• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.4
• /
• pp.236-244
• /
• 2024

A representative problem in domestic chestnut industry is the high loss of flesh due to excessive knife peeling in order to increase the peeling rate, resulting in a decrease in production efficiency. In this study, a prediction model for weight loss rate of chestnut by stage of knife peeling process was developed as undergarment study to optimize conditions of the machine. 51 control conditions of the two-stage blade peeler used in the experiment were derived and repeated three times to obtain a total of 153 data. Machine learning(ML) models including artificial neural network (ANN) and random forest (RF) were implemented to predict the weight loss rate by chestnut peel stage (after 1st peeling, 2nd peeling, and after final discharge). The performance of the models were evaluated by calculating the values of coefficient of determination (R), normalized root mean square error (nRMSE), and mean absolute error (MAE). After all peeling stages, RF model have better prediction accuracy with higher R values and low prediction error with lower nRMSE and MAE values, compared to ANN model. The final selected RF prediction model showed excellent performance with insignificant error between the experimental and predicted values. As a result, the proposed model can be useful to set optimum condition of knife peeling for the purpose of minimizing the weight loss of domestic chestnut flesh with maximizing peeling rate.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.1
• /
• pp.90-99
• /
• 2003

Java applets are downloaded from web server through internet and executed in Java Virtual Machine of clients' browser. Before execution of java applets, JVM checks bytecode program with bytecode verifier and performs runtime tests with interpreter. However, these tests will not protect against undesirable runtime behavior of java applets, such as denial of service attack, email forging attack, URL spoofing attack, and annoying sound attack. In order to protect malicious applets, a technique used in this paper is java bytecode modification. This technique is used to restrict applet behavior or insert code appropriate to profiling or other monitoring efforts. Java byte modification is divided into two general forms, class-level modification involving subclassing non-final classes and method-level modification used when control over objects from final classes or interface. This paper showed that malicious applets are controlled by java bytecode modification using proxy server. This implementation does not require any changes in the web sever, JVM or web browser.

• The Journal of Korean Academy of Prosthodontics
• /
• v.37 no.3
• /
• pp.358-374
• /
• 1999

This study measured the relative displacements of the five-unit fixed partial dentures as cast with the same fixed partial dentures sectioned and assembled by investment-soldering, solder-ing stand-soldering, and cast-joining techniques A total of fifteen specimens using a type IV gold alloy were one-piece cast as control and then sectioned and assembled five test specimens for each method were prepared. A computerized three dimensional coordinate measuring machine and specially designed cylinder for this study were used. Displacement was defined by six displacement variables for the each of cylinders incorporated in each casting: three component displacements(${\Delta}Lx,\;{\Delta}Ly,\;and\;{\Delta}Lz$) and rotational displacements(${\Delta}{\theta}x,\;{\Delta}{\theta}y,\;{\Delta}{\theta}z$). The global displacement was computed using the mathematical formula ${\Delta}R$ = Global displacement =$\sqrt{{(x'-x)}^2+{(y'-y)}^2+{(z'-z)}^2}$ Under the conditions of this study, the following conclusions were drawn: 1. The investment-soldering group showed the largest mean value of final global displacements, followed by stand-soldering group, cast-joining group and one-piece cast group. However, between the mean values of final global displacement for the cast-joining group and one-piece cast group, there was no significant difference. 2. For investment-soldering and stand-soldering groups, the greater global displacements were recorded in soldering phase than in indexing or investing phase. 3. For one-piece cast group, the displacements occured mostly in the casting phase. And for cast-joining group, there was no significant difference in global displacements among the fabricating procedures. 4. Intercentroidal distance decreased in framework-patterning, solder-indexing, solder-standing, and soldering phases, but increased in investment block-investing and casting phases. 5 Specially designed cylinder for touch-trigger type coordinate measuring machine was validated.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.5C
• /
• pp.523-531
• /
• 2003

Java applets are downloaded from web server through internet and executed in Java Virtual Machine of clients'browser. Before execution of java applets, JVM checks bytecode program with bytecode verifier and performs runtime tests with interpreter. However, these tests will not protect against undesirable runtime behavior of java applets, such as denial of service attack, email forging attack, URL spoofing attack, or annoying sound attack. In order to protect malicious applets, a technique used in this paper is java bytecode modification. This technique is used to restrict applet behavior or insert code appropriate to profiling or other monitoring efforts. Java byte modification is divided into two general forms, class-level modification involving subclassing non-final classes and method-level modification used when control over objects from final classes or interface. This paper showed that malicious applets are controlled by java bytecode modification using proxy server. This implementation does not require any changes in the web sever, JVM or web browser.

• Journal of Gifted/Talented Education
• /
• v.26 no.2
• /
• pp.319-346
• /
• 2016

In this study, by examining the characteristics of boys and girls which would appear in the performance process of Goldberg machine activities, it would be attempted to provide the implications for the development and teaching methods of gifted and talented programs. The object of study was organized into separate 2 groups of boys and girls by each, composed of a total of 16 people among 5th graders of the gifted class in elementary school, located in Gyeonggi province. The final assignment was to make the Goldberg machine in order to have the beads get to the target spot latest, in which the analysis was implemented qualitatively by participating in and observing the performance process of students. After dividing the Goldberg machine activities into the steps of planning, production, outcome, assessment and reflection, their analysis results are as follows: First, in the planning stage, the girls explained minutely the process of Goldberg machine in writing, whereas the boys represented it visually simply. Second, in the production stage, the boys showed the task commitment by trying to realize the machine as designed initially, but the girls showed their appearance to represent it simply and easily. Third, in the sophistication and efficiency of the machine production, the boys were superior to the girls, and in the creativity and diversity of the use of materials, the girls were more excellent. Fourth, in the assessment and reflection, the boys evaluated it individually, and the girls showed their appearance to evaluate it by reflecting others'thinking. Hence, when developing and teaching the gifted and talented programs, it would be required that the teaching and learning contents be recomposed by considering the gender, or that the various class strategies be sought. Further, the broader and more systematic studies, on the performance process of gifted students based on the gender, should be carried out.

