• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.11
• /
• pp.509-520
• /
• 2015

College Scholastic Ability Test(CSAT) is a primary test to evaluate the study achievement of high-school students and used by most universities for admission decision in South Korea. Because its level of difficulty is a significant issue to both students and universities, the government makes a huge effort to have a consistent difficulty level every year. However, the actual levels of difficulty have significantly fluctuated, which causes many problems with university admission. In this paper, we build two types of data-driven prediction models to predict correct answer rate and to identify significant factors for CSAT English test through accumulated test data of CSAT, unlike traditional methods depending on experts' judgments. Initially, we derive candidate question-specific factors that can influence the correct answer rate, such as the position, EBS-relation, readability, from the annual CSAT practices and CSAT for 10 years. In addition, we drive context-specific factors by employing topic modeling which identify the underlying topics over the text. Then, the correct answer rate is predicted by multiple linear regression and level of difficulty is predicted by classification tree. The experimental results show that 90% of accuracy can be achieved by the level of difficulty (difficult/easy) classification model, whereas the error rate for correct answer rate is below 16%. Points and problem category are found to be critical to predict the correct answer rate. In addition, the correct answer rate is also influenced by some of the topics discovered by topic modeling. Based on our study, it will be possible to predict the range of expected correct answer rate for both question-level and entire test-level, which will help CSAT examiners to control the level of difficulties.

• Journal of the Korean Society for information Management
• /
• v.35 no.4
• /
• pp.141-164
• /
• 2018

Recently, as the amount of academic literature has increased rapidly and complex researches have been actively conducted, researchers have difficulty in analyzing trends in previous research. In order to solve this problem, it is necessary to classify information in units of academic papers. However, in Korea, there is no academic database in which such information is provided. In this paper, we propose an automatic classification system that can classify domestic academic literature into multiple classes. To this end, first, academic documents in the technical science field described in Korean were collected and mapped according to class 600 of the DDC by using K-Means clustering technique to construct a learning set capable of multiple classification. As a result of the construction of the training set, 63,915 documents in the Korean technical science field were established except for the values in which metadata does not exist. Using this training set, we implemented and learned the automatic classification engine of academic documents based on deep learning. Experimental results obtained by hand-built experimental set-up showed 78.32% accuracy and 72.45% F1 performance for multiple classification.

• Journal of the Korean Society for information Management
• /
• v.13 no.2
• /
• pp.97-120
• /
• 1996

This study reports the results of an exploratory investigation into building number technique applied by a sample of novice students using Electronic Dewey(ED). For control purposes, a second sample used the same classification schedule but in print format. In this experiment, the DDC 20 in print had a higher success rate than ED. While the searches in ED took shorter time than DDC in print, students using DDC 20 in print were able to build more correct numbers than ED. Novice users did not take full advantage of system features of ED. Therefore, it is important to note the overall difficulty with which users could build numbers from the ED after little training. However, it is interesting to note that the students in the CD-ROM group were more interested in building numbers than their colleagues in the Print group. The correlation between the grade of English course and the score in experiment was negative in Print group. And the correlation between GPA and the score in experiment was a little positive. The correlation between the time and the score in experiment was quitenegative.

• Biomolecules & Therapeutics
• /
• v.27 no.3
• /
• pp.254-264
• /
• 2019

Atherosclerosis is a widespread and chronic progressive arterial disease that has been regarded as one of the major causes of death worldwide. It is caused by the deposition of cholesterol, fats, and other substances in the tunica intima which leads to narrowing of the blood vessels, loss of elasticity, and arterial wall thickening, thus causing difficulty in blood flow. Natural products have been used as one of the most important strategies for the treatment and prevention of cardiovascular diseases for a long time. In recent decades, as interests in natural products including medicinal herbs have increased, many studies regarding natural compounds that are effective against atherosclerosis have been conducted. The purpose of this review is to provide a brief overview of the natural compounds that have been used for the treatment and prevention of atherosclerosis, and their mechanisms of action based on recent research.

• Journal of Construction Engineering and Project Management
• /
• v.5 no.1
• /
• pp.32-39
• /
• 2015

Construction industry is an aggregate of information that diverse information is integrated and controlled. To implement successful construction projects, it can be said that the information management is very important. In particular, because information of construction sites is controlled in a form of documents, importance of the document management in construction has been increased. But, by controlling information through documents, there are difficult problems in writing and classification of the documents and preservation and utilization of the information. Also, due to incompletion of the information management system, difficulty in systematic info management arises. For this reason, this study intends to suggest the document information breakdown structure for controlling document info efficiently which is generated at construction sites. For this, through the examination of preceding studies, establishment of the concept of the document info breakdown structure, the space breakdown structure, and the info breakdown structure, availability of document information is intended to heighten.

