• Journal of the Korea Society of Computer and Information
• /
• v.20 no.10
• /
• pp.31-38
• /
• 2015

In this paper, we propose the ensemble-based classification model which extracts just new data patterns from the streaming-data by using clustering and generates new classification models to be added to the ensemble in order to reduce the number of data labeling while it keeps the accuracy of the existing system. The proposed technique performs clustering of similar patterned data from streaming data. It performs the data labeling to each cluster at the point when a certain amount of data has been gathered. The proposed technique applies the K-NN technique to the classification model unit in order to keep the accuracy of the existing system while it uses a small amount of data. The proposed technique is efficient as using about 3% less data comparing with the existing technique as shown the simulation results for benchmarks, thereby using clustering.

• Journal of Pharmaceutical Investigation
• /
• v.40 no.spc
• /
• pp.1-7
• /
• 2010

Since the introduction of the biopharmaceutics classification system (BCS) in 1995, it has viewed as an effective tool to categorize drugs in terms of prediction for bioavailability (BA) and bioequivalence (BE). The BCS consist of four drug categories: class I (highly soluble and highly permeable), class II (low soluble and highly permeable), class III (highly soluble and low permeable) and class IV (low soluble and low permeable), and almost all drugs belong to one of these categories. Likewise, classifying drugs into four categories according to their solubility and permeability is simple and relatively not controversial, and thus the FDA adopted the BCS as a science-based approach in establishing a series of regulatory guidance for the industry. Actually, many pharmaceutical companies have gained a lot of benefits, which directly connect to cost loss and failure decrease in the early stage of drug development. Recently, instead of solubility, using dissolution characteristics (e.g. intrinsic dissolution rate) have provided an improvement in the classification in correlating more closely with in vivo drug dissolution rather than solubility by itself. Furthermore, a newly modified-version of BCS, biopharmaceutics drug disposition classification system (BDDCS), which classify drugs into four categories according to solubility and metabolism, has been introduced and gained much attention as a new insight in respect with the drug classification. This report gives a brief overview of the BCS and its implication, and also introduces the recent new trend of drug classification.

• Journal of Korea Technology Innovation Society
• /
• v.11 no.4
• /
• pp.592-612
• /
• 2008

The convergence of technology as a major breakthrough in technology innovation has been generated and developed. With the advent of the 21st century characterized by the knowledge-based economy, it is happening even more frequently and vigorously than ever. However, study on classification system or definition of notions is necessary as the convergence of technology is still in the early stage. The existing classification system has limited application to the convergence of technology. With no standard available for the classification of the technology convergence, there has been much concern for duplicative R&D investment. On this ground, a new and reformed classification system for the convergence of technology and definition of notions are needed. Previous studies regard base technologies used in the technology convergence and new convergence technologies as important. However, this paper views convergence of technology as a dynamic phenomenon and puts an emphasis on the cause of technology convergence, not on the new technology itself and examines whether base technologies go back to their original state after the completion of technology convergence. This kind of approach will classify technology convergence by characteristics. Although more researches including quantitative analysis are necessary, this paper expects to offer help with further researches on classification system.

• Clean Technology
• /
• v.11 no.2
• /
• pp.91-104
• /
• 2005

We studied more effective Environmental R&D management system which can be used in the budget allocation of the national R&D budget by the newly opened governmental administration system of science and technology. As a first step, we surveyed classification systems for the environmental technology which can be applied to the analysis of the national R&D budget. Each organization, either inland and oversea, has its own classification system for the environmental technology. However, we found that each systems can lead different results. It would be beneficial if an efficient standard classification system could be generated. The new system classified the environmental technology by the environmental management area(media) like most present systems rather than by the technical characteristics. The new system included all the areas of the environmental management and maintained consistent principles. The new system would help efficient management of the national R&D of the environment technology, if the national standard classification can be changed by the new system. To the researchers and professionals the new system will help them noticing the status of the environmental R&D and avoiding duplicate research, so that they can do research more efficiently and can achieve maximized results.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.9 no.1
• /
• pp.125-155
• /
• 1999

