• Journal of the Korean Society for information Management
• /
• v.10 no.2
• /
• pp.111-143
• /
• 1993

There are two types of bibliographic database, the one IS bibliographic database with creativlty. the other is bibliographic database without creativlty, The purpose of this paper is to investigate the problems of bibliographic database with creatlvity by giving consideration of copyright law and the problems of bihographic database w~thout creat~vity by sui genem system. Thus thls paper provided the methods that resolve the problems of the current copyright law in order to protect all types of database being produced by way of the cooperation between the libraries and the information service centers. Secondly, this paper provlded the legal devlces to protect bibliographic database produced by the libraries wh~ch cannot be protected due to the lack of the creatlvity in the selection or arrangement of the material.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.6
• /
• pp.760-765
• /
• 2009

Since the features in object-images such as color and shape cannot clearly express the characteristic of objects, those features lead to vagueness of object-image recognition. Recently there have been studied on object-image recognition based on knowledge base in order to reduce the vagueness. However, because images are represented by numerical information but knowledge bases are represented by conceptual information, combining two kinds of information is difficult. In this paper, we compose knowledge base by using ontology to reduce the gap between the two kinds of information, and propose a method for object-image recognition to reduce the vagueness by using information on inner-object. Moreover, we confirm the usefulness of the proposed method through the experiments on object-image recognition in fruit domain.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05c
• /
• pp.1551-1554
• /
• 2003

최근 다양한 의료정보 시스템이 개발되어, 그 사용이 급증하고 있다. 이 들 각각의 의료정보 시스템에서 발생, 축적된 의료정보는 분산 이질의 형태를 가지며, 또한 같은 의미를 갖는 의료정보가 각기 다른 구조와 용어로 기술되어 축적되는 것이 일반적이다. 이와 같이 개별적으로 개발, 활용되어 온 의료정보를 웹 상에서 통합하여, 단일화 된 의료정보 검색 기능을 제공하기 위해서는 이들 의료정보의 의미적 연관성을 고려한 정보의 통합, 검색 기술의 개발이 필수적이다. 본 논문에서는 의미적 의료정보의 통합을 위한 UMLS와 LOINC 데이터베이스 기반의 연관 값 지식베이스의 설계 및 개발 방식을 제안한다. 웹 상에 존재하는 각종 분산 이질 형태의 의료정보는 XML을 공통 데이터 구조로 하여 통합되며, 정보 통합의 과정에서 연관 값 지식베이스를 참조하여 의미적 관련도가 높은 의료정보(구조 정보와 내용 정보)는 상호 연결되어, 진정한 의미의 정보 통합을 구현하게 된다. 지식베이스는 용어별로 식별자, 요소명, 연관값, 복수형, 동의어, 한글 이름 등의 필드틀 가지며, 현재 상담, 처방, 보험, 의료용어, 증상, 임상결과 등 적용분야 별로 작성된 연관 값 지식베이스가 구현되어 있다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.1000-1002
• /
• 2005

In the Bluetooth piconet in which up to 7 slave devices can be connected simultaneously at one network instance, the wireless data expected to be sent over to the RF interface should be sliced by the unit of 1 micrometer, which is a requirement in the specification of the Bluetooth version 1.1. In this contribution, we have designed a digital low-pass filter which is able to slice the unstable analog signal fed from the RF interface to the Baseband, by the uniform unit of 1 micrometer, and is also capable of removing the possible noise which can be caused by the analog circuit system. The low-pass filter operated well in the various modes of the Bluetooth RF embedded Baseband chip such as sleep mode, normal mode, and high-speed mode at 12MHz, 24MHz, and 48MHz respectively.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2011.10a
• /
• pp.208-209
• /
• 2011

