• The Korean Journal of Applied Statistics
• /
• v.35 no.6
• /
• pp.685-701
• /
• 2022

FDR is widely used in high-dimensional data inference since it provides more liberal criterion contrary to FWER which is known to be very conservative by controlling Type-1 errors. This paper proposes a sparse VHAR model estimation method controlling FDR by adapting the knockoff introduced by Barber and Candès (2015). We also compare knockoff with conventional method using adaptive Lasso (AL) through extensive simulation study. We observe that AL shows sparsistency and decent forecasting performance, however, AL is not satisfactory in controlling FDR. To be more specific, AL tends to estimate zero coefficients as non-zero coefficients. On the other hand, knockoff controls FDR sufficiently well under desired level, but it finds too sparse model when the sample size is small. However, the knockoff is dramatically improved as sample size increases and the model is getting sparser.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.51-62
• /
• 2022

AI-based speed prediction studies have been conducted quite actively. However, while the importance of explainable AI is emerging, the study of interpreting and reasoning the AI-based speed predictions has not been carried out much. Therefore, in this paper, 'Explainable Deep Graph Neural Network (GNN)' is devised to analyze the speed prediction and assess the nearby road influence for reasoning the critical contributions to a given road situation. The model's output was explained by comparing the differences in output before and after masking the input values of the GNN model. Using TOPIS traffic speed data, we applied our GNN models for the major congested roads in Seoul. We verified our approach through a traffic flow simulation by adjusting the most influential nearby roads' speed and observing the congestion's relief on the road of interest accordingly. This is meaningful in that our approach can be applied to the transportation network and traffic flow can be improved by controlling specific nearby roads based on the inference results.

• Smart Media Journal
• /
• v.12 no.9
• /
• pp.89-94
• /
• 2023

Recommender systems provide users with the most favorable items by analyzing explicit or implicit feedback of users on items. Recently, as the size of deep-learning-based models employed in recommender systems has increased, many studies have focused on reducing inference time while maintaining high recommendation accuracy. As one of them, a study on recommender systems with a knowledge distillation (KD) technique is actively conducted. By KD, a small-sized model (i.e., student) is trained through knowledge extracted from a large-sized model (i.e., teacher), and then the trained student is used as a recommendation model. Existing studies on KD for recommender systems have been mainly performed only for implicit feedback settings. Thus, in this paper, we try to investigate the performance and accuracy when applied to explicit feedback settings. To this end, we leveraged a total of five state-of-the-art KD methods and three real-world datasets for recommender systems.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.26 no.6
• /
• pp.85-96
• /
• 2023

This study proposes a method for forest vegetation monitoring using high-resolution aerial imagery captured by unmanned aerial vehicles(UAV) and deep learning technology. The research site was selected in the forested area of Mountain Dogo, Asan City, Chungcheongnam-do, and the target species for monitoring included Pinus densiflora, Quercus mongolica, and Quercus acutissima. To classify vegetation species at the pixel level in UAV imagery based on characteristics such as leaf shape, size, and color, the study employed the semantic segmentation method using the prominent U-net deep learning model. The research results indicated that it was possible to visually distinguish Pinus densiflora Siebold & Zucc, Quercus mongolica Fisch. ex Ledeb, and Quercus acutissima Carruth in 135 aerial images captured by UAV. Out of these, 104 images were used as training data for the deep learning model, while 31 images were used for inference. The optimization of the deep learning model resulted in an overall average pixel accuracy of 92.60, with mIoU at 0.80 and FIoU at 0.82, demonstrating the successful construction of a reliable deep learning model. This study is significant as a pilot case for the application of UAV and deep learning to monitor and manage representative species among climate-vulnerable vegetation, including Pinus densiflora, Quercus mongolica, and Quercus acutissima. It is expected that in the future, UAV and deep learning models can be applied to a variety of vegetation species to better address forest management.

