• Journal of the Semiconductor & Display Technology
• /
• 제19권4호
• /
• pp.121-125
• /
• 2020

With miniaturization of semiconductor, the manufacturing process become more complex, and undetected small changes in the state of the equipment have unexpectedly changed the process results. Fault detection classification (FDC) system that conducts more active data analysis is feasible to achieve more precise manufacturing process control with advanced machine learning method. However, applying machine learning, especially in supervised learning criteria, requires an arduous data labeling process for the construction of machine learning data. In this paper, we propose a semi-supervised learning to minimize the data labeling work for the data preprocessing. We employed equipment status variable identification (SVID) data and optical emission spectroscopy data (OES) in silicon etch with SF6/O2/Ar gas mixture, and the result shows as high as 95.2% of labeling accuracy with the suggested semi-supervised learning algorithm.

• Advances in nano research
• /
• 제14권3호
• /
• pp.225-234
• /
• 2023

In this study, we aim to use big data resources and statistical analysis to obtain a reliable instruction to reach high-quality and high yield agricultural yields. In this regard, soil type data, raining and temperature data as well as wheat production in each year are collected for a specific region. Using statistical methodology, the acquired data was cleaned to remove incomplete and defective data. Afterwards, using several classification methods in machine learning we tried to distinguish between different factors and their influence on the final crop yields. Comparing the proposed models' prediction using statistical quantities correlation factor and mean squared error between predicted values of the crop yield and actual values the efficacy of machine learning methods is discussed. The results of the analysis show high accuracy of machine learning methods in the prediction of the crop yields. Moreover, it is indicated that the random forest (RF) classification approach provides best results among other classification methods utilized in this study.

• Proceedings of the KSME Conference
• /
• 대한기계학회 2001년도 춘계학술대회논문집C
• /
• pp.122-127
• /
• 2001

Lasting machines for shoe manufacturing are continuously developed with the aid of automation and Computer Aided Manufacturing (CAM). Although automation and CAM techniques have tremendously reduced the labor in shoe manufacturing field, there still remain some parts manufactured by experts. In order to enhance the capability and efficiency of machines for labor-free shoe manufacturing, CAD data of a shoe last is indispensable. While CAD datarization takes the fundamental role in the shoe design as well as the shoe manufacturing, there has been little research for the CAD datarization of a shoe last. In this paper, a new procedure for CAD datarization of a shoe last using finite element patches is proposed and some data for the control part of the shoe lasting machine are extracted and interpolated from the CAD data. The outer line of a shoe-last sole is interpolated by a tension spline method and bonding lines are extracted from the shoe CAD data. Finally, initial setting data for the lasting machine are extracted from the last CAD data and initial setup parts of the lasting machine.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• 제8권4호
• /
• pp.19-28
• /
• 1999

In order to control thermal deformation of the machine origin of machine tools a empirical model and a compensation system have been developed, Prior to empirical modeling the volumetric error considering shape errors and joint errors of slides is formulated through the homogeneous transformation matrix (HTM) and kinematic chain. Simulation results of the HTM method show that the thermal error of the machine origin is more critical than position-dependent errors. In order to make a stable and effective software error compensation system the GMDH (Group Method of Data Handling) models are constructed to estimate the thermal deformation of the machine origin by measuring deformation data and temperature data. A test bar and gap sensors are used to measure the deformation data. In order to compensate the estimated error the work origin shift method is developed by implementing a digital I/O interface board between a CNC controller and an IBM PC. The method shifts the work origin as much as the amounts which are calculated by the pre-established thermal error model. The experiment results for a vertical machining center show that the thermal deformation of the machine origin is reduced within $\pm$5$mu extrm{m}$.

