• Membrane and Water Treatment
• /
• 제9권2호
• /
• pp.115-121
• /
• 2018

Artificial neural network (ANN) simulation is used to predict the dynamic change of permeate flux during wheat starch industry wastewater microfiltration with and without static turbulence promoter. The experimental program spans range of a sedimentation times from 2 to 4 h, for feed flow rates 50 to 150 L/h, at transmembrane pressures covering the range of $1{\times}10^5$ to $3{\times}10^5Pa$. ANN predictions of the wastewater microfiltration are compared with experimental results obtained using two different set of microfiltration experiments, with and without static turbulence promoter. The effects of the training algorithm, neural network architectures on the ANN performance are discussed. For the most of the cases considered, the ANN proved to be an adequate interpolation tool, where an excellent prediction was obtained using automated Bayesian regularization as training algorithm. The optimal ANN architecture was determined as 4-10-1 with hyperbolic tangent sigmoid transfer function transfer function for hidden and output layers. The error distributions of data revealed that experimental results are in very good agreement with computed ones with only 2% data points had absolute relative error greater than 20% for the microfiltration without static turbulence promoter whereas for the microfiltration with static turbulence promoter it was 1%. The contribution of filtration time variable to flux values provided by ANNs was determined in an important level at the range of 52-66% due to increased membrane fouling by the time. In the case of microfiltration with static turbulence promoter, relative importance of transmembrane pressure and feed flow rate increased for about 30%.

• Wind and Structures
• /
• 제36권6호
• /
• pp.393-404
• /
• 2023

Skewness and kurtosis are important higher-order statistics for simulating non-Gaussian wind pressure series on low-rise buildings, but their predictions are less studied in comparison with those of the low order statistics as mean and rms. The distribution gradients of skewness and kurtosis on roofs are evidently higher than those of mean and rms, which increases their prediction difficulty. The conventional artificial neural networks (ANNs) used for predicting mean and rms show unsatisfactory accuracy in predicting skewness and kurtosis owing to the limited capacity of shallow learning of ANNs. In this work, the deep neural networks (DNNs) model with the ability of deep learning is introduced to predict the skewness and kurtosis on a low-rise building. For obtaining the optimal generalization of the DNNs model, the hyper parameters are automatically determined by Bayesian Optimization (BO). Moreover, for providing a benchmark for future studies on predicting higher order statistics, the data sets for training and testing the DNNs model are extracted from the internationally open NIST-UWO database, and the prediction errors of all taps are comprehensively quantified by various error metrices. The results show that the prediction accuracy in this study is apparently better than that in the literature, since the correlation coefficient between the predicted and experimental results is 0.99 and 0.75 in this paper and the literature respectively. In the untrained cornering wind direction, the distributions of skewness and kurtosis are well captured by DNNs on the whole building including the roof corner with strong non-normality, and the correlation coefficients between the predicted and experimental results are 0.99 and 0.95 for skewness and kurtosis respectively.

• 한국음향학회지
• /
• 제42권4호
• /
• pp.329-344
• /
• 2023

본 논문에서는 소량 및 불균형 능동소나 데이터세트에 적용된 다양한 딥러닝 기반 표적식별기의 일반화 성능을 종합적으로 분석하였다. 서로 다른 시간과 해역에서 수집된 능동소나 실험 데이터를 이용하여 두 가지 능동소나 데이터세트를 생성하였다. 데이터세트의 각 샘플은 탐지 처리 이후 탐지된 오디오 신호로부터 추출된 시간-주파수 영역 이미지이다. 표적식별기의 신경망 모델은 다양한 구조를 가지는 22개의 Convolutional Neural Networks(CNN) 모델을 사용하였다. 실험에서 두 가지 데이터세트는 학습/검증 데이터세트와 테스트 데이터세트로 번갈아 가며 사용되었으며, 표적식별기 출력의 변동성을 계산하기 위해 학습/검증/테스트를 10번 반복하고 표적식별 성능을 분석하였다. 이때 학습을 위한 초매개변수는 베이지안 최적화를 이용하여 최적화하였다. 실험 결과 본 논문에서 설계한 얕은 층을 가지는 CNN 모델이 대부분의 깊은 층을 가지는 CNN 모델보다 견실하면서 우수한 일반화 성능을 가지는 것을 확인하였다. 본 논문은 향후 딥러닝 기반 능동소나 표적식별 연구에 대한 방향성을 설정할 때 유용하게 사용될 수 있다.

