• Journal of the Korean Society for information Management
• /
• v.29 no.1
• /
• pp.25-43
• /
• 2012

The ideas and quasi-ideas useful for human's creation were drawn out from documents and webpages with extraction methods used in idea mining, opinion mining, and topic signal mining. The extraction methods comprised (1) decisive cue phrases, (2) cue figures and sounds, (3) contextual signals, and (4) discourse segmentations, They tested on the idea samples, such as thoughts, plans, opinions, writings, figures, sounds, and formulas. Methods (1), (3), and (4) received largely positive evaluation, judging the efficiency of 4 methods by F measure, a mixture of recall and precision ratio. In particular, decisive cue phrase method was effective to search idea and contextual signal method was effective to detect quasi-idea.

• Journal of the Korean Society for Library and Information Science
• /
• v.39 no.3
• /
• pp.139-163
• /
• 2005

Various corpus-based approaches, rhetorical roles of discourse structure, and unifications of similar sentences were applied to construct the automatic Ext/Sums(extracts and summaries). Rhetorical roles of sentences like objective, method, background, result, conclusion, etc. for making elastic Ext/Sums were established and extraction engines according to respective role were prepared. The $90\%$ of Success rate in extracting the important sentences of sample articles was accomplished. Rearranging the selected sentences, it used unification of similar sentences using the cosine coefficient equation, deletion of unnecessary modification and insertion clauses, junction of short sentences, and connection of sentences able to link. They suggest the methods applying rhetorical roles of sentences, meaning and signature of noun and verb in clauses, and cue words and location will be researched to construct the more effective Ext/Sums.

• Journal of the Korean Data and Information Science Society
• /
• v.22 no.3
• /
• pp.467-476
• /
• 2011

We developed the hybrid model coupled with predictive model and business rule model for administration of injury by utilizing medical data of the National Health Insurance in Korea. We performed decision tree analysis using data mining methodology and used SAS Enterprise Miner 4.1. We also investigated under several business rule for benefits (expense paid by insurer) and claims of injury in National Health Insurance Corporation. We can see that the proposed hybrid model provides a quite efficient plausible results.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2014.06a
• /
• pp.89-91
• /
• 2014

전자해도는 IHO의 S-57 ENC Product Specification에 따라 해양의 모든 지형정보를 포함하는데, 해양의 지형정보에는 해안선, 등심선, 수심 그리고 항로표지 등이 포함된다. 전자해도가 포함하는 정보는 수로측량을 통해 제작된 측량원도를 이용하여 제작하는데, 대용량의 측량원도 자료는 매우 조밀하고 양이 방대하여 전자해도에 반영하기 위해서는 수심 정보를 발췌하여야 한다. 국립해양조사원에서는 측량원도의 정보를 발췌하기 위한 방안으로 수심편수규칙을 정하고 있으며, 편수 단계와 편수 후 검증 단계를 거쳐 전자해도의 수심 정보를 제작하고 있다. 그러나 편수와 편수 후 검증은 모두 수작업으로 진행하고 있어, 보다 효율적이고 정확한 방법으로 수심 편수 결과를 검증하기 위한 방안이 필요하다. 본 논문에서는 수심 편수 결과를 규칙에 맞도록 편수되었는지를 평가하는 방안을 연구하였다. 각 편수된 수심을 들로네 삼각분할 기법에 따라 영역을 삼각분할하고, 측량원도의 수심 정보와 삼각분할 된 삼각형과 비교 분석하여 수심이 수심편수규칙을 준수하는지를 비교분석하였다.

• Korean Architects
• /
• no.3 s.228
• /
• pp.62-67
• /
• 1988

본 논문은 미국 콜롬비아 대학의 Klaus Herdeg 교수가 일단의 건축과 학생들을 인솔하고 약 60일간 중국을 방문한 후 그 일원이었던 David Cohn씨가 AIA Journal(1985년)에 기고한 “The Search for National Forms and Modern Techniques”의 내용을 발췌, 번역한 것임.

• the Consultant
• /
• s.20
• /
• pp.87-121
• /
• 2004

본회가 연구·개발한 한국형중소기업컨설팅 표준모델 중 지난 회보를 통해 소개한 “진단기법”과“경영진단 체크리스트 표준모델”에 이어 이번 회보에서는“실행계획”과“통제방법”, “사후관리”등에 관한 컨설팅 방법(Methodology)을 발췌하여 소개 합니다.

• Proceedings of the Korean Society for Information Management Conference
• /
• 2001.08a
• /
• pp.181-184
• /
• 2001

본 논문에서는 적합한 색인어를 자동으로 추출해 내기 위해 잘 알려진 통계적 기법과 구문분석적 기법을 혼용하였다. 적용결과를 검색효율로 나타내지 않고 각 방법에 따라 추출된 단어들을 실증적으로 보여주어 성능에 대한 판단을 유도하였다. 빈도나 품사가 단독으로 사용된 것보다 동시에 적용된 것이 보다 좋은 결과를 가져왔다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.4
• /
• pp.793-799
• /
• 2021

Recently, there has been a growing interest in threat hunting presented to overcome the limitations of existing security solutions. Threat hunting is generally recognized as a technique for identifying and eliminating threats that exit inside the system. But, the definition is not clear, so there is confusion in terms with penetration testing, intrusion detection, and incident analysis. Therefore, in this paper, compare and analyze the definitions of threat hunting extracted from reports and papers to clarify their implications and compare with defense techniques.

• Journal of KIISE:Databases
• /
• v.31 no.2
• /
• pp.99-107
• /
• 2004

Education and training are expected to change dramatically due to the combined impact of the Internet, database, and multimedia technologies However, the distance learning is often impeded by the lack of effective tools and system to manage and retrieve the lecture contents effectively. This paper introduces a prototype system called COVA that enables remote users to access specific parts of interest by contents from a large lecture database. COVA includes several novel techniques to achieve the content-based lecture retrieval in distance teaming: (1) The XML-based semistructured model to represent lecture contents; (2) The technique to build structural summaries, i.e., schemas, of XML lecture databases; (3) Index structures to speed up the search to find appropriate lecture contents.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.3
• /
• pp.139-147
• /
• 2009

The memory based reasoning just stores in the memory in the form of the training pattern of the representative pattern. And it classifies through the distance calculation with the test pattern. Because it uses the techniques which stores the training pattern whole in the memory or in which it replaces training patterns with the representative pattern. Due to this, the memory in which it is a lot for the other machine learning techniques is required. And as the moreover stored training pattern increases, the time required for a classification is very much required. In this paper, We propose the EAS(Evaluation And Selection) algorithm in order to minimize memory usage and to improve classification performance. After partitioning the training space, this evaluates each partitioned space as MDL and PM method. The partitioned space in which the evaluation result is most excellent makes into the representative pattern. Remainder partitioned spaces again partitions and repeat the evaluation. We verify the performance of Proposed algorithm using benchmark data sets from UCI Machine Learning Repository.