• Journal of radiological science and technology
• /
• 제42권1호
• /
• pp.9-17
• /
• 2019

Although digital mammography is a representative method for breast cancer detection. It has a limitation in detecting and classifying breast tumor due to superimposed structures. Machine learning, which is a part of artificial intelligence fields, is a method for analysing a large amount of data using complex algorithms, recognizing patterns and making prediction. In this study, we proposed a technique to improve the diagnostic accuracy of energy-selective mammography by training data using the machine learning algorithm and using dual-energy measurements. A dual-energy images obtained from a photon-counting detector were used for the input data of machine learning algorithms, and we analyzed the accuracy of predicted tumor thickness for verifying the machine learning algorithms. The results showed that the classification accuracy of tumor thickness was above 95% and was improved with an increase of imput data. Therefore, we expect that the diagnostic accuracy of energy-selective mammography can be improved by using machine learning.

• Progress in Medical Physics
• /
• 제28권3호
• /
• pp.129-133
• /
• 2017

The machine log files recorded by a scanning control unit in proton beam therapy system have been studied to be used as a quality assurance method of scanning beam deliveries. The accuracy of the data in the log files have been evaluated with a standard calibration beam scan pattern. The proton beam scan pattern has been delivered on a gafchromic film located at the isocenter plane of the proton beam treatment nozzle and found to agree within ${\pm}1.0mm$. The machine data accumulated for the scanning beam proton therapy of five different cases have been analyzed using a statistical method to estimate any systematic error in the data. The high-precision scanning beam log files in line scanning proton therapy system have been validated to be used for off-line scanning beam monitoring and thus as a patient-specific quality assurance method. The use of the machine log files for patient-specific quality assurance would simplify the quality assurance procedure with accurate scanning beam data.

• Journal of Environmental Science International
• /
• 제26권4호
• /
• pp.521-527
• /
• 2017

Solar radiation forecasts are important for predicting the amount of ice on road and the potential solar energy. In an attempt to improve solar radiation predictability in Jeju, we conducted machine learning with various data mining techniques such as tree models, conditional inference tree, random forest, support vector machines and logistic regression. To validate machine learning models, the results from the simulation was compared with the solar radiation data observed over Jeju observation site. According to the model assesment, it can be seen that the solar radiation prediction using random forest is the most effective method. The error rate proposed by random forest data mining is 17%.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 한국정밀공학회 2001년도 춘계학술대회 논문집
• /
• pp.439-442
• /
• 2001

The Accuracy of a free-form artifact is affected by machine tool errors, machining process errors, environmental causes and other uncertainty. This paper deals with methodological approach about machine tool errors that are defined the relationship between CMM and OMM inspections of the free-form artifact. In order to analyze the measurement data, Reverse engineering was used. In other words, Surface of Free-Form Artifact is generated by NURBS surface approximation method. Finally, Volumetric error map is made to compare surface of CMM data with that of OMM data.

• Journal of the Korean Data and Information Science Society
• /
• 제25권2호
• /
• pp.455-464
• /
• 2014

Unlabeled examples are easier and less expensive to obtain than labeled examples. Semisupervised approaches are used to utilize such examples in an eort to boost the predictive performance. This paper proposes a novel semisupervised classication method named transductive least squares support vector machine (TLS-SVM), which is based on the least squares support vector machine. The proposed method utilizes the dierence convex algorithm to derive nonconvex minimization solutions for the TLS-SVM. A generalized cross validation method is also developed to choose the hyperparameters that aect the performance of the TLS-SVM. The experimental results conrm the successful performance of the proposed TLS-SVM.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 한국공작기계학회 1999년도 추계학술대회 논문집 - 한국공작기계학회
• /
• pp.554-559
• /
• 1999

The purpose of this study is the machining of texture shapes by the contour fitting data. The hardware of the system comprises PC and scanning system, CO2 laser machine. There are four steps, (1) text image loading using scanning shapes or 2D image files, (2) generation of contour fitting data by the line and arc, cubic Bezier curve, (3) generation of NC code from the contouring fitting data, (4) machining by the DNC system. It is developed a software package, with which can conduct a micro CAM system of CNC laser machine in the PC without economical burden.