• 한국지하수토양환경학회:학술대회논문집
• /
• 한국지하수토양환경학회 2005년도 총회 및 춘계학술발표회
• /
• pp.348-351
• /
• 2005

In this study, solute breakthrough curves through the unsaturated zone were predicted using artificial neural network (ANN) by numerical tests and laboratory experiments. In the numerical tests, applicability of ANN model to prediction of breakthrough curves was evaluated using synthetic data generated by HYDRUS-2D. An appropriate strategy of ANN application and input data form were recommended. The ANN model was validated by laboratory experiments comparing with HYDRUS-2D simulations. The results show that the ANN model can be an effective method for forecasting solute breakthrough curves through the unsaturated zone when hydraulic data are available.

• Journal of Information Processing Systems
• /
• 제13권5호
• /
• pp.1203-1212
• /
• 2017

Intrusion detection systems (IDSs) are crucial in this overwhelming increase of attacks on the computing infrastructure. It intelligently detects malicious and predicts future attack patterns based on the classification analysis using machine learning and data mining techniques. This paper is devoted to thoroughly evaluate classifier ensembles for IDSs in IEEE 802.11 wireless network. Two ensemble techniques, i.e. voting and stacking are employed to combine the three base classifiers, i.e. decision tree (DT), random forest (RF), and support vector machine (SVM). We use area under ROC curve (AUC) value as a performance metric. Finally, we conduct two statistical significance tests to evaluate the performance differences among classifiers.

• Elastomers and Composites
• /
• 제56권2호
• /
• pp.79-84
• /
• 2021

A strain energy density function is used to characterize the hyperelasticity of rubber-like materials. Conventional models, such as the Neo-Hookean, Mooney-Rivlin, and Ogden models, are widely used in automotive industries, in which the strain potential is derived from strain invariants or principal stretch ratios. A fitting procedure for experimental data is required to determine material constants for each model. However, due to the complexities of the mathematical expression, these models can only produce an accurate curve fitting in a specified strain range of the material. In this study, a hyperelastic model for Neodymium Butadiene rubber is developed by using the Artificial Neural Network. Comparing the analytical results to those obtained by conventional models revealed that the proposed model shows better agreement for both uniaxial and equibiaxial test data of the rubber.

• 한국정밀공학회지
• /
• 제22권10호
• /
• pp.186-194
• /
• 2005

This study was designed to investigate the mechanical properties of the coating layer on electronic galvanized sheet steel as a part of the ongoing research on the coated steel. Those properties were determined using nano-indentation, the finite element method, and artificial neural networks. First and foremost, the load-displacement curve (the loading-unloading curve) of coatings was derived from a nano-indentation test by CSM (continuous stiffness measurement) and was used to measure the elastic modulus and hardness of the coating layer. The properties derived were applied in FE simulations of a nano-indentation test, and the analytical results were compared with the experimental result. A numerical model for FE simulations was established for the coating layer and the substrate separately. Finally, to determine the mechanical properties of the coating, such as the stress-strain curve, functional equations of loading and unloading curves were introduced and computed using the neural networks method. The results show errors within $5\%$ in comparison with the load-displacement measured by a nano-indentation test.

• 한국전기전자재료학회논문지
• /
• 제12권1호
• /
• pp.56-61
• /
• 1999

Highly conductive oxide films of BaRuO$_3$ have been grown heteroepitaxially on (100) LaAlO$_3$ single crystalline substrates by using pulsed laser deposition. The films are c-axis oriented with an in-plane epitaxial relationship of <010><100>BaRuO$_3$ // <110>LaAlO$_3$. Atomic force microscopy (AFM) observation shows that they consist of a fine-arranged network of grains and have a mosaic microstructure. Generally temperature-dependent resistivity shows the transition from metallic curve to semiconductor-metallic twofold curve by the deposition conditions for Ru oxide based materials like SrRuO$_3$, CaRuO$_3$, BaRuO$_3$, etc.. This twofold curve comes from the structural similarity of Ru oxide based materials including BaRuO$_3$. We find that the distance of Ru-Ru bonding in the unit cell of BaRuO$_3$ as well as the grain boundary scattering could be the two important causes of these interesting conductive properties.

• 한국산학기술학회논문지
• /
• 제21권6호
• /
• pp.374-382
• /
• 2020

본 연구에서 저장탄약 신뢰성평가(ASRP: Ammunition Stockpile Reliability Program)의 데이터 특성을 고려하여 입력변수를 줄이는 정규화기법을 제안함으로써 분류성능의 저하 없이 저장탄약 신뢰성분류 인경신경망모델의 학습 속도향상을 목표로 하였다. 탄약의 성능에 대한 기준은 국방규격(KDS: Korea Defense Specification)과 저장탄약 시험절차서(ASTP: Ammunition Stockpile reliability Test Procedure)에 규정되어 있으며, 평가결과 데이터는 이산형과 연속형 데이터가 복합적으로 구성되어 있다. 이러한 저장탄약 신뢰성평가의 데이터 특성을 고려하여 입력변수는 로트 추정 불량률(estimated lot percent nonconforming) 또는 고장률로 정규화 하였다. 또한 입력변수의 unitary hypercube를 유지하기 위하여 최소-최대 정규화를 2차로 수행하는 2단계 정규화 기법을 제안하였다. 제안된 2단계 정규화 기법은 저장탄약 신뢰성평가 데이터를 이용하여 비교한 결과 최소-최대 정규화와 유사하게 AUC(Area Under the ROC Curve)는 0.95 이상이었으며 학습속도는 학습 데이터 수와 은닉 계층의 노드 수에 따라 1.74 ~ 1.99 배 향상되었다.

