• 한국정보과학회논문지:소프트웨어및응용
• /
• 제31권5호
• /
• pp.537-542
• /
• 2004

다층 퍼셉트론(Multi-Layer Perceptron, MLP) 구조는 그의 비선형 적합능력으로 인하여 매우 다양한 실제 문제에 적용되고 있다. 그러나 일반화된 MLP 구조의 적합능력은 은닉노드의 개수. 초기 가중 값 그리고 학습 회수 또는 학습 오차와 같은 구조인자(factor)들에 크게 영향을 받는다. 만약 이들 구조인자가 부적절하게 선택되면 일반화된 MLP 구조의 적합능력이 매우 왜곡될 수 있다. 따라서 MLP구조에 영향을 주는 인자들의 결합 영향을 살펴보는 것은 중요한 문제이다. 이 논문에서는 제어상자(controller box)를 통한 학습결과와 더불어 MLP구조를 일반화할 때 영향을 줄 수 있는 신경망의 일반적인 구조인자 들을 실증적으로 살펴보고 이들의 상대효과를 비교한다.

• 융합신호처리학회논문지
• /
• 제5권2호
• /
• pp.132-137
• /
• 2004

본 논문에서는 다층 신경회로망의 구조를 가지며, 백프로퍼게이션 학습 알고리즘을 이용한 적응신호처리 시스템을 구현하였다. 최소자승 알고리즘을 이용한 적응 잡음 제거기는 기준 신호와 잡음과의 상관도에 영향을 많이 받고, 정보 신호가 잡음에 비하여 상대적으로 작은 경우에 한계를 보이고 있다. 이와 같은 잡음에 대하여 본 논문에서 제안된 시스템은 좋은 성능을 보인다. 또한, 은닉층의 수와 노드 수를 다르게 구성했을 경우에 시스템의 출력에 미치는 결과에 대하여 분석하였다. 제안된 적응 신호처리 시스템의 장점을 알아보기 위하여 성능 평가의 기준이 되는 최소자승 알고리즘을 이용한 시스템과 비교하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권12호
• /
• pp.3152-3165
• /
• 2012

Wireless sensor networks (WSN) often employ asynchronous MAC scheduling, which allows each sensor node to wake up independently without synchronizing with its neighbor nodes. However, this asynchronous scheduling may not deal with collisions due to hidden terminals effectively. Although most of the existing asynchronous protocols exploit a random back-off technique to resolve collisions, the random back-off cannot secure a receiver from potentially repetitive collisions and may lead to a substantial increase in the packet latency. In this paper, we propose a new collision resolution algorithm called Transient Coordinator (TC) for asynchronous WSN MAC protocols. TC resolves a collision on demand by ordering senders' transmissions when a receiver detects a collision. To coordinate the transmission sequence both the receiver and the collided senders perform handshaking to collect the information and to derive a collision-free transmission sequence, which enables each sender to exclusively access the channel. According to the simulation results, our scheme can improve the average per-node throughput by up to 19.4% while it also reduces unnecessary energy consumption due to repetitive collisions by as much as 91.1% compared to the conventional asynchronous MAC protocols. This demonstrates that TC is more efficient in terms of performance, resource utilization, and energy compared to the random back-off scheme in dealing with collisions for asynchronous WSN MAC scheduling.

• 한국기계가공학회지
• /
• 제18권9호
• /
• pp.52-57
• /
• 2019

The key technology of the fourth industrial revolution is artificial intelligence and machine learning. In this study, FMEA was performed on fuel pumps used as key items in most systems to identify major failure components, and artificial neural networks were built using big data. The main failure mode of the fuel pump identified by the test was coil damage due to overheating. Based on the artificial neural network built, machine learning was conducted to predict the failure and the mean error rate was 4.9% when the number of hidden nodes in the artificial neural network was three and the temperature increased to $140^{\circ}C$ rapidly.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.2162-2179
• /
• 2019

The performance of text classification is highly related to the feature selection methods. Usually, two tasks are performed when a feature selection method is applied to construct a feature set; 1) assign score to each feature and 2) select the top-N features. The selection of top-N features in the existing filter-based feature selection methods is biased by their discriminative power and the empirical process which is followed to determine the value of N. In order to improve the text classification performance by presenting a more illustrative feature set, we present an approach via a potent representation learning technique, namely DBN (Deep Belief Network). This algorithm learns via the semantic illustration of documents and uses feature vectors for their formulation. The nodes, iteration, and a number of hidden layers are the main parameters of DBN, which can tune to improve the classifier's performance. The results of experiments indicate the effectiveness of the proposed method to increase the classification performance and aid developers to make effective decisions in certain domains.

• 한국컴퓨터정보학회논문지
• /
• 제23권12호
• /
• pp.43-48
• /
• 2018

In this paper, more efficient classification result could be obtained by applying the combination of the Hidden Markov Model and SVM Model to HMSV algorithm gene expression data which simulated the stochastic flow of gene data and clustering it. In this paper, we verified the HMSV algorithm that combines independently learned algorithms. To prove that this paper is superior to other papers, we tested the sensitivity and specificity of the most commonly used classification criteria. As a result, the K-means is 71% and the SOM is 68%. The proposed HMSV algorithm is 85%. These results are stable and high. It can be seen that this is better classified than using a general classification algorithm. The algorithm proposed in this paper is a stochastic modeling of the generation process of the characteristics included in the signal, and a good recognition rate can be obtained with a small amount of calculation, so it will be useful to study the relationship with diseases by showing fast and effective performance improvement with an algorithm that clusters nodes by simulating the stochastic flow of Gene Data through data mining of BigData.

