• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.7
• /
• pp.191-198
• /
• 2021

Since the size of training dataset become large and the model is getting deeper to achieve high accuracy in deep learning, the deep neural network training requires a lot of computation and it takes too much time with a single node. Therefore, distributed deep learning is proposed to reduce the training time by distributing computation across multiple nodes. In this study, we propose hybrid allreduce strategy that considers the characteristics of each layer and communication and computational overlapping technique for synchronization of distributed deep learning. Since the convolution layer has fewer parameters than the fully-connected layer as well as it is located at the upper, only short overlapping time is allowed. Thus, butterfly allreduce is used to synchronize the convolution layer. On the other hand, fully-connecter layer is synchronized using ring all-reduce. The empirical experiment results on PyTorch with our proposed scheme shows that the proposed method reduced the training time by up to 33% compared to the baseline PyTorch.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.18-23
• /
• 2018

FPSO consists of topside modularized plants for production of crude oil, and hullside structures that serve as support for the topside and storage of produced crude oil. The structural behavior of the FPSO topside module and its supporting hull depends on the interface structure that connects them, and the interface structure consists of a combination of individual unit support structures called Module Support Seat (MSS). Types of interface structures are various and, accordingly, the basic design of the FPSO topside module structure is greatly influenced, so various design methods should be considered from the initial design phase. Structural design of FPSO topside module requires consideration of the number of MSSs, connection type, and structural analysis options such as the range of finite element models, load conditions, and boundary conditions for verification of structural strength. In this study, the comparison combination cases for the above considerations were derived and the strength evaluation was performed, and the structural behavior characteristics of the topside module were compared and analyzed through a detailed review of the analysis results. The results of this study are considered to be a good reference for designing a more reliable topside module structure.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.23 no.1
• /
• pp.13-24
• /
• 2021

Tunnel boring machine (TBM) is widely used for tunnel excavation in hard rock and soft ground. In the perspective of TBM-based tunneling, one of the main challenges is to drive the machine optimally according to varying geological conditions, which could significantly lead to saving highly expensive costs by reducing the total operation time. Generally, drilling investigations are conducted to survey the geological ground before the TBM tunneling. However, it is difficult to provide the precise ground information over the whole tunnel path to operators because it acquires insufficient samples around the path sparsely and irregularly. To overcome this issue, in this study, we proposed a geological type classification system using the TBM operating data recorded in a 5 s sampling rate. We first categorized the various geological conditions (here, we limit to granite) as three geological types (i.e., rock, soil, and mixed type). Then, we applied the preprocessing methods including outlier rejection, normalization, and extracting input features, etc. We adopted a deep neural network (DNN), which has 6 hidden layers, to classify the geological types based on TBM operating data. We evaluated the classification system using the 10-fold cross-validation. Average classification accuracy presents the 75.4% (here, the total number of data were 388,639 samples). Our experimental results still need to improve accuracy but show that geology information classification technique based on TBM operating data could be utilized in the real environment to complement the sparse ground information.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.6
• /
• pp.477-483
• /
• 2021

In a military maintenance system, the standard maintenance man-hour of weapon systems is a tool to estimate the maintenance capabilities of maintenance units, provide standards for determining the maintenance needs and workload, and provide basic data for establishing a maintenance plan. The standard maintenance man-hours of major weapon systems have already been derived and used, but the standard maintenance man-hour in a wartime maintenance environment has not been computed. Therefore, the standard wartime maintenance man-hours need to be derived and This study proposes a process and method of computing the maintenance man-hours. In addition, this work suggests the criteria of collecting and screening data that is necessary for estimating the standard maintenance man-hours and introduces a methodology for analyzing the characteristics of maintenance man-hour distribution in the process. The proposed process first designs a model that reflects the wartime maintenance environment, selects statistical techniques, collects maintenance data, analyzes the descriptive statistics, estimates the distribution, and finally presents representative values of maintenance man-hour. Based on the proposed method, the standard wartime maintenance man-hours of the four weapon systems were calculated, and the distribution of the maintenance man-hours was analyzed to follow a lognormal distribution, and the method presented reliable results.