The objectives of this study were (1) an investigation of the current status of the hazardous substances, (2) a comparative analysis of classification system of hazardous substances between Korea and other countries, (3) a development of the new classification system in Korea, finally a review of prerequisites for effective enforcement of the new system. The results are summarized as follows : 1) The backbone of classification system of hazardous substances in Korea is based on Japanese classification system and EC(European Committee) guideline of hazardous substance. There are many problems in managing and handling of hazardous substances due to discordant of Japanese and EC system. It is desirable to revise the classification system to be harmonized with international guideline for example, guideline of IFCS(The Inter-governmental Forum on Chemical Safety) and EC guideline. 2) There are several problems in definitions of corrosive, sensitizing and irritation in MSDS Code of Ministry of Labour. It is desirable to reform those definitions. 3) Among the hazardous substances under the current system, there are several substances such as, beta-propiolactone, methyl bromide, ethyleneimine, etc that are not used and produced in Korea. It is desirable to exclude the substances from the list of controlled substances. 4) The section 39, about 'substance that designated to attach warning label' in Korea Occupational Safety and Health Act(OSHAct) should be eliminated, because above section is unnecessary under the MSDS system. 5) The researchers recommend to establish a new list of hazardous substance which are controlled by OSHAct. It is desirable that the new list is consist of two types. The first type is 'a list of the specially controlled substances' and the other is 'the generally controlled substances'. 6) It is recommended that the specially controlled substances should include hazardous substances that are highly toxic and widely used in Korea. And the generally controlled substances should include hazardous substances that can be analysed by instrument and carry threshold. limit values(TLV's).

• The Journal of Korean Association of Computer Education
• /
• v.7 no.5
• /
• pp.83-91
• /
• 2004

Through this research, it was possible to set up classification system between 'Harmful information site' and 'General site' that badly effect to teenagers emotional health. To intercept those entire harmful information sites, it using contents basis isolating. Instead of using existing methods, it picks most frequent using composed key words and adds all those harmful words' harmfulness degree point by using 'ICEC(Information Communication Ethics Committee)' suggested harmful word classification. To testify harmful information blocking system, to classify the harmful information site, set standard harmfulness degree point as 3.5 by the result of a fore study, after that pick up a hundred of each 'Harmful information site' and 'General site' randomly to classify them through new classification system. By this classification could found this new classification system classified 78% of 'Harmful Site' to "Harmful information site' and 96% of 'General Site' to 'General site'. As a result, successfully confirm validity of this new classification system.

• Proceedings of the KIEE Conference
• /
• 2006.10c
• /
• pp.470-472
• /
• 2006

Sensor networks are the results of convergence of very important technologies such as wireless communication and micro electromechanical systems. In recent years, sensor networks found a wide applicability in various fields such as environment and health, industry scene system monitoring, etc. A very important step for these many applications is pattern classification and recognition of data collected by sensors installed or deployed in different ways. But, pattern classification and recognition are sometimes difficult to perform. Systematic approach to pattern classification based on modem learning techniques like Multivariate Gaussian mixture models, can greatly simplify the process of developing and implementing real-time classification models. This paper proposes a new recognition system which is hierarchically composed of many sensor nodes having the capability of simple processing and wireless communication. The proposed system is able to perform context classification of sensed data using the Multivariate Gaussian function. In order to verify the usefulness of the proposed system, it was applied to intelligent dust collecting system.

• Journal of the Korea Safety Management & Science
• /
• v.24 no.4
• /
• pp.61-70
• /
• 2022

In spite of lab safety act for over 10 years, over 100 safety accidents in the laboratory have been constantly occurring. The ideal safety management system is to prevent accidents by differential classifying and managing laboratory regulatory materials according to the risk level. In order to approach this system, in-depth interviews with safety managers were first conducted to identify the current status of safety management in domestic university laboratories. And then through comparative analysis of safety management systems in domestic and foreign laboratories, a new regulatory substance classification standard based on the analysis of the hazards and the classification of risk grades, and a safety management system are proposed. From this study, it will contribute to the creation of a safe laboratory environment by differential classification and management laboratory regulatory materials based on the risk level.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.110.3-110
• /
• 2002

$\textbullet$ Lane detection by TFALDA $\textbullet$ SVM for large scale data and multiclass classification problem $\textbullet$ TLC Classification $\textbullet$ Lateral offset estimation by IPT $\textbullet$ Lane departure warning by a fuzzy system $\textbullet$ Experimental results by HiLS $\textbullet$ Conclusion

• Journal of the Korea Society for Simulation
• /
• v.18 no.1
• /
• pp.63-70
• /
• 2009

This project presents a classification methodology for malicious codes in Windows OS (Operating System) environment, develops a test classification system. Thousands of malicious codes are brought in every day. In a result, classification system is needed to analyzers for supporting information which newly brought malicious codes are a new species or a variety. This system provides the similarity for analyzers to judge how much a new species or a variety is different to the known malicious code. It provides to save time and effort, to less a faulty analysis. This research includes the design of classification system and test system. We classify the malicious codes to 9 groups and then 9 groups divide the clusters according to the each property. This system provides the similarity for analyzers to save time and effort. It is used prospect system of malicious code in the future.