• Journal of Internet Computing and Services
• /
• v.20 no.5
• /
• pp.1-8
• /
• 2019

This paper carried out a study to reduce damage from jellyfish whose population has increased due to global warming. The emergence of jellyfish on the beach could result in casualties from jellyfish stings and economic losses from closures. This paper confirmed from the preceding studies that the pattern of jellyfish's appearance is predictable through machine learning. This paper is an extension of The prediction model of emergence of Busan coastal jellyfish using SVM. In this paper, we used deep neural network to expand from the existing methods of predicting the existence of jellyfish to the classification by index. Due to the limitations of the small amount of data collected, the 84.57% prediction accuracy limit was sought to be resolved through data expansion using bootstraping. The expanded data showed about 7% higher performance than the original data, and about 6% better performance compared to the transfer learning. Finally, we used the test data to confirm the prediction performance of jellyfish appearance. As a result, although it has been confirmed that jellyfish emergence binary classification can be predicted with high accuracy, predictions through indexation have not produced meaningful results.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.1
• /
• pp.103-116
• /
• 2021

Analysis of runoff is substantial for effective water management in the watershed. Runoff occurs by reaction of a watershed to the rainfall and has non-linearity and uncertainty due to the complex relation of weather and watershed factors. ANN (Artificial Neural Network), which learns from the data, is one of the machine learning technique known as a proper model to interpret non-linear data. The performance of ANN is affected by the ANN's structure, the number of hidden layer nodes, learning rate, and activation function. Especially, the activation function has a role to deliver the information entered and decides the way of making output. Therefore, It is important to apply appropriate activation functions according to the problem to solve. In this paper, ANN models were constructed to estimate runoff with different activation functions and each model was compared and evaluated. Sigmoid, Hyperbolic tangent, ReLU (Rectified Linear Unit), ELU (Exponential Linear Unit) functions were applied to the hidden layer, and Identity, ReLU, Softplus functions applied to the output layer. The statistical parameters including coefficient of determination, NSE (Nash and Sutcliffe Efficiency), NSEln (modified NSE), and PBIAS (Percent BIAS) were utilized to evaluate the ANN models. From the result, applications of Hyperbolic tangent function and ELU function to the hidden layer and Identity function to the output layer show competent performance rather than other functions which demonstrated the function selection in the ANN structure can affect the performance of ANN.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.11
• /
• pp.1500-1506
• /
• 2020

In this paper, we study a scheduling problem based on reinforcement learning for overlay device-to-device (D2D) communication networks. Even though various technologies for D2D communication networks using Q-learning, which is one of reinforcement learning models, have been studied, Q-learning causes a tremendous complexity as the number of states and actions increases. In order to solve this problem, D2D communication technologies based on Deep Q Network (DQN) have been studied. In this paper, we thus design a DQN model by considering the characteristics of wireless communication systems, and propose a distributed scheduling scheme based on the DQN model that can reduce feedback and signaling overhead. The proposed model trains all parameters in a centralized manner, and transfers the final trained parameters to all mobiles. All mobiles individually determine their actions by using the transferred parameters. We analyze the performance of the proposed scheme by computer simulation and compare it with optimal scheme, opportunistic selection scheme and full transmission scheme.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.6
• /
• pp.1013-1021
• /
• 2020

Malware Authorship Attribution is a research field for identifying malware by comparing the author characteristics of unknown malware with the characteristics of known malware authors. The authorship attribution method using binaries has the advantage that it is easy to collect and analyze targeted malicious codes, but the scope of using features is limited compared to the method using source code. This limitation has the disadvantage that accuracy decreases for a large number of authors. This study proposes a method of 'Defining semantic features from binaries' and 'Defining allowable ranges for redundant features using the concept of survival network' to complement the limitations in the identification of binary authors. The proposed method defines Opcode-based graph features from binary information, and defines the allowable range for selecting unique features for each author using the concept of a survival network. Through this, it was possible to define the feature definition and feature selection method for each author as a single technology, and through the experiment, it was confirmed that it was possible to derive the same level of accuracy as the source code-based analysis with an improvement of 5.0% accuracy compared to the previous study.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.23-31
• /
• 2022

This study analyzed the actual frozen container operation data of Starcool provided by H Shipping. Through interviews with H's field experts, only Critical and Fatal Alarms among the four failure alarms were defined as failures, and it was confirmed that using all variables due to the nature of frozen containers resulted in cost inefficiency. Therefore, this study proposes a method for detecting failure of frozen containers through characteristic importance and PCA techniques. To improve the performance of the model, we select variables based on feature importance through tree series models such as XGBoost and LGBoost, and use PCA to reduce the dimension of the entire variables for each model. The boosting-based XGBoost and LGBoost techniques showed that the results of the model proposed in this study improved the reproduction rate by 0.36 and 0.39 respectively compared to the results of supervised learning using all 62 variables.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.33-41
• /
• 2022

In this article, we proposed to predict natural gas (NG) leakage levels through feature selection based on a factor analysis (FA) of the integrating the Korean Meteorological Agency data and natural gas leakage data for considering complex factors. The paper has been divided into three modules. First, we filled missing data based on the linear interpolation method on the integrated data set, and selected essential features using FA with OrdinalEncoder (OE)-based normalization. The dataset is labeled by K-means clustering. The final module uses four algorithms, K-nearest neighbors (KNN), decision tree (DT), random forest (RF), Naive Bayes (NB), to predict gas leakage levels. The proposed method is evaluated by the accuracy, area under the ROC curve (AUC), and mean standard error (MSE). The test results indicate that the OrdinalEncoder-Factor analysis (OE-F)-based classification method has improved successfully. Moreover, OE-F-based KNN (OE-F-KNN) showed the best performance by giving 95.20% accuracy, an AUC of 96.13%, and an MSE of 0.031.