• 한국물환경학회지
• /
• 제29권3호
• /
• pp.343-353
• /
• 2013

Flow duration curve (FDC) can be developed by linking the daily flow data of stream flow monitoring network to 8-day interval flow data of the unit watersheds for the management of Total Maximum Daily Loads. This study investigated the applicable method for the development of long term FDC with the selection of the stream flow reference sites, and suggested the development of the FDC in 4 river basins. Out of 142 unit watersheds in 4 river basins, 107 unit watersheds were shown to estimate daily flow data for the unit watersheds from 2006 to 2010. Short term FDC could be developed in 64 unit watersheds (45%) and long term FDC in 43 unit watersheds (30%), while other 35 unit watersheds (25%) were revealed to have difficulties in the development of FDC itself. Limits in the development of the long term FDC includes no stream monitoring sites in certain unit watersheds, short duration of stream flow data set and missing data by abnormal water level measurements on the stream flow monitoring sites. To improve these limits, it is necessary to install new monitoring sites in the required areas, to keep up continuous monitoring and make normal water level observations on the stream flow monitoring sites, and to build up a special management system to enhance data reliability. The development of long term FDC for the unit watersheds can be established appropriately with the normal and durable measurement on the selected reference sites in the stream flow monitoring network.

• Journal of Periodontal and Implant Science
• /
• 제48권2호
• /
• pp.114-123
• /
• 2018

Purpose: The aim of the current study was to develop a computer-assisted detection system based on a deep convolutional neural network (CNN) algorithm and to evaluate the potential usefulness and accuracy of this system for the diagnosis and prediction of periodontally compromised teeth (PCT). Methods: Combining pretrained deep CNN architecture and a self-trained network, periapical radiographic images were used to determine the optimal CNN algorithm and weights. The diagnostic and predictive accuracy, sensitivity, specificity, positive predictive value, negative predictive value, receiver operating characteristic (ROC) curve, area under the ROC curve, confusion matrix, and 95% confidence intervals (CIs) were calculated using our deep CNN algorithm, based on a Keras framework in Python. Results: The periapical radiographic dataset was split into training (n=1,044), validation (n=348), and test (n=348) datasets. With the deep learning algorithm, the diagnostic accuracy for PCT was 81.0% for premolars and 76.7% for molars. Using 64 premolars and 64 molars that were clinically diagnosed as severe PCT, the accuracy of predicting extraction was 82.8% (95% CI, 70.1%-91.2%) for premolars and 73.4% (95% CI, 59.9%-84.0%) for molars. Conclusions: We demonstrated that the deep CNN algorithm was useful for assessing the diagnosis and predictability of PCT. Therefore, with further optimization of the PCT dataset and improvements in the algorithm, a computer-aided detection system can be expected to become an effective and efficient method of diagnosing and predicting PCT.

• 정보보호학회논문지
• /
• 제25권2호
• /
• pp.467-478
• /
• 2015

사물인터넷에 대한 관심이 증가하면서 사물인터넷에 적합한 여러 보안 기술들이 연구되고 있다. 특히 디바이스 센서 네트워크 영역에서는 사물인터넷의 특성상 저사양 디바이스의 사용이 증가하고 다양화되었다. 그러나 현재의 인증 기술등의 보안 기술을 저전력 저사양 디바이스에 그대로 적용하기에 어려움이 있고, 이로 인해 보안 위협도 증가하였다. 따라서 사물인터넷의 센서 네트워크 통신상의 엔티티간 인증 프로토콜이 연구되고 있다. 2014년 Porambage 등은 타원곡선 암호 알고리즘에 기반한 센서 네트워크 인증 프로토콜을 제안하여 사물인터넷 환경의 안전성을 향상하고자 하였지만, 취약성이 존재하였다. 이에 따라 본 논문에서는 Porambage 등이 제안한 타원곡선 암호 알고리즘 기반 인증 프로토콜의 취약성을 분석하고, 사물인터넷 환경에서 센서 네트워크에 대한 개선된 인증 프로토콜을 제안한다.

• 한국조경학회지
• /
• 제35권6호
• /
• pp.64-73
• /
• 2008

The purpose of this study is to classify urban green space, to assess an imbalance by an administrative district (Dong), and to establish the management zone of urban green spaces for the construction of an environmental city in Changwon. The spatial data of 1:5,000 digital maps, park data in Changwon, land cover by the Ministry of Environment, and IKONOS satellite images from 2003 were used for this analysis. The assessment of the imbalance of urban green spaces was analyzed with the Lorenz curve and Gini's coefficient. The establishment of the management zone was performed by network analysis of GIS. The results of this study are as follows: the urban green spaces were classified as a park green space, a natural green space, and a riparian green space. According to the results of assessment of the imbalance of green spaces, Gini's coefficient was analyzed at higher than 0.4. Thus, the spatial imbalance of urban green spaces in Changwon was evident. The management zones to solve the imbalance were established: "rich zone", "fair zone", "poor zone" and "broken zone". Therefore, the rich and fair zones which have rich green spaces must maintain the good conditions through analysis of the green network and a survey of civic attitudes. The poor and broken zones which have poor green spaces must improve quality and quantity through creation of additional green spaces, construction of an eco-industrial park, and utilization of children's parks and pocket parks.