• Transactions of the Korean Society of Automotive Engineers
• /
• v.10 no.5
• /
• pp.29-34
• /
• 2002

This study presents the process of the continuous-time system identification for unknown nonlinear systems. The Radial Basis Function(RBF) error filtering identification model is introduced at first. This identification scheme includes RBF network to approximate unknown function of nonlinear system which is structured by affine form. The neural network is trained by the adaptive law based on Lyapunov synthesis method. The identification scheme is applied to engine and the performance of RBF error filtering Identification model is verified by the simulation with a three-state engine model. The simulation results have revealed that the values of the estimated function show favorable agreement with the real values of the engine model. The introduced identification scheme can be effectively applied to model-based nonlinear control.

• Industrial Engineering and Management Systems
• /
• v.15 no.3
• /
• pp.215-223
• /
• 2016

Design of personnel teams has been studied from diverse perspectives; the most common are the people and systems requirements perspectives. All these point of view are linked, which is the reason why it is necessary to study them simultaneously. Considering this gap, a decision making model is developed based on factors, models, and requirements mentioned in the literature. The model is applied to a real case. The findings indicate that the Personnel Behavior Based Lean model (PBBL) can be converted into a decision making model for the selection of team members. The study is focused not only on the individual candidates' knowledge, skills, and aptitudes, but also on how the model considers the company requirements, conflicts, and the importance of each person to the project.

• Journal of information and communication convergence engineering
• /
• v.19 no.3
• /
• pp.136-141
• /
• 2021

Research on wireless sensor networks has focused on the monitoring and characterization of large-scale physical environments and the tracking of various environmental or physical conditions, such as temperature, pressure, and wind speed. We propose a stochastic mobility model that can be applied to a MANET (Mobile Ad-hoc NETwork). environment, and apply this mobility model to a newly proposed clustering-based routing protocol. To verify its stability and durability, we compared the proposed stochastic mobility model with a random model in terms of energy efficiency. The FND (First Node Dead) was measured and compared to verify the performance of the newly designed protocol. In this paper, we describe the proposed mobility model, quantify the changes to the mobile environment, and detail the selection of cluster heads and clusters formed using a fuzzy inference system. After the clusters are configured, the collected data are sent to a base station. Studies on clustering-based routing protocols and stochastic mobility models for MANET applications have shown that these strategies improve the energy efficiency of a network.

• Journal of Information Technology Applications and Management
• /
• v.27 no.6
• /
• pp.141-151
• /
• 2020

The job mismatch between job seekers and SMEs is becoming more and more intensifying with the serious difficulties in youth employment. In this study, a bi-directional content-based metric learning model is proposed to recommend suitable jobs for job seekers and suitable job seekers for SMEs, respectively. The proposed model not only enables bi-directional recommendation, but also enables HR matching without relearning for new job seekers and new job offers. As a result of the experiment, the proposed model showed superior performance in terms of precision, recall, and f1 than the existing collaborative filtering model named NCF+GMF. The proposed model is also confirmed that it is an evolutionary model that improves performance as training data increases.

• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.579-586
• /
• 2009

4D simulation integrates the 3 dimensional model of a building with the construction schedule, and it leads to the possibility of virtually checking the construction process and the building itself in advance. However, the existing problem of 4D simulation is the difference between the demands of architects, engineers, and construction site workers in 4D simulation. This study suggests the possible way to enhance the availability of 4D simulation, considering more the practical demands from the construction site. In order to conduct this study, we build a 3D BIM model of a business-residential complex and link the model with the pre-defined construction schedule in order to make a 4D simulation. This study concludes with the optimized 4D simulation methodology based on BIM model considering the demands in perspective of the construction site. It would contribute to the harmonic collaboration among architects, engineers, and labors in the construction site when using 4D simulation based on BIM model.

• Annual Conference of KIPS
• /
• 2023.05a
• /
• pp.624-626
• /
• 2023

Credit scoring is a technique used by financial institutions to assess the creditworthiness of potential borrowers. This involves evaluating a borrower's credit history to predict the likelihood of defaulting on a loan. This paper presents an ensemble of two Transformer based models within a framework for discriminating the default risk of loan applications in the field of credit scoring. The first model is FinBERT, a pretrained NLP model to analyze sentiment of financial text. The second model is FT-Transformer, a simple adaptation of the Transformer architecture for the tabular domain. Both models are trained on the same underlying data set, with the only difference being the representation of the data. This multi-modal approach allows us to leverage the unique capabilities of each model and potentially uncover insights that may not be apparent when using a single model alone. We compare our model with two famous ensemble-based models, Random Forest and Extreme Gradient Boosting.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.5
• /
• pp.1089-1099
• /
• 2013

