• MALSORI
• /
• no.59
• /
• pp.53-68
• /
• 2006

We report the evaluation results of the Korean speech recognition platform called ECHOS. The platform has an object-oriented and reusable architecture so that researchers can easily evaluate their own algorithms. The platform has all intrinsic modules to build a large vocabulary speech recognizer: Noise reduction, end-point detection, feature extraction, hidden Markov model (HMM)-based acoustic modeling, cross-word modeling, n-gram language modeling, n-best search, word graph generation, and Korean-specific language processing. The platform supports both lexical search trees and finite-state networks. It performs word-dependent n-best search with bigram in the forward search stage, and rescores the lattice with trigram in the backward stage. In an 8000-word continuous speech recognition task, the platform with a lexical tree increases 40% of word errors but decreases 50% of recognition time compared to the HTK platform with flat lexicon. ECHOS reduces 40% of recognition errors through incorporation of cross-word modeling. With the number of Gaussian mixtures increasing to 16, it yields word accuracy comparable to the previous lexical tree-based platform, Julius.

• Journal of Korean Society on Water Environment
• /
• v.36 no.6
• /
• pp.636-650
• /
• 2020

A modeling system that can consider the overall water environment and be used to integrate hydrology, water quality, and aquatic ecosystem on a watershed scale is essential to support decision-making in integrated water resources management (IWRM). In adapting imported models for evaluating the unique water environment in Korea, a platform perspective is becoming increasingly important. In this study, a modeling platform is defined as an ecosystem that continuously grows and provides sustainable values through voluntary participation- and interaction-of all stakeholders- not only experts related to model development, but also model users and decision-makers. We assessed the conceptual values provided by the IWRM modeling platform in terms of openness, transparency, scalability, and sustainability. I We also reviewed the technical aspects of functional and spatial integrations in terms of socio-economic factors and user-centered multi-scale climate-forecast information. Based on those conceptual and technical aspects, we evaluated potential modeling platforms such as Source, FREEWAT, Object Modeling System (OMS), OpenMI, Community Surface-Dynamics Modeling System (CSDMS), and HydroShare. Among them, CSDMS most closely approached the values suggested in model development and offered a basic standard for easy integration of existing models using different program languages. HydroShare showed potential for sharing modeling results with the transparency expected by model user-s. Therefore, we believe that can be used as a reference in development of a modeling platform appropriate for managing the unique integrated water environment in Korea.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.7
• /
• pp.654-661
• /
• 2015

Stairs are the most popular obstacles in buildings and factories. To enlarge the application areas of a field robotic platform, stair-climbing is very important mission. One important reason why a stair-climbing is difficult is that stairs are various in sizes. To achieve autonomous climbing of various-sized stairs, dynamic modeling is essential. In this research, an inverse dynamic modeling is performed to enable an autonomous stair climbing. Stair-climbing robotic platform with flip locomotion, named FilpBot, is analyzed. The FlipBot platform has advantages of robust stair-climbing of various sizes with constant speed, but the autonomous operation is not yet capable. Based on external constraints and the postures of the robot, inverse dynamic models are derived. The models are switched by the constraints and postures to analyze the continuous motion during stair-climbing. The constraints are changed according to the stair size, therefore the analysis results are different each other. The results of the inverse dynamic modeling are going to be used in motor design and autonomous control of the robotic platform.

• Journal of KIBIM
• /
• v.4 no.1
• /
• pp.1-7
• /
• 2014

The construction method using PC(Precast Concrete) has been widely used because the method can shorten the construction period and improve construction quality. In this paper, through the analysis of present design process for PC railway platform, design parameters on the geometry and properties were extracted and 3D information models for PC railway platform were constructed by the parametric modeling technique. Furthermore, the interface module was developed to link 3D models to the structural analysis/design sheet and database program using VBA(Visual Basic Application). This information model could be used in various areas including structural analysis and design, 2D drawing, quantity estimation and 4D simulation including clash detection.

• Korean Journal of Computational Design and Engineering
• /
• v.20 no.1
• /
• pp.65-74
• /
• 2015

Structure and/or fluid analysis is gradually increased by an essential design process in the small and medium-sized businesses (SMBs) because of the needs for a rapid design process and the certification about the supplement of the parts by the large business (LB). In this paper, we developed the web-based ezSIM platform installed in the resources integrated system server. The ezSIM platform is based on the heterogeneous linux and windows operating system for the user-friendly connection with the part of the analysis for the SMBs. The procedure of the structure/fluid analysis service module using the public software and the license-free open code in the ezSIM platform was explained. The convenience of the ezSIM platform service was presented by the reaction rate of the graphic motion compared with that of a local PC and the solving and pre-post processing interface compared with that of the KISTI supercomputer. The web-based ezSIM platform service was identified as a useful and essential platform to the SMBs for the usage of the structure and/or fluid analysis procedure.