• 한국전자통신학회논문지
• /
• 제12권5호
• /
• pp.923-932
• /
• 2017

본 논문에서는 기존의 문제점인 얼굴 움직임이 있을 시 시선 식별이 어려운 점과 사용자에 따른 교정작업이 필요하다는 점을 해결하고자 새로운 시선 식별 시스템과 얼굴인식에 필요한 GRNN(: Generalized Regression Neural Network) 알고리즘을 제안한다. Kalman필터를 사용하여 현재 머리의 위치정보를 이용하여 미래위치를 추정하였고 얼굴의 진위 여부를 판단하기 위해서 얼굴의 특징요소를 구조적 정보와 비교적 처리시간이 빠른 수평, 수직 히스토그램 분석법을 이용하여 얼굴의 요소를 검출한다. 그리고 적외선 조명기를 구성하여 밝은 동공효과를 얻어 동공을 실시간으로 검출, 추적하였고 동공-글린트 벡터를 추출하여 의료정보 보호에 도움을 주고자 한다.

• 한국멀티미디어학회논문지
• /
• 제6권1호
• /
• pp.136-147
• /
• 2003

주어진 문제에 대하여 최적의 성능을 가지는 신경회로망을 얻기 위해서는 학습을 통한 매개변수의 최적화 (parameter optimization)와 모델 선택을 통한 구조 최적화(structure optimization )의 통합적인 과정이 필요하다. 본 논문에서는, 각 세부 방법들의 특성을 고려하여, 공통의 특성을 갖는 방법들을 결합함으로써 효율적이면서도 일반화 성능을 높이는 총체적인 신경회로망 최적화 방법을 제안한다. 먼저 다양한 오차 함수를 사용할 수 있는 자연 기울기 강하 학습에 적응적 정규화 방법을 도입함으로써 가중치 매개변수(weight parameter)들을 최적화한다. 그리고 이렇게 최적화된 매개변수(parameter)들에 자연 프루닝(natural pruning)을 적용하여 불필요한 요소들을 제저하여 최적화 된 구조를 생성한다. 반복적인 과정에 의하여 후보 모델들을 구성하고 베이시안 정보 기준(Bayesian Information Criterion: BIC )을 이 용하여 최적의 모델을 평가하여 선택하는 방법을 제안하였다. 벤치마크 데이터에 대한 실험을 통하여 제안하는 방법의 구조 최적화 능력과 일반화 성능의 우수성을 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권8호
• /
• pp.2948-2963
• /
• 2015

In the field of computer vision, visual surveillance systems have recently become an important research topic. Growth in this area is being driven by both the increase in the availability of inexpensive computing devices and image sensors as well as the general inefficiency of manual surveillance and monitoring. In particular, the ultimate goal for many visual surveillance systems is to provide automatic activity recognition for events at a given site. A higher level of understanding of these activities requires certain lower-level computer vision tasks to be performed. So in this paper, we propose an intelligent activity recognition model that uses a structure learning method and a classification method. The structure learning method is provided as a K2-learning algorithm that generates Bayesian networks of causal relationships between sensors for a given activity. The statistical characteristics of the sensor values and the topological characteristics of the generated graphs are learned for each activity, and then a neural network is designed to classify the current activity according to the features extracted from the multiple sensor values that have been collected. Finally, the proposed method is implemented and tested by using PETS2013 benchmark data.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2012년도 학술발표회
• /
• pp.195-195
• /
• 2012

인공신경망 모형은 복잡하고 비선형의 입력과 출력 관계를 잘 반영할 수 있어서 유출 모델링에 널리 적용되어 왔다. 그러나 인공신경망 모형은 강우나 유역특성의 공간적 분포를 반영하는 것이 어려우며 물리적 개념이 결여되어 있는 단점이 있다. 본 연구에서는 유역특성과 물리적 개념을 반영할 수 있는 물리기반 모형과 인공신경망 모형의 장점들을 조합하여 물리기반 모형의 일 유출량 해석 능력을 향상하기 위하여 SWAT 모형과 인공신경망(ANN)을 연계하였다. SWAT-ANN 연계모형은 두 단계로 구성되어 진다. 첫 번째 단계에서는 관측 자료를 이용하여 SWAT 모형을 보정한다. 두 번째 단계에서는 첫 번째 단계에서 계산한 소유역별 SWAT 모형의 유출결과를 ANN의 입력자료로 이용하여 SWAT-ANN 연계모형을 구축한다. SCE-UA 최적화 방법을 적용하여 SWAT 모형의 매개변수들을 보정하였고, ANN 학습은 3층의 feed-forward 역전파 알고리즘에 기초한 Bayesian Regularization 방법을 적용하였다. ANN 은닉층의 뉴런 및 전달함수는 시행착오를 통하여 적절한 ANN 구조를 설정하여 SWAT-ANN 연계모형의 일유출량을 모의하였다. 여러 가지 통계적 오차기준을 이용하여 보청천 유역에서 SWAT-ANN 연계모형의 결과와 SWAT 단독 모형의 결과를 비교하였다. SWAT-ANN 연계모형이 SWAT 단독 모형보다 더 우수한 결과를 나타내어 일 유출량 해석을 위한 SWAT-ANN 연계모형의 유용성을 확인할 수 있었다.