The Bradley-Terry model is widely used for analysis of pairwise preference data. We explain that the popularity of Bradley-Terry model is gained due to not only easy computation but also some nice asymptotic properties when the model is misspecified. For information retrieval required to analyze big ranking data, we propose to use a pseudo likelihood based on the Bradley-Terry model even when the true model is different from the Bradley-Terry model. We justify using the Bradley-Terry model by proving that the estimated ranking based on the proposed pseudo likelihood is consistent when the true model belongs to the class of Thurstone models, which is much bigger than the Bradley-Terry model.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.34 no.4
• /
• pp.15-26
• /
• 2009

Bass diffusion model have played a central role in studying the diffusion of the new products since 1969, the year of publication of Bass model. Almost 750 publications based on the Bass diffusion model have explored extensions and applications. Extension models can be divided into two types. One is the model containing marketing-mix variables and the other is the model containing additional parameters. This paper presents another extension model of the latter type. Our model allows the time varying coefficients of innovation and imitation. Two pieces approximation of time varying coefficients is introduced and it's parameters are estimated based on NLS(Non-Linear Mean Square) method. Empirical studies are performed and the results show that our model is superior to the basic Bass model and the NUI(Non-Uniform Influence) model which is the well-known extension of the Bass model. The model developed in this paper is, also, transformed into the Bass model with the ready potential adopters in order to enhance the descriptive power.

• Wind and Structures
• /
• v.39 no.4
• /
• pp.305-311
• /
• 2024

This paper reviews measurements of wind profiles in the atmospheric boundary layer in strong wind (thermally neutral) conditions over open water and the ocean, and the variation of the roughness parameters with mean wind speed. Based on the wind data recorded on the coast of the island of Frøya (Norway) in the 1980s, and dropwindsonde profiles in hurricanes, the paper shows that 'capping', or saturation, of the surface drag coefficient becomes apparent at a mean wind speed at 10m height of about 25 m/s. Wind speed models used in the offshore industries were investigated, (the ISO model, the API 'tropical cyclone' model and the IEC model). The ISO model, although based on good quality data from Frøya, does not allow for the saturation of the roughness above about 25-30 m/s, even though that was apparent in the Frøya data. 'Capping' of the aerodynamic roughness length for wind speeds greater than 28 m/s is represented appropriately in the API 'tropical cyclone'model, and hence the model represents the mean wind properties reasonably well in severe tropical cyclone conditions. However, the turbulence intensities in the API 'tropical cyclone' model, based on over-land measurements (ESDU), are overpredicted for winds over the ocean, at heights above 20m. The IEC models are entirely based on over-land measurements, and hence are not representative of over-water conditions such as those required for offshore wind farms. New model profiles for over-ocean strong winds are proposed for wind speeds up to hurricane strength, based on the ISO profiles, but with capping of the surface drag coefficient at a value of 0.0025, at a mean wind speed at 10m height of 25 m/s. The proposed turbulence intensity model is also a revision of the ISO profile, also with capping above 25 m/s. The proposed model profiles are in better general agreement with recorded data in strong winds than those currently specified in international standards, and are applicable to all wind speeds in synoptic-scale events, including those in tropical cyclones, typhoons and hurricanes. As well as the Frøya data, the revised strong-wind models are supported by measurements from Atlantic hurricanes, gales in the North Sea, landfalling typhoons in Japan and Cyclone 'Yasi' in Queensland, Australia.

• Journal of Ship and Ocean Technology
• /
• v.10 no.2
• /
• pp.11-23
• /
• 2006

The model exchange from CAD system to CAE system in valid and effective manner is the major issue of automatic analysis modelling of ship structure. However, model exchange approaches based on the neutral CAD file have resulted in invalid model exchange that could not properly reflect the characteristics of CAD model and CAE model of ship structure. This paper presents the new approach of n-to-n mapping to exchange ship structure model in heterogeneous CAD/CAE environments. In this study, the common model called 'unified ship model for analysis' to directly extract proper information from different CAD systems for ship structural analysis is proposed. Moreover, a command language based model interfacing technique to construct an idealized model for analysis job is also proposed. The proposed approach has been actually implemented in DSME CAD/CAE environment of ship structure such as TRIBON system, PATRAN system and FLUENT system. The applicability and effectiveness of the proposed approach was verified by applying it to the real analysis project for fore-body of ship and block lifting. This application results show that the proposed approach can be effectively used for heterogeneous CAD/CAE environment.