• Smart Structures and Systems
• /
• v.20 no.1
• /
• pp.61-74
• /
• 2017

One of the most reliable and simplest tools for structural vibration control in civil engineering is Tuned Mass Damper, TMD. Provided that the frequency and damping parameters of these dampers are tuned appropriately, they can reduce the vibrations of the structure through their generated inertia forces, as they vibrate continuously. To achieve the optimal parameters of TMD, many different methods have been provided so far. In old approaches, some formulas have been offered based on simplifying models and their applied loadings while novel procedures need to model structures completely in order to obtain TMD parameters. In this paper, with regard to the nonlinear decision-making of fuzzy systems and their enough ability to cope with different unreliability, a method is proposed. Furthermore, by taking advantage of both old and new methods a fuzzy system is designed to be operational and reduce uncertainties related to models and applied loads. To design fuzzy system, it is required to gain data on structures and optimum parameters of TMDs corresponding to these structures. This information is obtained through modeling MDOF systems with various numbers of stories subjected to far and near field earthquakes. The design of the fuzzy systems is performed by three methods: look-up table, the data space grid-partitioning, and clustering. After that, rule weights of Mamdani fuzzy system using the look-up table are optimized through genetic algorithm and rule weights of Sugeno fuzzy system designed based on grid-partitioning methods and clustering data are optimized through ANFIS (Adaptive Neuro-Fuzzy Inference System). By comparing these methods, it is observed that the fuzzy system technique based on data clustering has an efficient function to predict the optimal parameters of TMDs. In this method, average of errors in estimating frequency and damping ratio is close to zero. Also, standard deviation of frequency errors and damping ratio errors decrease by 78% and 4.1% respectively in comparison with the look-up table method. While, this reductions compared to the grid partitioning method are 2.2% and 1.8% respectively. In this research, TMD parameters are estimated for a 15-degree of freedom structure based on designed fuzzy system and are compared to parameters obtained from the genetic algorithm and empirical relations. The progress up to 1.9% and 2% under far-field earthquakes and 0.4% and 2.2% under near-field earthquakes is obtained in decreasing respectively roof maximum displacement and its RMS ratio through fuzzy system method compared to those obtained by empirical relations.

• Economic and Environmental Geology
• /
• v.36 no.3
• /
• pp.243-255
• /
• 2003

The mathematical models for GIS-based spatial data integration have been developed for geological applications such as mineral potential mapping or landslide susceptibility analysis. Among various models, the effectiveness of fuzzy logic based integration of multiple sets of geological data is investigated and discussed. Unlike a traditional target-driven fuzzy integration approach, we propose a data-driven approach that is derived from statistical relationships between the integration target and related spatial geological data. The proposed approach consists of four analytical steps; data representation, fuzzy combination, defuzzification and validation. For data representation, the fuzzy membership functions based on the likelihood ratio functions are proposed. To integrate them, the fuzzy inference network is designed that can combine a variety of different fuzzy operators. Defuzzification is carried out to effectively visualize the relative possibility levels from the integrated results. Finally, a validation approach based on the spatial partitioning of integration targets is proposed to quantitatively compare various fuzzy integration maps and obtain a meaningful interpretation with respect to future events. The effectiveness and some suggestions of the schemes proposed here are illustrated by describing a case study for landslide susceptibility analysis. The case study demonstrates that the proposed schemes can effectively identify areas that are susceptible to landslides and ${\gamma}$ operator shows the better prediction power than the results using max and min operators from the validation procedure.

