• 산업경영시스템학회지
• /
• 제45권2호
• /
• pp.65-76
• /
• 2022

Recently, not only traditional statistical techniques but also machine learning algorithms have been used to make more accurate bankruptcy predictions. But the insolvency rate of companies dealing with financial institutions is very low, resulting in a data imbalance problem. In particular, since data imbalance negatively affects the performance of artificial intelligence models, it is necessary to first perform the data imbalance process. In additional, as artificial intelligence algorithms are advanced for precise decision-making, regulatory pressure related to securing transparency of Artificial Intelligence models is gradually increasing, such as mandating the installation of explanation functions for Artificial Intelligence models. Therefore, this study aims to present guidelines for eXplainable Artificial Intelligence-based corporate bankruptcy prediction methodology applying SMOTE techniques and LIME algorithms to solve a data imbalance problem and model transparency problem in predicting corporate bankruptcy. The implications of this study are as follows. First, it was confirmed that SMOTE can effectively solve the data imbalance issue, a problem that can be easily overlooked in predicting corporate bankruptcy. Second, through the LIME algorithm, the basis for predicting bankruptcy of the machine learning model was visualized, and derive improvement priorities of financial variables that increase the possibility of bankruptcy of companies. Third, the scope of application of the algorithm in future research was expanded by confirming the possibility of using SMOTE and LIME through case application.

• 한국정보교육학회:학술대회논문집
• /
• 한국정보교육학회 2021년도 학술논문집
• /
• pp.369-376
• /
• 2021

인공지능(AI) 기술의 발전에 따라 많은 분야에서 인공지능 활용 방안에 대한 논의가 활발하게 일어나고 있으며 교육 분야에서도 인공지능 인재 양성을 위한 각종 정책이 추진되고 있다. 본 연구에서는 인공지능 기술을 활용한 로봇 프로그래밍 프레임워크를 제안하고 이를 기반으로 머신러닝(Machine Learning) 분야에서 높은 빈도로 활용되는 파이썬(Python)과 교육 현장의 활용도가 높은 교육용 로봇을 활용하여 인공지능(AI) 교육 프로그램을 제안하였다. 국제자동차공학회(SAE)에서 제시하는 자율주행자동차 수준(0~5단계)을 4단계로 단순화하고 이를 기반으로 로봇에 부착된 카메라가 선(객체)을 인지(Perception)하고 검출(Object detection)하여 스스로 움직일 수 있는 라인 디텍터(Line Detector)를 만드는 것을 목표로 하였다. 개발된 프로그램은 단순히 특정 프로그래밍 언어를 활용하여 주어진 문제를 해결하는 정형화된 형태가 아니라 생활 속의 복잡하고 비구조화된 문제를 자기주도적으로 정의하고 인공지능(AI) 기술을 기반으로 해결하는 경험을 가지는데 그 의의가 있다.

• 스마트미디어저널
• /
• 제11권2호
• /
• pp.39-52
• /
• 2022

In the context of the evolution of automation and intelligence, deep learning and machine learning algorithms have been widely applied in aquaculture in recent years, providing new opportunities for the digital realization of aquaculture. Especially, water quality management deserves attention thanks to its importance to food organisms. In this study, we proposed an end-to-end deep learning-based TabNet model for water quality prediction. From major indexes of water quality assessment, we applied novel deep learning techniques and machine learning algorithms in innovative fish aquaculture to predict the number of water cells counting. Furthermore, the application of deep learning in aquaculture is outlined, and the obtained results are analyzed. The experiment on in-house data showed an optimistic impact on the application of artificial intelligence in aquaculture, helping to reduce costs and time and increase efficiency in the farming process.

• Nuclear Engineering and Technology
• /
• 제55권11호
• /
• pp.4282-4286
• /
• 2023

Nuclear energy is estimated by the machine learning method as the mathematical quantifications where neural networking is the major algorithm of the data propagations from input to output. As the aspect of nuclear energy, the other energy sources of the traditional carbon emission-characterized oil and coal are compared. The artificial intelligence (AI) oriented algorithm like the intelligence of a robot is applied to the modeling in which the mimicking of biological neurons is utilized in the mathematical calculations. There are graphs for nuclear priority weighted by climate factor and for carbon dioxide mitigation weighted by climate factor in which the carbon dioxide quantities are divided by the weighting that produces some results. Nuclear Priority and CO2 Mitigation values give the dimensionless values that are the comparative quantities with the normalization in 2010. The values are 1.0 in 2010 of the graphs which are changed to 24.318 and 0.0657 in 2040, respectively. So, the carbon dioxide emissions could be reduced in this study.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2005년도 춘계학술대회 논문집
• /
• pp.1228-1231
• /
• 2005

This paper reports about the development of scanning probe microscopy (SPM) tip with multi-walled carbon nanotube (MWNT). For making a carbon nanotube (CNT) modified tips, AC electric field which causes the dielectrophoresis was used for alignment and deposition of CNTs to the metal coated SPM tip. By dropping the MWNT solution and applying an electric field between an SPM tip and an electrode, MWNTs which were dispersed into a diluted solution were directly assembled onto the apex of the SPM tips due to the attraction by the dielectrophoretic force. In this paper, we investigate experimental conditions about the alignment of the CNT to tip axis according to the change of the angle between a tip and an electrode. Experimental results are presented, and then fabricated CNT tips are showed and measurement results for 15nm gold particles are compared with that of the conventional silicon tip.

