• International Journal of Computer Science & Network Security
• /
• 제23권5호
• /
• pp.148-162
• /
• 2023

Classification systems can significantly assist the medical sector by allowing for the precise and quick diagnosis of diseases. As a result, both doctors and patients will save time. A possible way for identifying risk variables is to use machine learning algorithms. Non-surgical technologies, such as machine learning, are trustworthy and effective in categorizing healthy and heart-disease patients, and they save time and effort. The goal of this study is to create a medical intelligent decision support system based on machine learning for the diagnosis of heart disease. We have used a mixed feature creation (MFC) technique to generate new features from the UCI Cleveland Cardiology dataset. We select the most suitable features by using Least Absolute Shrinkage and Selection Operator (LASSO), Recursive Feature Elimination with Random Forest feature selection (RFE-RF) and the best features of both LASSO RFE-RF (BLR) techniques. Cross-validated and grid-search methods are used to optimize the parameters of the estimator used in applying these algorithms. and classifier performance assessment metrics including classification accuracy, specificity, sensitivity, precision, and F1-Score, of each classification model, along with execution time and RMSE the results are presented independently for comparison. Our proposed work finds the best potential outcome across all available prediction models and improves the system's performance, allowing physicians to diagnose heart patients more accurately.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2003년도 ISIS 2003
• /
• pp.644-647
• /
• 2003

This paper describes a neural network based machine vision system designed for inspecting yellow beans in real time. The system consists of a camera. lights, a belt conveyor, air ejectors, and a computer. Beans are conveyed in four lines on a belt and their images are taken by a monochrome line scan camera when they fall down from the belt. Beans are separated easily from their background on images by back-lighting. After analyzing the image, a decision is made by a multilayer artificial neural network (ANN) trained by the error back-propagation (EBP) algorithm. We use the global mean, variance and local change of gray levels of a bean for the input nodes of the network. In an our experiment, the system designed could process about 520kg/hour.

• 대한전기학회논문지
• /
• 제43권3호
• /
• pp.492-503
• /
• 1994

In this study, the control and scheduling of the flexible manufacturing cell (FMC) is discussed, which can perform the mixed production and relieve the effect of machine failure. The control of the FMC isvery complex task due to the property of multiple jobs and the dynamically changing states. For effective control of proposed FMC, the hierarchical scheme is introduced and the functions of each levels are defined. Especially for the control functions of shop floor level and cell level, the intelligent scheduler is implemented. To show the efficiency of the intelligent scheduler, the production method fo the existing assembly lines was evaluated and compared with the proposed intelligent FMC method. The results from the production performance show that the proposed method is superior to the existing method in various performance indices.

• 한국공작기계학회논문집
• /
• 제14권4호
• /
• pp.55-60
• /
• 2005

In this study, vibration control of the enclosed plate which is typical two dimensional structure are performed by using intelligent control method and piezo ceramic as an actuator attached at the plate. The characteristics of the plate with piezo ceramic are examined to identify the natural frequencies. Then, the plate is controlled under the sinusoidal excitation by acoustic source and the reduction level of vibration is examined via frequency response and time sequence. The control result shows the effect of the intelligent control method as a control scheme compared with that of the previous filtered-X LMS algorithm.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.956-959
• /
• 2004

Recently the monitoring system of tool setting in high speed precision machining center is required for manufacturing products that have highly complex and small shape, high precision and high function. It is very important to reduce time to setup tool in order to improve the machining precision and the productivity and to protect the breakage of cutting tool as the shape of product is smaller and more complex. Generally, the combination of errors that geometrical clamping error of fixing tool at the spindle of machining tool and the asynchronized error of driving mechanism causes that the run-out of tool reaches to 3$^{\sim}$20 times of the thickness of cutting chip. And also the run-out is occurred by the misalignment between axis of tool shank and axis of spindle and spindle bearing in high speed rotation. Generally, high speed machining is considered when the rotating speed is more than 8,000 rpm. At that time, the life time of tool is reduced to about 50% and the roughness of machining surface is worse as the run-out is increased to 10 micron. The life time of tool could be increased by making monitoring of tool-setup easy, quick and precise in high speed machining tool. This means the consumption of tool is much more reduced. And also it reduces the manufacturing cost and increases the productivity by reducing the tool-setup time of operator. In this study, in order to establish the concept of tool-setup monitoring the measuring method of the geometrical error of tool system is studied when the spindle is stopped. And also the measuring method of run-out, dynamic error of tool system, is studied when the spindle is rotated in 8,000${\sim}$60,000 rpm. The dynamic phenomena of tool-setup are analyzed by implementing the monitoring system of rotating tool system and the non-contact measuring system of micro displacement in high speed.