• International Journal of Computer Science & Network Security
• /
• 제21권9호
• /
• pp.31-40
• /
• 2021

The increasing number of credit card fraud cases has become a considerable problem since the past decades. This phenomenon is due to the expansion of new technologies, including the increased popularity and volume of online banking transactions and e-commerce. In order to address the problem of credit card fraud detection, a rule-based approach has been widely utilized to detect and guard against fraudulent activities. However, it requires huge computational power and high complexity in defining and building the rule base for pattern matching, in order to precisely identifying the fraud patterns. In addition, it does not come with intelligence and ability in predicting or analysing transaction data in looking for new fraud patterns and strategies. As such, Data Mining and Machine Learning algorithms are proposed to overcome the shortcomings in this paper. The aim of this paper is to highlight the important techniques and methodologies that are employed in fraud detection, while at the same time focusing on the existing literature. Methods such as Artificial Neural Networks (ANNs), Support Vector Machines (SVMs), naïve Bayesian, k-Nearest Neighbour (k-NN), Decision Tree and Frequent Pattern Mining algorithms are reviewed and evaluated for their performance in detecting fraudulent transaction.

• 대기
• /
• 제16권1호
• /
• pp.19-31
• /
• 2006

The purpose of this study is to develop the statistical model to predict sea level pressure of typhoon period in south coast of the Korean Peninsula. Seven typhoons, which struck south coast of the Korean Peninsula, are selected for this study, and the data for analysis include the central pressure and location of typhoon, and sea level pressure and location of 19 observing site. Models employed in this study are the first order regression, the second order regression and the neural network. The dependent variable of each model is a 3-hr interval sea level pressure at each station. The cause variables are the central pressure of typhoon, distance between typhoon center and observing site, and sea level pressure of 3 hrs before, whereas the indicative variable reveals whether it is before or after typhoon passing. The data are classified into two groups - one is the full data obtained during typhoon period and the other is the data that sea level pressure is less than 1000 hPa. The stepwise selection method is used in the regression model while the node number is selected in the neural network by the Schwarz's Bayesian Criterion. The performance of each model is compared in terms of the root-mean square error. It turns out that the neural network shows better performance than other models, and the case using the full data produces similar or better results than the case using the other data.

• 한국컴퓨터정보학회논문지
• /
• 제21권6호
• /
• pp.9-19
• /
• 2016

In finance literature, stock liquidity showing how stocks can be cashed out in the market has received rich attentions from both academicians and practitioners. The reasons are plenty. First, it is known that stock liquidity affects significantly asset pricing. Second, macroeconomic announcements influence liquidity in the stock market. Therefore, stock liquidity itself affects investors' decision and managers' decision as well. Though there exist a great deal of literature about stock liquidity in finance literature, it is quite clear that there are no studies attempting to investigate the stock liquidity issue as one of decision making problems. In finance literature, most of stock liquidity studies had dealt with limited views such as how much it influences stock price, which variables are associated with describing the stock liquidity significantly, etc. However, this paper posits that stock liquidity issue may become a serious decision-making problem, and then be handled by using data mining techniques to estimate its future extent with statistical validity. In this sense, we collected financial data set from a number of manufacturing companies listed in KRX (Korea Exchange) during the period of 2010 to 2013. The reason why we selected dataset from 2010 was to avoid the after-shocks of financial crisis that occurred in 2008. We used Fn-GuidPro system to gather total 5,700 financial data set. Stock liquidity measure was computed by the procedures proposed by Amihud (2002) which is known to show best metrics for showing relationship with daily return. We applied five data mining techniques (or classifiers) such as Bayesian network, support vector machine (SVM), decision tree, neural network, and ensemble method. Bayesian networks include GBN (General Bayesian Network), NBN (Naive BN), TAN (Tree Augmented NBN). Decision tree uses CART and C4.5. Regression result was used as a benchmarking performance. Ensemble method uses two types-integration of two classifiers, and three classifiers. Ensemble method is based on voting for the sake of integrating classifiers. Among the single classifiers, CART showed best performance with 48.2%, compared with 37.18% by regression. Among the ensemble methods, the result from integrating TAN, CART, and SVM was best with 49.25%. Through the additional analysis in individual industries, those relatively stabilized industries like electronic appliances, wholesale & retailing, woods, leather-bags-shoes showed better performance over 50%.