• Advances in environmental research
• /
• 제6권4호
• /
• pp.265-279
• /
• 2017

The climate change has made adverse effects on land surface temperature for many regions of the world. Several climatic studies focused on different downscaling techniques for climatological parameters of different regions. For statistical downscaling of any hydrological parameters, conventional Neural Network Models were used in common. However, it seems that in any modeling study, uncertainty is a vital aspect when making any predictions about the performance. In this paper, Gamma Test is performed to determine the data length selection for training to minimize the uncertainty in model development. Another measure to improve the data quality and model development are wavelet transforms. Hence, Gamma Test with Wavelet decomposed Feedforward Neural Network (GT-WNN) model is developed and tested for downscaled land surface temperature of Patna Urban, Bihar. The results of GT-WNN model are compared with GT-FFNN and conventional Feedforward Neural Network (FFNN) model. The effectiveness of the developed models is illustrated by Root Mean Square Error and Coefficient of Correlation. Results showed that GT-WNN outperformed the GT-FFNN and conventional FFNN in downscaling the land surface temperature. The land surface temperature is forecasted for a period of 2015-2044 with GT-WNN model for Patna Urban in Bihar. In addition, the significance of the probable changes in the land surface temperature is also found through Mann-Kendall (M-K) Test for Summer, Winter, Monsoon and Post Monsoon seasons. Results showed an increasing surface temperature trend for summer and winter seasons and no significant trend for monsoon and post monsoon season over the study area for the period between 2015 and 2044. Overall, the M-K test analysis for the annual data shows an increasing trend in the land surface temperature of Patna Urban.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2006년도 춘계학술발표대회
• /
• pp.1223-1226
• /
• 2006

P2P(Peer to Peer) 기술은 1990년대 후반기부터 산업계 및 학계에 주목을 받고 있는 기술 분야중의 하나로 이 기술의 장점은 인터넷 환경에 산재하여 있는 컴퓨팅 파워, 공간, 네트워크 대역을 인터넷 기반으로 효과적으로 활용하여 협력작업을 가능하게 한다는데 있다. 최근에는 모바일 환경 응용을 위한 P2P 디바이스 탐색 분야에 관심사가 증대되고 있으며, P2P 시스템은 중앙통제 장치가 결여 되어 있기 때문에 중앙통제 장치 개입을 최소로 하면서 P2P를 운영하기 위한 효율적인 기법 및 체계가 요구되고 있다. 본 논문에서는 기존의 접근방법을 검토하여 FFNN(feed forward neural network)을 이용한 디바이스 탐색 기법을 제시한다. 제시한 FFNN은 BP(back propagation) 알고리즘을 통해 훈련하고 디바이스를 탐색한다. 제시한 시스템의 성능을 보이기 위해 일정한 계산량을 가지는 작업을 에이전트를 활용, 탐색된 디바이스간에 분배하여 처리한다. 본 논문에서는 제한된 자원을 가지는 디바이스 간에 P2P를 사용하는 기법에 대해 제시하였다.

• 말소리와 음성과학
• /
• 제7권4호
• /
• pp.27-33
• /
• 2015

This paper proposes an audio event classification method using Deep Neural Networks (DNN). The proposed method applies Feed Forward Neural Network (FFNN) to generate event probabilities of ten audio events (dog barks, engine idling, and so on) for each frame. For each frame, mel scale filter bank features of its consecutive frames are used as the input vector of the FFNN. These event probabilities are accumulated for the events and the classification result is determined as the event with the highest accumulated probability. For the same dataset, the best accuracy of previous studies was reported as about 70% when the Support Vector Machine (SVM) was applied. The best accuracy of the proposed method achieves as 79.23% for the UrbanSound8K dataset when 80 mel scale filter bank features each from 7 consecutive frames (in total 560) were implemented as the input vector for the FFNN with two hidden layers and 2,000 neurons per hidden layer. In this configuration, the rectified linear unit was suggested as its activation function.

• Structural Monitoring and Maintenance
• /
• 제3권3호
• /
• pp.297-314
• /
• 2016

As the study of internal corrosion of pipeline need a large number of experiments as well as long time, so there is a need for new computational technique to expand the spectrum of the results and to save time. The present work represents a new non-destructive evaluation (NDE) technique for detecting the internal corrosion inside pipeline by evaluating the dielectric properties of steel pipe at room temperature by using electrical capacitance sensor (ECS), then predict the effect of pipeline environment temperature (${\theta}$) on the corrosion rates by designing an efficient artificial neural network (ANN) architecture. ECS consists of number of electrodes mounted on the outer surface of pipeline, the sensor shape, electrode configuration, and the number of electrodes that comprise three key elements of two dimensional capacitance sensors are illustrated. The variation in the dielectric signatures was employed to design electrical capacitance sensor (ECS) with high sensitivity to detect such defects. The rules of 24-electrode sensor parameters such as capacitance, capacitance change, and change rate of capacitance are discussed by ANSYS and MATLAB, which are combined to simulate sensor characteristic. A feed-forward neural network (FFNN) structure are applied, trained and tested to predict the finite element (FE) results of corrosion rates under room temperature, and then used the trained FFNN to predict corrosion rates at different temperature using MATLAB neural network toolbox. The FE results are in excellent agreement with an FFNN results, thus validating the accuracy and reliability of the proposed technique and leads to better understanding of the corrosion mechanism under different pipeline environmental temperature.