• 한국인공지능학회지
• /
• 제6권2호
• /
• pp.23-27
• /
• 2018

Korea has a high proportion of self-employment. Many of them start the food business since it does not require high-techs and it is possible to start the business relatively easily compared to many others in business categories. However, the closure rate of the business is also high due to excessive competition and market saturation. Cafés and restaurants are examples of food business where the business analysis is highly important. However, for most of the people who want to start their own business, it is difficult to conduct systematic business analysis such as trade area analysis or to find information for business analysis. Therefore, in this paper, we predicted business status with simple information using Microsoft Azure Machine Learning Studio program. Experimental results showed higher performance than the number of attributes, and it is expected that this artificial intelligence model will be helpful to those who are self-employed because it can easily predict the business status. The results showed that the overall accuracy was over 60 % and the performance was high compared to the number of attributes. If this model is used, those who prepare for self-employment who are not experts in the business analysis will be able to predict the business status of stores in Seoul with simple attributes.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.621-625
• /
• 2005

This paper presents a new hybrid serial-parallel robot(HSPR), which has six degrees of freedom driven by ball screw linear actuators and motored joints. This hybrid robot design presents a compromise between high rigidity of fully parallel manipulators and extended workspace of serial manipulators. The hybrid robot has a large, singularity-free workspace and high stiffness. Therefore, the presented kinematic structure of the hybrid robot is particularly suitable for multi-tasking machining processes such as milling, drilling, deburring and grinding. In addition to the machining processes, the hybrid robot can be used for welding, fixturing, material handling and so on. The study on design of the hybrid robot is performed. A kinematic analysis and mechanism description of the hybrid robot with six-controlled degree of freedom is presented. In the virtual design works by DADS, workspace and force analysis are discussed. A numerical model is treated to demonstrate our analysis and to determine the range of permissible extension of the struts. Also, we determine some important design parameters for the hybrid robot.

• International Journal of Precision Engineering and Manufacturing
• /
• 제6권2호
• /
• pp.19-25
• /
• 2005

In this paper, the feedforward control with least mean square (LMS) adaptive algorithm is proposed and examined to reduce rotating error by runout of an active magnetic bearing system. Using eddy-current type gap sensors for control, the electrical runout caused by non-uniform material properties of sensor target produces rotational error amplified in feedback control loop, so this runout should be eliminated to increase rotating accuracy. The adaptive feedforward controller is designed and examined its tracking performances and stability numerically with established frequency response function. The designed feedforward controller was applied to a grinding spindle system which is manufactured with a 5.5 kW internal motor and 5-axis active magnetic bearing system including 5 eddy current gap sensors which have approximately 15∼30㎛ of electrical runout. According to the experimental results, the error signal in radial bearings is reduced to less than 5 ,Urn when it is rotating up to 50,000 rpm due to applying the feedforward control for first order harmonic frequency, and corresponding vibration of the spindle is also removed.

• 한국생산제조학회지
• /
• 제23권6호
• /
• pp.547-554
• /
• 2014

CVD SiC is a perfect material used for molds/dies in hot press molding of glass lens. In its fabrication process, nano-precision polishing is essential finally. For this purpose, a novel polishing method using MCF (Magnetic Compound Fluid) slurry is proposed. In this method, MCF slurry is supplied into a given gap between the workpiece and a MCF slurry carrier, and constrained within the polishing zone by magnetic forces from permanent magnet. In this paper, after an experimental rig used to actually realize the proposed method has been constructed, the fundamental polishing characteristics of CVD SiC such as the effects of process parameters including MCF slurry composition on work-surface roughness were experimentally investigated. As a result, nano-precision surface finish of CVD SiC was successfully attained with MCF slurry and the optimum process parameters for obtaining the smoothest work-surface were determined.

• Journal of Magnetics
• /
• 제16권4호
• /
• pp.431-434
• /
• 2011

The reduction of the Nb content in the $(Fe_{0.75}B_{0.20}Si_{0.05})_{96}Nb_4$ bulk metallic glass (BMG) has been studied. The glass-forming ability (GFA) is reduced by decreasing the Nb content, but it can be enhanced by replacing partially Fe by Co. Furthermore, the saturation magnetization of the $(Fe_{0.8}Co_{0.2})_{76}B_{18}Si_3Nb_3$ BMG is 1.35 T, being with 13% larger than that of the base alloy $(Fe_{0.75}B_{0.20}Si_{0.05})_{96}Nb_4$. $(Fe_{0.8}Co_{0.2})_{76}B_{18}Si_3Nb_3$ BMG exhibits slightly larger $B_{800}$ (the magnetic flux density at 800 A/m) and smaller core losses (20%-30%) compared with the commercial Fe-6.5 mass% Si steel.