• Nuclear Engineering and Technology
• /
• 제51권3호
• /
• pp.723-730
• /
• 2019

Acquiring instrumentation signals generated from nuclear power plants (NPPs) is essential to maintain nuclear reactor integrity or to mitigate an abnormal state under normal operating conditions or severe accident circumstances. However, various safety-critical instrumentation signals from NPPs cannot be accurately measured on account of instrument degradation or failure under severe accident circumstances. Reactor vessel (RV) water level, which is an accident monitoring variable directly related to reactor cooling and prevention of core exposure, was predicted by applying a few signals to deep neural networks (DNNs) during severe accidents in NPPs. Signal data were obtained by simulating the postulated loss-of-coolant accidents at hot- and cold-legs, and steam generator tube rupture using modular accident analysis program code as actual NPP accidents rarely happen. To optimize the DNN model for RV water level prediction, a genetic algorithm was used to select the numbers of hidden layers and nodes. The proposed DNN model had a small root mean square error for RV water level prediction, and performed better than the cascaded fuzzy neural network model of the previous study. Consequently, the DNN model is considered to perform well enough to provide supporting information on the RV water level to operators.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2020년도 학술발표회
• /
• pp.170-170
• /
• 2020

Due to the development of technologies, complex computation of huge data set is possible with a prevalent personal computer. Therefore, machine learning methods have been widely applied in the hydrologic field such as regression-based regional frequency analysis (RFA). The main purpose of this study is to compare two frameworks of RFA based on the artificial neural network (ANN) models: quantile regression technique (QRT-ANN) and parameter regression technique (PRT-ANN). As an output layer of the ANN model, the QRT-ANN predicts quantiles for various return periods whereas the PRT-ANN provides prediction of three parameters for the generalized extreme value distribution. Rainfall gauging sites where record length is more than 20 years were selected and their annual maximum rainfalls and various hydro-meteorological variables were used as an input layer of the ANN model. While employing the ANN model, 70% and 30% of gauging sites were used as training set and testing set, respectively. For each technique, ANN model structure such as number of hidden layers and nodes was determined by a leave-one-out validation with calculating root mean square error (RMSE). To assess the performances of two frameworks, RMSEs of quantile predicted by the QRT-ANN are compared to those of the PRT-ANN.

• Journal of Information Processing Systems
• /
• 제20권2호
• /
• pp.200-214
• /
• 2024

Amid the flood of data, social network analysis is beneficial in searching for its hidden context and verifying several pieces of information. This can be used for detecting the spread model of infectious diseases, methods of preventing infectious diseases, mining of small groups and so forth. In addition, community detection is the most studied topic in social network analysis using graph analysis methods. The objective of this study is to examine signed attributed social networks and identify the maximal balanced cliques that are both absolute and fair. In the same vein, the purpose is to ensure fairness in complex networks, overcome the "information cocoon" bottleneck, and reduce the occurrence of "group polarization" in social networks. Meanwhile, an empirical study is presented in the experimental section, which uses the personal information of 77 employees of a research company and the trust relationships at the professional level between employees to mine some small groups with the possibility of "group polarization." Finally, the study provides suggestions for managers of the company to align and group new work teams in an organization.

• Steel and Composite Structures
• /
• 제51권4호
• /
• pp.417-440
• /
• 2024

Artificial neural networks (ANN) have been the focus of several studies when it comes to evaluating the pile's bearing capacity. Nonetheless, the principal drawbacks of employing this method are the sluggish rate of convergence and the constraints of ANN in locating global minima. The current work aimed to build four ANN-based prediction models enhanced with methods from the black hole algorithm (BHA), league championship algorithm (LCA), shuffled complex evolution (SCE), and symbiotic organisms search (SOS) to estimate the carrying capacity of piles in cold climates. To provide the crucial dataset required to build the model, fifty-eight concrete pile experiments were conducted. The pile geometrical properties, internal friction angle 𝛗 shaft, internal friction angle 𝛗 tip, pile length, pile area, and vertical effective stress were established as the network inputs, and the BHA, LCA, SCE, and SOS-based ANN models were set up to provide the pile bearing capacity as the output. Following a sensitivity analysis to determine the optimal BHA, LCA, SCE, and SOS parameters and a train and test procedure to determine the optimal network architecture or the number of hidden nodes, the best prediction approach was selected. The outcomes show a good agreement between the measured bearing capabilities and the pile bearing capacities forecasted by SCE-MLP. The testing dataset's respective mean square error and coefficient of determination, which are 0.91846 and 391.1539, indicate that using the SCE-MLP approach as a practical, efficient, and highly reliable technique to forecast the pile's bearing capacity is advantageous.