• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.36 no.3
• /
• pp.169-176
• /
• 2020

The authors developed a nodal network model that simulates the flow of moist air and the thermal behavior of a target area. The nodal network model was enhanced using a parameter estimation technique based on the measured temperature, humidity, and schedule data. However, the nodal model is not good enough for predicting humidity of the target space, having 55.6% of CVRMSE. It is because re-evaporation effect could not be modeled due to uncertain factors in the field measurement. Hence, a data-driven model was introduced using an artificial neural network (ANN). It was found that the data-driven model is suitable for predicting the condensation compared to the nodal model satisfying ASHRAE Guideline with 3.36% of CVRMSE for temprature, relative humidity, and surface temperature on average. The model will be embedded in automated devices for real-time predictive control, to minimize the risk of surface condensation at dressroom in an apartment housing.

• MALSORI
• /
• no.50
• /
• pp.51-69
• /
• 2004

In this paper, we present a one-pass semi-dynamic network decoding framework that inherits both advantages of fast decoding speed from static network decoders and memory efficiency from dynamic network decoders. Our method is based on the novel language model network representation that is essentially of finite state machine (FSM). The static network derived from the language model network [1][2] is partitioned into smaller subnetworks which are static by nature or self-structured. The whole network is dynamically managed so that those subnetworks required for decoding are cached in memory. The network is near-minimized by applying the tail-sharing algorithm. Our decoder is evaluated on the 25k-word Korean broadcast news transcription task. In case of the search network itself, the network is reduced by 73.4% from the tail-sharing algorithm. Compared with the equivalent static network decoder, the semi-dynamic network decoder has increased at most 6% in decoding time while it can be flexibly adapted to the various memory configurations, giving the minimal usage of 37.6% of the complete network size.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.8
• /
• pp.3621-3640
• /
• 2016

In the process of k-coverage of the target node, there will be a lot of data redundancy forcing the phenomenon of congestion which reduces network communication capability and coverage, and accelerates network energy consumption. Therefore, this paper proposes a novel energy balanced k-coverage control algorithm based on probability model (EBKCCA). The algorithm constructs the coverage network model by using the positional relationship between the nodes. By analyzing the network model, the coverage expected value of nodes and the minimum number of nodes in the monitoring area are given. In terms of energy consumption, this paper gives the proportion of energy conversion functions between working nodes and neighboring nodes. By using the function proportional to schedule low energy nodes, we achieve the energy balance of the whole network and optimizing network resources. The last simulation experiments indicate that this algorithm can not only improve the quality of network coverage, but also completely inhibit the rapid energy consumption of node, and extend the network lifetime.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.13 no.6
• /
• pp.48-55
• /
• 2005

In this paper, a mathematical model and a simulation method for the response time analysis of Local Interconnect Network(LIN) based network systems are proposed. Network-induced delays in a network based control system can vary widely according to the transmission time of message and the overhead time of transmission. Therefore, in order to design a distributed control system using LIN network, a method to predict and verify the timing behavior of LIN protocol is required at the network design phase. Furthermore, a simulation environment based on a timing model of LIN protocol is beneficial to predict the timing behavior of LIN. The model equation is formulated with six timing parameters deduced from timing properties of LIN specification. Additionally, LIN conformance test equations to verify LIN device driver are derived with timing constraints of the parameters. The proposed model equation and simulation method are validated with a result that is measured at real LIN based network system.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2009.03a
• /
• pp.456-463
• /
• 2009

The preconsolidation pressure has been commonly determined by oedometer test. However, it can also be determined by in-situ test, such as piezocone test with theoretical and(or) empirical correlations. Recently, Neural Network(NN) theory was applied and some models were proposed to estimate the preconsolidation pressure or OCR. However, since the optimization process of synaptic weights of NN model is dependent on the initial synaptic weights, NN models which are trained with different initial weights can't avoid the variability on prediction result for new database even though they have same structure and use same transfer function. In this study, Committee Neural Network(CNN) model is proposed to improve the initial weight dependency of multi-layered neural network model on the prediction of preconsolidation pressure of soft clay from piezocone test result. It was found that even though the NN model has the optimized structure for given training data set, it still has the initial weight dependency, while the proposed CNN model can improve the initial weight dependency of the NN model and provide a consistent and precise inference result than existing NN models.