In this paper, we introduce a Naive Bayes learning algorithm for learning and reasoning in Map-Reduce model based environment. For this purpose, we use Apache Mahout to execute Distributed Naive Bayes on University of California, Irvine (UCI) benchmark data sets. From the experimental results, we see that Apache Mahout' s Distributed Naive Bayes algorithm is comparable to WEKA' s Naive Bayes algorithm in terms of performance. These results indicates that in the future Big Data environment, Map-Reduce model based systems such as Apache Mahout can be promising for machine learning usage.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.1
• /
• pp.9-16
• /
• 2002

Join operation exert a great effect on the performance of system in temporal database as in the relational database. Especially, as for the temporal join, the optimization of interval partition decides the performance of query processing. In this paper, to improve the efficiency of parallel join query in temporal database. I proposed Minimum Interval Partition(MIP) scheme that time interval partitioning. The validity of this MIP algorithm that decides minimum breakpoint of the partition is proved by example scenario and I confirmed improved efficiency as compared with existing partition algorithm.

• Journal of KIISE:Databases
• /
• v.32 no.1
• /
• pp.86-99
• /
• 2005

In this paper, we suggest a system which automatically verifies and tunes transaction processing database applications based on program analysis technology. This system automatically verifies two kinds of transaction processing errors. The first case is the un-closed transaction. In this case, data is not updated as expected or performance of overall system can decrease seriously by locking some database tables until the process terminates. The second case is the miss-use of transaction isolation(inking) level. This causes runtime exception or abnormal termination of the program depending on runtime environment. This system automatically tunes two kinds of inefficient definition of transaction processing which decrease the performance of overall system. The first case happens when opened transaction is closed too late. And the second case happens when transaction isolation level is set too high.

• Journal of KIISE
• /
• v.42 no.8
• /
• pp.1065-1070
• /
• 2015

The performance of many natural language processing applications depends on the knowledge base as a major resource. WordNet, YAGO, Cyc, and BabelNet have been extensively used as knowledge bases in English. In this paper, we propose a method to construct a YAGO-style knowledge base automatically for Korean (hereafter, K-YAGO) from Wikipedia and YAGO. The proposed system constructs an initial K-YAGO simply by matching YAGO to info-boxes in Wikipedia. Then, the initial K-YAGO is expanded through the use of a machine learning technique. Experiments with the initial K-YAGO shows that the proposed system has a precision of 0.9642. In the experiments with the expanded part of K-YAGO, an accuracy of 0.9468 was achieved with an average macro F1-measure of 0.7596.

• Resources Recycling
• /
• v.30 no.5
• /
• pp.3-9
• /
• 2021

In Korea, base metals such as lead, zinc, copper, tin, nickel, and aluminum have a polarized supply and demand structure. Despite the presence of world-class producers of lead, zinc, and copper, and their production is insufficient. And there are no domestic producers of tin, nickel, and aluminum, Thus, most of the domestic demand is dependent on imports. Therefore, it is necessary to prepare for the risk of supply interruption, such as the disruption of the import of base metals or interruption of domestic production. In this study, we estimated the quantity required to respond to the risk of import disruption, the quantity to which the market can respond, and the quantity to which the government needs to respond for six base metals (copper, lead, zinc, aluminum, nickel, and tin).

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.185-189
• /
• 2019

관계추출기의 학습을 위해서는 많은 학습 데이터가 필요한데, 사람이 모으게 되면 많은 비용이 필요하여 원격 지도 학습을 이용한 데이터 수집이 많은 연구에서 사용되고 있다. 원격 지도 학습은 지식베이스를 기반으로 학습 데이터를 자동으로 만들어 내는 방식이기에 비용이 거의 들지 않지만, 지식베이스의 질과 양에 영향을 받는다. 본 연구는 원격 지도 학습을 기본으로 관계추출기의 성능을 향상 시키고, 지식베이스를 확장하는 방안으로 반복학습을 제안한다. 실험을 적은 비용으로 빠르게 진행하기 위해 반복학습을 자동화 하는 시스템을 설계하여 실험을 하였고, 이 시스템으로 관계추출기의 성능이 향상 될 수 있는 가능성을 보였으며, 반복학습을 통한 지식베이스의 확장 방안을 제시한다.