• 디지털콘텐츠학회 논문지
• /
• 제8권3호
• /
• pp.393-401
• /
• 2007

P2P systems have gained a lot of research interests and popularity over the years and have the capability to unleash and distribute awesome amounts of computing power, storage and bandwidths currently languishing - often underutilized - within corporate enterprises and every Internet connected home in the world. Since there is no central control over resources or devices and no before hand information about the resources or devices, device discovery remains a substantial problem in P2P environment. In this paper, we cover some of the current solutions to this problem and then propose our feed forward neural network (FFNN) based solution for device discovery in mobile P2P environment. We implements feed forward neural network (FFNN) trained with back propagation (BP) algorithm for device discovery and show, how large computation task can be distributed among such devices using agent technology. It also shows the possibility to use our architecture in home networking where devices have less storage capacity.

• Ocean Systems Engineering
• /
• 제7권4호
• /
• pp.319-328
• /
• 2017

Higher prediction efficacy is a very challenging task in any field of engineering. Due to global warming, there is a considerable increase in the global sea level. Through this work, an attempt has been made to find the sea level variability due to climate change impact at Haldia Port, India. Different statistical downscaling techniques are available and through this paper authors are intending to compare and illustrate the performances of three regression models. The models: Wavelet Neural Network (WNN), Minimax Probability Machine Regression (MPMR), Feed-Forward Neural Network (FFNN) are used for projecting the sea level variability due to climate change at Haldia Port, India. Model performance indices like PI, RMSE, NSE, MAPE, RSR etc were evaluated to get a clear picture on the model accuracy. All the indices are pointing towards the outperformance of WNN in projecting the sea level variability. The findings suggest a strong recommendation for ensembled models especially wavelet decomposed neural network to improve projecting efficiency in any time series modeling.

• International Journal of Fluid Machinery and Systems
• /
• 제3권4호
• /
• pp.369-378
• /
• 2010

The creation of the hydraulic turbine flow factor map will undoubtedly benefit its design by decreasing both the design cycle time and product cost. In this paper, the geometry and flow variables, which effectively affect the flow factor, are proposed, analyzed and determined. These flow variables are further used to create the operating condition maps by using different model approaches categorized into Response Surface Method (RSM) and Artificial Neural Network (ANN). The accuracies of models created by different approaches are compared and the performances of model approaches are analyzed. The influences of chosen variables and the combination of Principle Component Analysis (PCA) and model approaches are also studied. The comparison results between predicted and actual flow factors suggest that two-hidden-layer Feed-forward Neural Network (FFNN), and one.hidden-layer FFNN with PCA has the best performance on forming this mapping, and are accurate sufficiently for hydraulic turbine design.

• The Plant Pathology Journal
• /
• 제38권4호
• /
• pp.395-402
• /
• 2022

To predict rice blast, many machine learning methods have been proposed. As the quality and quantity of input data are essential for machine learning techniques, this study develops three artificial neural network (ANN)-based rice blast prediction models by combining two ANN models, the feed-forward neural network (FFNN) and long short-term memory, with diverse input datasets, and compares their performance. The Blast_Weathe long short-term memory r_FFNN model had the highest recall score (66.3%) for rice blast prediction. This model requires two types of input data: blast occurrence data for the last 3 years and weather data (daily maximum temperature, relative humidity, and precipitation) between January and July of the prediction year. This study showed that the performance of an ANN-based disease prediction model was improved by applying suitable machine learning techniques together with the optimization of hyperparameter tuning involving input data. Moreover, we highlight the importance of the systematic collection of long-term disease data.

• 디지털콘텐츠학회 논문지
• /
• 제18권1호
• /
• pp.183-190
• /
• 2017

본 논문에서는 깊은 신경망을 이용한 오디오 이벤트 검출 방법을 제안한다. 오디오 입력의 매 프레임에 대한 오디오 이벤트 확률을 feed-forward 신경망을 적용하여 생성한다. 매 프레임에 대하여 멜 스케일 필터 뱅크 특징을 추출한 후, 해당 프레임의 전후 프레임으로부터의 특징벡터들을 하나의 특징벡터로 결합하고 이를 feed-forward 신경망의 입력으로 사용한다. 깊은 신경망의 출력층은 입력 프레임 특징값에 대한 오디오 이벤트 확률값을 나타낸다. 연속된 5개 이상의 프레임에서의 이벤트 확률값이 임계값을 넘을 경우 해당 구간이 오디오 이벤트로 검출된다. 검출된 오디오 이벤트는 1초 이내에 동일 이벤트로 검출되는 동안 하나의 오디오 이벤트로 유지된다. 제안된 방법으로 구현된 오디오 이벤트 검출기는 UrbanSound8K와 BBC Sound FX자료에서의 20개 오디오 이벤트에 대하여 71.8%의 검출 정확도를 보였다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제9권10호
• /
• pp.309-316
• /
• 2020

우리는 응급실을 방문한 65세 이상 노인환자의 의료 데이터를 각각 피드 포워드 신경망과 합성곱 신경망에 학습하여 사망률을 예측하였다. 의료 데이터는 노인환자의 성별, 연령, 체온, 심박 수 등의 기초적인 정보뿐 아니라 과거 병력, 다양한 혈액 검사 및 배양 검사 결과 등 다양하고 복잡한 정보를 포함하여 총 99가지의 자질로 구성된다. 이 중 사망률 예측에 크게 기여하는 자질을 선택하기 위해 랜덤 포레스트를 이용하여 자질의 중요도를 계산하였고, 그 결과 중요도가 높은 상위 80개의 자질을 선택하였다. 선택된 자질을 각각 피드 포워드 신경망과 합성곱 신경망의 학습에 사용하여 두 신경망의 성능을 비교하였다. 합성곱 신경망 학습을 위해 의료 데이터를 고정된 크기의 이미지로 변환하였으며 합성곱 신경망이 피드 포워드 신경망을 이용한 것보다 성능이 좋았다. 합성곱 신경망의 사망률 예측 성능으로 테스트 데이터에 대해 F1 점수는 56.9, AUC는 92.1을 각각 얻었다.