• 한국진공학회지
• /
• 제14권1호
• /
• pp.35-39
• /
• 2005

본 실험에서는 단위요소 사이에 채널을 갖는 Elementwise Patterned Stamp (이하 EPS)를 이용하여 싱글스탭 (single step)으로 4인치 웨이퍼를 임프린트 하는 공정을 수행하였다. 단위요소간의 간격이 3m인 EPS를 이용한 임프린트에서 50 - 100nm급의 패턴을 성공적으로 형성하였다. 그러나 임프린트 과정 중 EPS의 채널 부분에서 웨이퍼의 미소변형이 발생하여 단위요소의 미충전과 불균일한 잔여층이 형성되는 문제들이 발생하였다. 본 논문에서는 이러한 웨이퍼의 미소변형이 단위요소 충전과 패턴형성에 미치는 영향을 확인해 보기 위해 웨이퍼의 두께를 100 - 500㎛로 변화시켜가며 임프린트 실험을 수행하였고, 유한요소법(Finite Element Method, FEM)을 이용한 수치모사를 통하여 실험결과를 확인하였다. 또한 웨이퍼의 미소변형이 발생하는 또 다른 요인인 EPS의 채널 폭을 3mm, 2mm, 1mm로 변화시키며 수행한 수치모사를 통하여 안정된 임프린트 조건을 제시하였다.

• International Journal of Control, Automation, and Systems
• /
• 제1권1호
• /
• pp.156-161
• /
• 2003

An overview of our approach to autonomous navigation is presented showing how motion information can be integrated into existing navigation schemes. Particular attention is given to our short range motion estimation scheme which utilises a number of unique assumptions regarding the nature of the visual environment allowing a direct fusion of visual and range information. Graduated non-convexity is used to solve the resulting non-convex minimisation problem. Experimental results show the advantages of our fusion technique.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2000년도 추계학술대회 학술발표 논문집
• /
• pp.409-412
• /
• 2000

본 논문에서는 기존의 결정론적 볼츠만 머신의 학습알고리듬을 수정하여 양자화결합을 갖는 볼츠만 머신에도 적용할 수 있는 알고리듬을 제안하였다. 제안한 알고리듬은 2-입력 XOR문제와 3-입력 패리티문제에 적용하여 성능을 분석하였다. 그 결과 하중이 대폭적으로 양자화된 네트워크도 학습이 가능하다는 것은 은닉 뉴런수를 증가시키면 한정된 하중값의 범위로 유지할 수 있다는 것을 보여주었다. 또한 1회에 갱신하는 하중의 개수 m$_{s}$를 제어함으로써 학습계수를 제어하는 효과가 얻어지는 것을 확인하였다..

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2002년도 춘계학술대회 및 임시총회
• /
• pp.19-22
• /
• 2002

분류자 시스템은 유전자 알고리즘(Genetic Algorithm : GA)을 이용하여 새로운 규칙 집합을 발견하는 시스템이다 또 로봇 축구 시뮬레이션 게임(SimuroSot)은 시간에 따라 상태가 변화하는 동적인 시스템이다 본 논문에서는 GBML(Genetic Based Machine Learning)의 한 갈래이자 미시간 접근 방법을 기반으로 하는 Zeroth Level Classifier System(ZCS)을 SimuroSot에 적용하여 게임 전략을 구성하는 새로운 규칙의 발견과 학습에 의한 축구 로봇의 행동전략 알고리즘을 제안하고, 시뮬레이션을 통하여 본 전략의 유용성을 확인한다

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
• /
• pp.719-724
• /
• 1998

This paper presents a systematic developement of a formal approach to inference in approximate reasoning. We introduce some measures of similarity and discuss their properties. Using the concept of similarity index we formulate two methods for inferring from vague knowledge. In order to illustrate the effectiveness of the proposed technique we use it to develop a vowel recognition system.