• Proceedings of the ESK Conference
• /
• 1996.10a
• /
• pp.47-55
• /
• 1996

This study developed a comprehensive and easily applicable nuclear reactor control system evaluation method using reactor operators behavioral and mental workload database. A proposed control panel design cycle consists of the 5 steps: (1) finding out inconvenient, erroneous, and mentally stressful factors for the proposed design through evaluative experiments, (2) drafting improved design alternatives considering detective factors found out in the step (1), (3) comparative experiements for the design alternatives, (4) selecting a best design alternative, (5) returning to the step (1) and repeating the design cycle. Reactor operators behavioral and mental workload database collected from evaluative experiments in the step (1) and comparative experiments in the step (3) of the design cycle have a key roll in finding out defective factors and yielding the criteria for selection of the proposed reactor control systems. The behavioral database was designed to include the major informations about reactor operators' control behaviors: beginning time of operations, involved displays, classification of observational behaviors, dehaviors, decisions, involved control devices, classification of control behaviors, communications, emotional status, opinions for man-machine interface, and system event log. The database for mental workload scored from various physiological variables-EEG, EOG, ECG, and respir- ation pattern-was developed to indicate the most stressful situation during reactor control operations and to give hints for defective design factors. An experimental test for the evaluation method applied to the Compact Nuclear Simulator (CNS) installed in Korea Atomic Energy Research Institute (KAERI) suggested that some defective design factors of analog indicators should be improved and that automatization of power control to a target level would give relaxation to the subject operators in stressful situation.

• Korean Journal of Agricultural Science
• /
• v.49 no.2
• /
• pp.269-284
• /
• 2022

The development of radish collectors has the potential to increase radish yields while decreasing the time and dependence on human labor in a variety of field activities. Stress and fatigue analyses are essential to ensure the optimal design and machine life of any agricultural machinery. The objectives of this research were to analyze the stress and fatigue of major components of a tractor-mounted radish collector under dynamic load conditions in an effort to increase the design dependability and dimensions of the materials. An experiment was conducted to measure the shaft torque of stem-cutting and transferring conveyor motors using rotary torque sensors at different tractor ground speeds with and without a load. The Smith-Watson-Topper mean stress equation and the rain-flow counting technique were utilized to determine the required shear stress with the distribution of the fatigue life cycle. The severity of the operation was assessed using Miner's theory. All running conditions produced more than 10⁷ of high cycle fatigue strength. Furthermore, the highest severity levels for motor shafts used for stem cutting and transferring and for transportation joints and cutting blades were 2.20, 4.24, 2.07, and 1.07, and 1.97, 3.81, 1.73, and 1.07, respectively, with and without a load condition, except for 5.24 for a winch motor shaft under a load. The stress and fatigue analysis presented in this study can aid in the selection of the most appropriate design parameters and material sizes for the successful construction of a tractor-mounted radish collector, which is currently under development.

• Journal of Korea Society of Industrial Information Systems
• /
• v.14 no.4
• /
• pp.1-7
• /
• 2009

This paper presents a process routing (PR) algorithm with multiple objectives. PR determines the optimum sequence of operations for transforming a raw material into a completed part within the available machining resources. In any computer aided process planning (CAPP) system, selection of the machining operation sequence is one of the most critical activities for manufacturing a part and for the technical specification in the part drawing. Here, the goal could be to generate the sequence that optimizes production time, production cost, machine utilization or with multiple these criteria. The Pareto Stratum Niche Cubicle (PS NC) GA has been adopted to find the optimum sequence of operations that optimize two conflicting criteria; production cost and production quality. The numerical analysis shows that the proposed PS NC GA is both effective and efficient to the PR problem.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2022.10a
• /
• pp.259-259
• /
• 2022

Soil salinity is a major factor that reduces crop yields. The amount of soil affected by salinity is about 83 million hectares (FAO 2000), which is increasing due to the effects of climate change. In soybean [Glycine max (L.) Merr.], nutritional properties such as protein, starch, and sucrose content together with biomass and yield tends to reduce due to excessive salt. As a result of QTL mapping using the 169 F_2:3 population from the KA-1285 (salt-tolerant) × Daepung (salt-sensitive) in a previous study, two major QTLs (Gm03_39796778 and Gm03_40600088) related to salt tolerance were found on chromosome 3. In this study, the CDS region of the Gmsalt3 gene was analyzed using the ABI 3730x1 DNA Analyzer (Macrogen, Korea). The sequence of Gmsalt3 gene in KA-1285 was compared with Williams 82.a4.vl and PI483463 (Glycine soja). Two transversions were found at exon6 in KA-1285 and PI483463. Currently, whole genome sequencing and variation analysis using the Illumine Novaseq 6000 machine (Illumina, USA) are in progress. The results of this study can provide useful molecular markers for the selection of salt-tolerant soybeans and can be used as basic data for future salt-tolerant gene research.