• Journal of Information Technology Applications and Management
• /
• v.31 no.1
• /
• pp.163-175
• /
• 2024

This study conducts topic modeling analysis on four electric scooter sharing platforms: Alpaca, SingSing, Kickgoing, and Beam. Using user review data, the study aims to identify key topics and issues associated with each platform, as well as uncover common themes across platforms. The analysis reveals that users primarily express concerns and preferences related to application usability, service mobility, and parking/accessibility. Additionally, each platform exhibits unique characteristics and challenges. Alpaca users generally appreciate convenience and enjoyment but express concerns about safety and service areas. SingSing faces issues with application functionality, while Kickgoing users encounter connectivity problems and device usability issues. Beam receives overall positive feedback, but users express dissatisfaction with application usability and parking. Based on these findings, scooter sharing service providers should focus on enhancing application features, stability, and expanding service coverage to meet user expectations and improve customer satisfaction. Furthermore, highlighting platform-specific strengths and providing tailored services can enhance competitiveness and foster continuous service growth and development.

• Journal of Korean Association for Spatial Structures
• /
• v.16 no.4
• /
• pp.29-36
• /
• 2016

The objective of this research is to development of a parametric design system for membrane structures. The parametric design platform for the spatial structures has been designed and implemented. Rhino3D is used as a 3D graphic kernel and Grasshopper is introduced as a parametric modeling engine. Modeling components such as structural members, loading conditions, and support conditions are developed for structural modeling of the spatial structures. The interface module with commercial structural analysis programs is implemented. An iterative generation algorithm for design alternatives is a part of the design platform. This paper also proposes a design approach for the parametric design of Spoke Wheel membrane structures. A parametric modeling component is designed and implemented. SOFiSTik is examined to interact with the design platform as the structural analysis module. The application of the developed interface is to design optimally Spoke Wheel Shaped Ductile Membrane Structure using parametric design. It is possible to obtain objective shape by controlling the parameter using a parametric modeling designed for shape finding of spoke wheel shaped ductile membrane structure. Recently, looking at the present Construction Trends, It has increased the demand of the large spatial structure. But, It requires a lot of time for Modeling design and the Structural analysis. Finally an optimization process for membrane structures is proposed.

• MALSORI
• /
• no.51
• /
• pp.151-165
• /
• 2004

For educational and research purposes, a Korean speech recognition platform is designed. It is based on an object-oriented architecture and can be easily modified so that researchers can readily evaluate the performance of a recognition algorithm of interest. This platform will save development time for many who are interested in speech recognition. The platform includes the following modules: Noise reduction, end-point detection, met-frequency cepstral coefficient (MFCC) and perceptually linear prediction (PLP)-based feature extraction, hidden Markov model (HMM)-based acoustic modeling, n-gram language modeling, n-best search, and Korean language processing. The decoder of the platform can handle both lexical search trees for large vocabulary speech recognition and finite-state networks for small-to-medium vocabulary speech recognition. It performs word-dependent n-best search algorithm with a bigram language model in the first forward search stage and then extracts a word lattice and restores each lattice path with a trigram language model in the second stage.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.6
• /
• pp.968-974
• /
• 2017

In this paper, modeling of brushless DC motor (BLDC) driving system for platform screen door control applied fuel cell power generation system has been proposed. At first the system configuration and operational principle of the developed fuel cell simulator has been investigated and the design of BLDC motor driving system is studied and the overall performance and dynamics of the proposed system could be effectively examined by simulation. PSIM simulation program is implemented to verify the performance and compatibility of the fuel cell power generation system and BLDC motor control system modeling.

• Journal of KIBIM
• /
• v.9 no.4
• /
• pp.62-74
• /
• 2019

Because the existing BIM platform is based on user services, the focus is on the development of SaaS (Software as a Service), which provides business services online. However, since a harbor is a security facility, the harbor BIM platform is preferably provided in a private form, rather than relying on the infrastructure environment provided by external cloud providers. Therefore, this study analyzes and reviews the main functions to be provided as SaaS services of the harbor BIM platform. The goal is to build a cloud-based harbor BIM platform that can provide this service to users. To this end, we built IaaS (Infrastructure as a Service) environment of the harbor BIM platform based on the open source Open Stack and integrate and develop PaaS environment with Open Shift applied with IaaS. We applied the GPU to the harbor BIM platform to verify the performance of the harbor BIM platform, and found that the rendering and loading times are improved. In particular, it is expected to reduce the cost of introduction and provide it as the basic cloud environment of similar BIM platform for infrastructure facilities.