• Science of Emotion and Sensibility
• /
• v.25 no.3
• /
• pp.15-32
• /
• 2022

This study seeks to clarify the mechanisms of anthropomorphism and positive user experience. This study adopts the "computers are social actors" (CASA) paradigm to verify the causal relationship between social response and anthropomorphism and correctly explicate this paradigm. The intimacy-forming and anthropomorphizing effects of deep self-disclosure in interpersonal relationships were replicated in relationships between humans and conversational agents to induce both social response and anthropomorphism. Then, the mediating effect of intimacy on the anthropomorphizing effect of deep self-disclosure was explored with psychological models that revealed the causal relationships between social connections, including intimacy and anthropomorphism. Furthermore, we explored how intimacy and anthropomorphism trigger positive user experiences. The results demonstrated that the deeper the self-disclosure depth was, the more intimate and humanly the agent was perceived and the more positive the user experience was. In addition, the effect of self-disclosure depth on anthropomorphism and positive user experience was completely mediated by intimacy. This means that when using a computer with interpersonal characteristics, people anthropomorphize it and have a positive experience because people react socially to objects with social cues. This study bridges the gap between the CASA paradigm and anthropomorphism research, suggesting the possibility of psychological explanations for the principle of human-computer interactions. In addition, it explicates the mechanism of anthropomorphism and positive user experience, emphasizing the importance of social response-that is, intimacy.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.51 no.4
• /
• pp.76-89
• /
• 2023

One of the technologies in urban agriculture, hydroponics cultivation, has primarily focused on technological development, resulting in a lack of research on urban agriculture's cultural utilization aspects, encompassing cultural values associated with urban residents' leisure activities. Therefore, this study aimed to identify the participation constraints perceived by school community members when implementing urban farming activities using hydroponics and understand the structural relationships between the variables that influence decision-making from the perspective of leisure activities in urban farming. As a result, participation constraints in urban farming activities utilizing hydroponics were first categorized into intrinsic, interpersonal, and structural factors. Second, the results of hypothesis model verification showed that interpersonal constraints significantly influenced the participants' willingness to participate and their expected behavior. This study found the multidimensional perceptions of school community members regarding hydroponic urban farming conducted in urban spaces, particularly rooftops, and revealed the influence of decision-making factors on participation when conducting urban farming activities using hydroponic cultivation.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.219-240
• /
• 2023

Conversational AI which allows users to interact with satisfaction is a long-standing research topic. To develop conversational AI, it is necessary to build training data that reflects real conversations between people, but current Korean datasets are not in question-answer format or use honorifics, making it difficult for users to feel closeness. In this paper, we propose a conversation dataset (KOMUChat) consisting of 30,767 question-answer sentence pairs collected from online communities. The question-answer pairs were collected from post titles and first comments of love and relationship counsel boards used by men and women. In addition, we removed abuse records through automatic and manual cleansing to build high quality dataset. To verify the validity of KOMUChat, we compared and analyzed the result of generative language model learning KOMUChat and benchmark dataset. The results showed that our dataset outperformed the benchmark dataset in terms of answer appropriateness, user satisfaction, and fulfillment of conversational AI goals. The dataset is the largest open-source single turn text data presented so far and it has the significance of building a more friendly Korean dataset by reflecting the text styles of the online community.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.6D
• /
• pp.809-817
• /
• 2008

This paper deals with the development of ViscoElastic Continuum Damage Finite Element Program (VECD-FEP++) and its verification with the results from both field and laboratory accelerated pavement tests. Damage characteristics of asphalt concrete mixture have been defined by Schapery's work potential theory, and uniaxial constant crosshead rate tests were carried out to be used for damage model implementation. VECD-FEP++ predictions were compared with strain responses (longitudinal and transverse strains) under moving wheel loads running at different constant speeds. To this end, an asphalt pavement section (A5) of Korea Expressway Corporation Test Road (KECTR) instrumented with strain gauges were loaded with a dump truck. Also, a series of accelerated pavement fatigue tests have been conducted at pavement sections surfaced with four asphalt concrete mixtures (Dense-graded, SBS, Terpolymer, CR-TB). Planar strain responses were in good agreement with field measurements at base layers, whereas strains at both surface and intermediate layers were found different from simulation results due to the complexity of tire-road contact pressures. Finally, fatigue characteristics of four asphalt mixtures were reasonably described with VECD-FEP++.