• Korean Management Science Review
• /
• v.9 no.1
• /
• pp.31-40
• /
• 1992

The size of software collection at any organization is increasing rapidly with the use of computer becomes popular for various applications. This trend points out the importance of proper 'catalog'or 'directory' for software collections, however, only 'title list' is found in practice. The problem of no proper directory with software collection would be more serious than the problem with book collection since browsing is not possible with softwares. Many of the expensive softwares will be a waste of money to purchase and to hold if no catalog about them is provided. In this paper the difficulties in applying AACR2 to software cataloging and the most important characteristic of computer software are discussed as well as presenting the necessity of localized information due to the requirement of a machine to be used, the difficulty using subject-based-call-number scheme, an expected way of searching software to design a location indicator, software qualifications to be cataloged, and so on. In addition, an analysis of randomly selected softwares after a cataloging practice is followed.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.7
• /
• pp.1100-1104
• /
• 2017

A detector-based approach shows the limited performances for the defect inspections such as shallow fine cracks and indistinguishable defects from background. Deep learning technique is widely used for object recognition and it's applications to detect defects have been gradually attempted. Deep learning requires huge scale of learning data, but acquisition of data can be limited in some industrial application. The possibility of applying CNN which is one of the deep learning approaches for surface defect inspection is investigated for industrial parts whose detection difficulty is challenging and learning data is not sufficient. VOV is adopted for pre-processing and to obtain a resonable number of ROIs for a data augmentation. Then CNN method is applied for the classification. Three CNN networks, AlexNet, VGGNet, and mofified VGGNet are compared for experiments of defects detection.

• Journal of Integrative Natural Science
• /
• v.7 no.2
• /
• pp.79-86
• /
• 2014

Amyloid oligomers are believed to play important causal roles in many types of amyloid-related degenerative diseases. Many different laboratories have reported amyloid oligomers that differ in size, morphology, toxicity, and method of preparation or purification, raising the question of the structural relationships among these oligomer preparations. The structural plasticity that has been reported to occur in amyloid formed from the same protein sequence indicates that it is quite possible that different oligomer preparations may represent distinct structural variants. In view of the difficulty in determining the precise structure of amyloids, conformation- and epitope-specific antibodies may provide a facile means of classifying amyloid oligomer structures. Conformation-dependent antibodies that recognize generic epitopes that are specifically associated with distinct aggregation states of many different amyloid-forming sequences indicate that there are at least two fundamentally distinct types of amyloid oligomers: fibrillar and prefibrillar oligomers. Classification of amyloid oligomers according to their underlying structures may be a more useful and rational approach than relying on differences in size and morphology.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.3
• /
• pp.491-497
• /
• 2007

As dispersed generators was driven in condition interconnecting with utility, it could cause a variety of new effects to the original distribution system that was running as considered only the one-way power flow. Therefore, the protection devices that is builted in distribution system should be designed to be able to operate with disposing of not only a fault of the generator, but also utility condition. Especially, the fault of the feeder interconnected with Dispersed Generator can cause the islanding phenomenon of open DG(Dispersed Generators). This phenomenon has many problems such as a machinery damage, electricity qualify degradation and a difficulty of the system recovery. In the fault therefore, we must separate Dispersed Generator from the system quickly. In this paper, for the fault classification of the interconnected DG and the outside feeder we judge the fault of the interconnected DG and the outside feeder in HMI through data provided by IED(Intelligent Electronic Device) on the network and decide whether it operates or not by sending the result to each relay.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2485-2487
• /
• 2000

In this paper, we use a fuzzy algorithm to detect and diagnose the error which is caused by time delay of the computer-controlled system. Generally, a computer-controlled system is composed of computer and process. And they communicate the data each other. In data communication, error occurs by some reasons, such as noise, disturbance, hardware defect, etc. Time delay is one of the reasons. And time delay makes it difficult to distinguish whether the system really has a problem or not. Therefore, we need to detect and diagnose the error from time delay. For difficulty of modeling and ambiguity of classification, we use a fuzzy algorithm. To verify the better performance of the proposed algorithm, we exemplified by some simulation results.