• Journal of Korean Society of Transportation
• /
• v.19 no.5
• /
• pp.119-134
• /
• 2001

This paper proposed a dynamic transit vehicle simulation model and a dynamic transit passengers simulation model, which can simultaneously simulate the transit vehicles and passengers traveling on a transit network, and also developed an algorithm of dynamic departure time choice model based on individual passenger. The proposed model assumes that each passenger's behavior is heterogeneous based on stochastic process by relaxing the assumption of homogeneity among passengers and travelers have imperfect information and bounded rationality to more actually represent and to simulate each passenger's behavior. The proposed model integrated a inference and preference reforming procedure into the learning and decision making process in order to describe and to analyze the departure time choices of transit passengers. To analyze and evaluate the model an example transit line heading for work place was used. Numerical results indicated that in the model based on heterogeneous passengers the travelers' preference influenced more seriously on the departure time choice behavior, while in the model based on homogeneous passengers it does not. The results based on homogeneous passengers seemed to be unrealistic in the view of rational behavior. These results imply that the aggregated travel demand models such as the traditional network assignment models based on user equilibrium, assuming perfect information on the network, homogeneity and rationality, might be different from the real dynamic travel demand patterns occurred on actual network.

• Journal of Korea Water Resources Association
• /
• v.53 no.5
• /
• pp.383-393
• /
• 2020

The main objective of this study is to provide a robust model for estimating parameters of the Clark unit hydrograph (UH) using the observed rainfall-runoff data in the Soyangang dam basin. In general, HEC-1 and HEC-HMS models, developed by the Hydrologic Engineering Center, have been widely used to optimize the parameters in Korea. However, these models are heavily reliant on the objective function and sample size during the optimization process. Moreover, the optimization process is carried out on the basis of single rainfall-runoff data, and the process is repeated for other events. Their averaged values over different parameter sets are usually used for practical purposes, leading to difficulties in the accurate simulation of discharge. In this sense, this paper proposed a hierarchical Bayesian model for estimating parameters of the Clark UH model. The proposed model clearly showed better performance in terms of Bayesian inference criterion (BIC). Furthermore, the result of this study reveals that the proposed model can also be applied to different hydrologic fields such as dam design and design flood estimation, including parameter estimation for the probable maximum flood (PMF).

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.5
• /
• pp.169-176
• /
• 2020

Recently, deep neural networks (DNNs) have brought revolutions to many mobile and embedded devices by providing human-level machine intelligence for various applications. However, high inference accuracy of such DNNs comes at high computational costs, and, hence, there have been significant efforts to reduce computational overheads of DNNs either by compressing off-the-shelf models or by designing a new small footprint DNN architecture tailored to resource constrained devices. One notable recent paradigm in designing small footprint DNN models is sharing parameters in several layers. However, in previous approaches, the parameter-sharing techniques have been applied to large deep networks, such as ResNet, that are known to have high redundancy. In this paper, we propose a parameter-sharing method for already parameter-efficient small networks such as ShuffleNetV2. In our approach, small templates are combined with small layer-specific parameters to generate weights. Our experiment results on ImageNet and CIFAR100 datasets show that our approach can reduce the size of parameters by 15%-35% of ShuffleNetV2 while achieving smaller drops in accuracies compared to previous parameter-sharing and pruning approaches. We further show that the proposed approach is efficient in terms of latency and energy consumption on modern embedded devices.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.5
• /
• pp.495-503
• /
• 2014

In recent days, a study of the intrusion detection system collecting and analyzing network data, packet or logs, has been actively performed to response the network threats in computer security fields. In particular, Bayesian network has advantage of the inference functionality which can infer with only some of provided data, so studies of the intrusion system based on Bayesian network have been conducted in the prior. However, there were some limitations to calculate high detection performance because it didn't consider the problems as like complexity of the relation among network packets or continuos input data processing. Therefore, in this paper we proposed two methodologies based on K-menas clustering to improve detection rate by reforming the problems of prior models. At first, it can be improved by sophisticatedly setting interval range of nodes based on K-means clustering. And for the second, it can be improved by calculating robust CPT through applying weighted-leaning based on K-means clustering, too. We conducted the experiments to prove performance of our proposed methodologies by comparing K_WTAN_EM applied to proposed two methodologies with prior models. As the results of experiment, the detection rate of proposed model is higher about 7.78% than existing NBN(Naive Bayesian Network) IDS model, and is higher about 5.24% than TAN(Tree Augmented Bayesian Network) IDS mode and then we could prove excellence our proposing ideas.