• Korean Journal of Agricultural Science
• /
• v.47 no.4
• /
• pp.769-782
• /
• 2020

Interest in fruit and vegetables has increased due to changes in consumer consumption patterns, socioeconomic status, and family structure. This study determined the factors influencing the demand for fruit and vegetables (strawberries, paprika, tomatoes and cherry tomatoes) using a panel of Rural Development Administration household-level purchases from 2010 to 2018 and compared the ability to the prediction performance. An artificial neural network model was constructed, linking household characteristics with final food expenditure. Comparing the analysis results of the artificial neural network with the results of the panel model showed that the artificial neural network accurately predicted the pattern of the consumer panel data rather than the fixed effect model. In addition, the prediction for strawberries was found to be heavily affected by the number of families, retail places and income, while the prediction for paprika was largely affected by income, age and retail conditions. In the case of the prediction for tomatoes, they were greatly affected by age, income and place of purchase, and the prediction for cherry tomatoes was found to be affected by age, number of families and retail conditions. Therefore, a more accurate analysis of the consumer consumption pattern was possible through the artificial neural network model, which could be used as basic data for decision making.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.1
• /
• pp.115-123
• /
• 2016

In this study, we collect various side effect pairs which are appeared frequently at many drugs, and select side effect pairs that have higher severity. For every selected side effect pair, we extract common genetic networks which are shared by side effects' genes and drugs' target genes based on PPI(Protein-Protein Interaction) network. For this work, firstly, we gather drug related data, side effect data and PPI data. Secondly, for extracting common genetic network, we find shortest paths between drug target genes and side effect genes based on PPI network, and integrate these shortest paths. Thirdly, we develop a classification model which uses this common genetic network as a classifier. We calculate similarity score between the common genetic network and genetic network of a drug for classifying the drug. Lastly, we validate our classification model by means of AUC(Area Under the Curve) value.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.2
• /
• pp.91-97
• /
• 2019

In automatic map generalization, the formalization of cartographic principles is important. This study proposes and evaluates the selection method for road network generalization that analyzes existing maps using reverse engineering and formalizes the selection rules for the road network. Existing maps with a 1:5,000 scale and a 1:25,000 scale are compared, and the criteria for selection of the road network data and the relative importance of each network object are determined and analyzed using $T{\ddot{o}}pfer^{\prime}s$ Radical Law as well as the logistic regression model. The selection model derived from the analysis result is applied to the test data, and road network data for the 1:25,000 scale map are generated from the digital topographic map on a 1:5,000 scale. The selected road network is compared with the existing road network data on the 1:25,000 scale for a qualitative and quantitative evaluation. The result indicates that more than 80% of road objects are matched to existing data.

• Journal of information and communication convergence engineering
• /
• v.19 no.4
• /
• pp.241-247
• /
• 2021

Particulate matter has emerged as a serious global problem, necessitating highly reliable information on the matter. Therefore, various algorithms have been used in studies to predict particulate matter. In this study, we compared the prediction performance of neural network models that have been actively studied for particulate matter prediction. Among the neural network algorithms, a deep neural network (DNN), a recurrent neural network, and long short-term memory were used to design the optimal prediction model using a hyper-parameter search. In the comparative analysis of the prediction performance of each model, the DNN model showed a lower root mean square error (RMSE) than the other algorithms in the performance comparison using the RMSE and the level of accuracy as metrics for evaluation. The stability of the recurrent neural network was slightly lower than that of the other algorithms, although the accuracy was higher.

• Journal of Digital Convergence
• /
• v.12 no.8
• /
• pp.309-319
• /
• 2014

Virtual Network service is a key characteristics of future Internet which is debate internationally. There are two kinds of network virtualization technologies considered lately: network functions virtualization and virtual network approaches. Several national wide research networks including US's GENI project have experimented technologies for future Internet and in particular network virtualization is one of key issues. Representative Korean research network, KREONET, is working on deploying virtual network framework as a preliminary for future Ineternet using the virtualization model of SDN/OpenFlow which is typical network model of future Internet. This paper proposes a stepwise model to bring virtual network services in KREONET. Firstly, we requirements of KREONET users' virtual network service and network resource management and network deploying virtual network. Finally, we verify the adequacy of our virtual network model for KREONET.