• Journal of KIISE:Software and Applications
• /
• v.31 no.3
• /
• pp.369-378
• /
• 2004

In partially free word order languages like Korean and Japanese, the rule-based method is effective for text chunking, and shows the performance as high as machine learning methods even with a few rules due to the well-developed overt Postpositions and endings. However, it has no ability to handle the exceptions of the rules. Exception handling is an important work in natural language processing, and the exceptions can be efficiently processed in memory-based teaming. In this paper, we propose a hybrid of rule-based method and memory-based learning for Korean text chunking. The proposed method is primarily based on the rules, and then the chunks estimated by the rules are verified by memory-based classifier. An evaluation of the proposed method on Korean STEP 2000 corpus yields the improvement in F-score over the rules or various machine teaming methods alone. The final F-score is 94.19, while those of the rules and SVMs, the best machine learning method for this task, are just 91.87 and 92.54 respectively.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.5
• /
• pp.161-168
• /
• 2020

Predicting outcome of the sports enables teams to establish their strategy by analyzing variables that affect overall game flow and wins and losses. Many studies have been conducted on the prediction of the outcome of sports events through statistical techniques and machine learning techniques. Predictive performance is the most important in a game prediction model. However, statistical and machine learning models show different optimal performance depending on the characteristics of the data used for learning. In this paper, we propose a new ensemble model to predict English Premier League soccer games using statistical models and the machine learning models which showed good performance in predicting the results of the soccer games and this model is possible to select a model that performs best when predicting the data even if the data are different. The proposed ensemble model predicts game results by learning the final prediction model with the game prediction results of each single model and the actual game results. Experimental results for the proposed model show higher performance than the single models.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.8
• /
• pp.17-25
• /
• 2013

TIn this paper, we propose a model automatically scoring a student's answer for a descriptive problem by using a voting method. Considering the model construction cost, the proposed model does not separately construct the automatic scoring model per problem type. In order to utilize features useful for automatically scoring the descriptive answers, the proposed model extracts feature values from the results, generated by comparing the student's answer with the answer sheet. For the purpose of improving the precision of the scoring result, the proposed model collects the scoring results classified by a few machine learning based classifiers, and unanimously selects the scoring result as the final result. Experimental results show that the single machine learning based classifier C4.5 takes 83.00% on precision while the proposed model improve the precision up to 90.57% by using three machine learning based classifiers C4.5, ME, and SVM.

• Journal of Convergence for Information Technology
• /
• v.9 no.8
• /
• pp.248-253
• /
• 2019

Early diagnosis of mild cognitive impairment (MCI) can reduce the incidence of dementia. This study developed the MCI prediction model for the elderly in Korea. The subjects of this study were 3,240 elderly (1,502 men, 1,738 women) aged 65 and over who participated in the Korean Longitudinal Survey of Aging (KLoSA) in 2012. Outcome variables were defined as MCI prevalence. Explanatory variables were age, marital status, education level, income level, smoking, drinking, regular exercise more than once a week, average participation time of social activities, subjective health, hypertension, diabetes Respectively. The prediction model was developed using Restricted Boltzmann Machine (RBM) neural network. As a result, age, sex, final education, subjective health, marital status, income level, smoking, drinking, regular exercise were significant predictors of MCI prediction model of rural elderly people in Korea using RBM neural network. Based on these results, it is required to develop a customized dementia prevention program considering the characteristics of high risk group of MCI.

• Journal of International Academy of Physical Therapy Research
• /
• v.12 no.1
• /
• pp.2261-2266
• /
• 2021

Background: Deep learning related research works on website medical images have been actively conducted in the field of health care, however, articles related to the musculoskeletal system have been introduced insufficiently, deep learning-based studies on classifying orthopedic manual therapy images would also just be entered. Objectives: To create a deep learning model that categorizes cervical mobilization images and establish a web application to find out its clinical utility. Design: Research and development. Methods: Three types of cervical mobilization images (central posteroanterior (CPA) mobilization, unilateral posteroanterior (UPA) mobilization, and anteroposterior (AP) mobilization) were obtained using functions of 'Download All Images' and a web crawler. Unnecessary images were filtered from 'Auslogics Duplicate File Finder' to obtain the final 144 data (CPA=62, UPA=46, AP=36). Training classified into 3 classes was conducted in Teachable Machine. The next procedures, the trained model source was uploaded to the web application cloud integrated development environment (https://ide.goorm.io/) and the frame was built. The trained model was tested in three environments: Teachable Machine File Upload (TMFU), Teachable Machine Webcam (TMW), and Web Service webcam (WSW). Results: In three environments (TMFU, TMW, WSW), the accuracy of CPA mobilization images was 81-96%. The accuracy of the UPA mobilization image was 43~94%, and the accuracy deviation was greater than that of CPA. The accuracy of the AP mobilization image was 65-75%, and the deviation was not large compared to the other groups. In the three environments, the average accuracy of CPA was 92%, and the accuracy of UPA and AP was similar up to 70%. Conclusion: This study suggests that training of images of orthopedic manual therapy using machine learning open software is possible, and that web applications made using this training model can be used clinically.