• Korean Journal of Remote Sensing
• /
• v.36 no.2_2
• /
• pp.249-261
• /
• 2020

Coastal monitoring using multiple platforms/sensors is a very important tools for accurately understanding the changes in offshore marine environment and disaster with high temporal and spatial resolutions. However, integrated observation studies using multiple platforms and sensors are insufficient, and none of them have been evaluated for efficiency and limitation of convergence. In this study, we aimed to suggest an integrated observation method with multi-remote sensing platform and sensors, and to diagnose the utility and limitation. Integrated in situ surveys were conducted using Rhodamine WT fluorescent dye to simulate various marine disasters. In September 2019, the distribution and movement of RWT dye patches were detected using satellite (Kompsat-2/3/3A, Landsat-8 OLI, Sentinel-3 OLCI and GOCI), unmanned aircraft (Mavic 2 pro and Inspire 2), and manned aircraft platforms after injecting fluorescent dye into the waters of the South Sea-Yeosu Sea. The initial patch size of the RWT dye was 2,600 ㎡ and spread to 62,000 ㎡ about 138 minutes later. The RWT patches gradually moved southwestward from the point where they were first released,similar to the pattern of tidal current flowing southwest as the tides gradually decreased. Unmanned Aerial Vehicles (UAVs) image showed highest resolution in terms of spatial and time resolution, but the coverage area was the narrowest. In the case of satellite images, the coverage area was wide, but there were some limitations compared to other platforms in terms of operability due to the long cycle of revisiting. For Sentinel-3 OLCI and GOCI, the spectral resolution and signal-to-noise ratio (SNR) were the highest, but small fluorescent dye detection was limited in terms of spatial resolution. In the case of hyperspectral sensor mounted on manned aircraft, the spectral resolution was the highest, but this was also somewhat limited in terms of operability. From this simulation approach, multi-platform integrated observation was able to confirm that time,space and spectral resolution could be significantly improved. In the future, if this study results are linked to coastal numerical models, it will be possible to predict the transport and diffusion of contaminants, and it is expected that it can contribute to improving model accuracy by using them as input and verification data of the numerical models.

• Korean Journal of Environment and Ecology
• /
• v.29 no.3
• /
• pp.333-343
• /
• 2015

This research aims at identifying the goshawk's possible and replaceable breeding ground by using the MaxEnt prediction model which has so far been insufficiently used in Korea, and providing evidence to expand possible protection areas for the goshawk's breeding for the future. The field research identified 10 goshawk's nests, and 23 appearance points confirmed during the 3rd round of environmental research were used for analysis. 4 geomorphic, 3 environmental, 7 distance, and 9 weather factors were used as model variables. The final environmental variables were selected through non-parametric verification between appearance and non-appearance coordinates identified by random sampling. The final predictive model (MaxEnt) was structured using 10 factors related to breeding ground and 7 factors related to appearance area selected by statistics verification. According to the results of the study, the factor that affected breeding point structure model the most was temperature seasonality, followed by distance from mixforest, density-class on the forest map and relief energy. The factor that affected appearance point structure model the most was temperature seasonality, followed by distance from rivers and ponds, distance from agricultural land and gradient. The nature of the goshawk's breeding environment and habit to breed inside forests were reflected in this modeling that targets breeding points. The northern central area which is about $189.5 km^2$(2.55 %) is expected to be suitable breeding ground. Large cities such as Cheongju and Chungju are located in the southern part of Chungcheongbuk-do whereas the northern part of Chungcheongbuk-do has evenly distributed forests and farmlands, which helps goshawks have a scope of influence and food source to breed. Appearance point modeling predicted an area of $3,071 km^2$(41.38 %) showing a wider ranging habitat than that of the breeding point modeling due to some limitations such as limited moving observation and non-consideration of seasonal changes. When targeting the breeding points, a specific predictive area can be deduced but it is difficult to check the points of nests and it is impossible to reflect the goshawk's behavioral area. On the other hand, when targeting appearance points, a wider ranging area can be covered but it is less accurate compared to predictive breeding point since simple movements and constant use status are not reflected. However, with these results, the goshawk's habitat can be predicted with reasonable accuracy. In particular, it is necessary to apply precise predictive breeding area data based on habitat modeling results when enforcing an environmental evaluation or establishing a development plan.