는 아토피성 피부염의 증상개선에 도움이 되는 것으로 보이나 현재 근거는 부족하다. 우리나라에서 많이 쓰이는 한약제제들 역시 엄정한 임상연구를 거쳐 그 효과를 평가하고 근거를 구축해야 할 것이다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.403-403
• /
• 2019

수자원을 효율적으로 관리하고 사용하는 것은 확보한 수자원을 확보한 목표에 맞게 시,공간적으로 적절하게 분배 시키는 것이다. 따라서 저수지 운영의 최종 목표는 댐 건설 목적에 따라 확보된 물을 유입량, 저수량, 용수 수요등을 감안하여 댐 운영 목표에 맞게 최적으로 적절한 양의 물을 적절한 시기에 방류하는 것이다.(손덕환, 2004) 현재 댐군의 운영방법은 확정론적인 방법과 추계학적인 방법이 주로 이용되고 있으나 본 연구에서는 최근 연구가 많이 이루어지고 있는 인공신경망을 적용하여 운영방법으로써의 적용성을 검토하고자한다. 연구대상지로는 수력발전소가 포함된 한강의 충주 다목적댐을 선정하였다. 인공신경망은 입력층에서 출력층사이에 은닉층이 존재하는 다중신경망을 활용하였으며 출력층은 방류량으로 설정하여 발전방류와 수문방류를 구분하여 설정하였다. 방류량 결정을 위한 입력층 구성은 선행 연구들을 참고하여 예측 유입량, 현재 수위, 발전량, 용수 수요량 등을 설정하여 입력층으로 구성하였다. 학습기간의 방류량 자료를 학습하고 검정기간을 통해 실제 이루어진 방류량과 모의된 방류량의 차이를 비교, 분석하여 댐 운영방법으로써의 인공신경망의 적용성을 검토하고자하였다.

• Journal of Korea Water Resources Association
• /
• v.37 no.9
• /
• pp.771-780
• /
• 2004

The purpose of this study is to test a development of system for flood runoff forecasting using neural network model. As the forecasting models for flood runoff the neural network model was tested with the observed flood data at Gongju and Buyeo stations. The neural network model consists of input layer, hidden layer, and output layer. For the flood events tested rainfall and runoff data were the input to the input layer and the flood runoff data were used in the output layer. To make a choice the forecasting model which would make up of runoff forecasting system properly, real-time runoff of river when flood periods were forecasted by using neural network model and state-space model. A comparison of the results obtained by the two forecasting models indicated the superiority and reliability of the neural network model over the state-space model. The neural network model was modified to work in the Web and developed to be the basic model of the forecasting system for the flood runoff. The neural network model developed to be used in the Web was loaded into the server and was applied to the main stream of Geum river. For the main stage gauging stations mentioned above the applicability of the selected forecasting model, the Neural Network Model, was verified in the Web.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.14 no.1
• /
• pp.37-53
• /
• 2012

There is considerable uncertainty in ground properties used in tunnel designs. In this study, a back analysis was performed to find optimal ground properties based on the artificial neural network facility of MATLAB program of using tunnel monitoring data. Total 81 data were constructed by changing elastic modulus and coefficient of lateral pressure which have great influence on tunnel convergence. A sensitivity analysis was conducted to establish an optimal training model by varying the number of hidden layers, the number of nodes, learning rate, and momentum. Meanwhile, the optimal training model was selected by comparing MSE (Mean Squared Error) and coefficient of determination ($R^2$) and was used to find the correct elastic moduli of layers and the coefficient of lateral pressure. In future, it is expected that the suggested method of this study can be applied to determine the optimum tunnel support pattern under given ground conditions.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.23 no.1
• /
• pp.57-67
• /
• 2013

Lately, intrusive incidents (including system hacking, viruses, worms, homepage alterations, and data leaks) have not involved the distribution of an virus or worm, but have been designed to acquire private information or trade secrets. Because an attacker uses advanced intelligence and attack techniques that conceal and alter data in a computer, the collector cannot trace the digital evidence of the attack. In an initial incident response first responser deals with the suspect or crime scene data that needs investigative leads quickly, in accordance with forensic process methodology that provides the identification of digital evidence in a systematic approach. In order to an effective initial response to first responders, this paper analyzes the collection data such as user usage profiles, chronology timeline, and internet data according to CFFPM(computer forensics field triage process model), proceeds to design, and implements a collection application to deploy the client/server architecture on the Windows based computer.

• Journal of Korea Multimedia Society
• /
• v.13 no.5
• /
• pp.754-762
• /
• 2010

Nowadays, email-based targeted attacks using malcode-bearing documents have been steadily increased. To improve the success rate of the attack and avoid anti-viruses, attackers mainly employ zero-day exploits and relevant social engineering techniques. In this paper, we propose an architecture of the email vaccine cloud system to prevent targeted attacks using malcode-bearing documents. The system extracts attached document files from email messages, performs behavior analysis as well as signature-based detection in the virtual machine environment, and completely removes malicious documents from the messages. In the process of behavior analysis, the documents are regarded as malicious ones in cases of creating executable files, launching new processes, accessing critical registry entries, connecting to the Internet. The email vaccine cloud system will help prevent various cyber terrors such as information leakages by preventing email based targeted attacks.

• The Korean Journal of Applied Statistics
• /
• v.32 no.6
• /
• pp.899-908
• /
• 2019

Massive in-depth neural networks with numerous parameters are powerful machine learning methods, but they have overfitting problems due to the excessive flexibility of the models. Dropout is one methods to overcome the problem of oversized neural networks. It is also an effective method that randomly drops input and hidden nodes from the neural network during training. Every sample is fed to a thinned network from an exponential number of different networks. In this study, instead of feeding one sample for each thinned network, two or more samples are used in fitting for one thinned network known as a Hybrid Dropout. Simulation results using real data show that the new method improves the stability of estimates and reduces the minimum error for the verification data.