• Annual Conference on Human and Language Technology
• /
• 2006.10e
• /
• pp.15-21
• /
• 2006

본 논문에서는 올바른 한국어 생활과 한국어 실력 향상을 위하여, 한국어 어휘에 대한 풍부한 정보를 담고 있는 한국어사전, 사용자 어휘지능망(User-Word Intelligent Network : U-WIN)등의 언어자원을 이용한 자동문제생성기술을 소개하고, 이를 이용한 한국어 어휘학습시스템을 제시한다. 대부분의 학습시스템에서 사용하는 문제 은행식 출제 방식의 문제점을 해소할 수 있는 하나의 방법으로서, 기존의 한국어 어휘문제의 문항을 분석하여 8가지 문제 유형으로 재편성하고, 각 유형별 자동 문제 생성패턴에 따라 언어자원이 가지고 있는 한국어 어휘의 형태적 정보, 의미적 정보를 이용하여 한국어 어휘 문제를 자동 출제하는 한국어 어휘학습시스템을 구현하였다.

• Proceedings of the Korean Society for Information Management Conference
• /
• 1996.08a
• /
• pp.53-56
• /
• 1996

한자로 기술된 용어를 한글로 자동변환하게 되면 생성된 한글이 한글의 음운구조에 맞지 않는 일이 생기게 된다. 이런 문제가 생기는 원인은 해당 한자에 대응하는 정확한 한글을 입력하여 변환하지 않았을 때와 해당 한자의 음이 없거나 한자와의 음운구조가 달라 생기는 때로 대별할 수 있다. 본고에서는 이를 해결하기 위하여 KS C 5601 표준코드(이하 표준코드라 한다)를 기준으로 한자의 다음자를 조사하고 이 다음자가 포함되어 있는 용어를 이용하여 매핑파일을 구축함으로써 바른 한글음을 자동생성할 수 있는 방안을 논한다.

• Journal of Intelligence and Information Systems
• /
• v.8 no.2
• /
• pp.51-69
• /
• 2002

In the supply chain environment on the web, collaborative problem solving and case-based modeling has been getting more important, because it is difficult to cope with diverse problem requirements and inefficient to manage many models as well. Hence, the approach on case-based modeling is required. This paper provides a framework that generates a goal model based on multiple cases, modeling knowledge, and forward chaining and it also develops a search algorithm through sensitivity analysis to reduce the modeling effort.

• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.2
• /
• pp.91-102
• /
• 2012

Computer aided language learning has become popular. But the level of automation of constructing a Korean learning assistant system is not so high because a practical language learning system needs large scale knowledge resources, which is very hard to acquire. In this paper, we propose a Korean learning assistant system that utilizes easily obtainable knowledge resources like a corpus, web documents and a lexicon. Our system has three modules - problem solving, pronunciation marker and writing assistant. Automatic problem generator uses a corpus and a lexicon to make problems with one correct answer and three distracters, then verifies their suitability by utilizing frequency information from web documents. We analyze pronunciation rules for a pronunciation marker and recommend appropriate words and sentences in real-time by using data extracted from a corpus. In experiment, we evaluate 400 automatically generated problems, which show 89.9% problem suitability and 64.9% example suitability.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.1_2
• /
• pp.52-61
• /
• 2006

A crossword puzzle is one of the popular word games around the world in which you work out the answers and write them in the white squares of a pattern of small black and white squares. As the technology of computers develops, some people worked about making and solving crossword puzzle games, which led them to a commercial use. However, almost all of these commercial programs are ones where you do ready-made puzzles with a fixed size because it is very difficult to make puzzles in a certain size, picking up some among a great number of words to fit for them. Furthermore, these programs are only for a very few languages, such as English, French, not for Korean. Accordingly, ore took a look at what should be considered to make an automatic puzzle-generating program for Korean, and in this paper we implemented Korizzle, a system making the puzzles automatically. We introduce the algorithm used for Korizzle and evaluate the its performance.

• Journal of Internet Computing and Services
• /
• v.9 no.2
• /
• pp.89-105
• /
• 2008

So far, most of research on automated test data generation(ATDG) deals with programs without pointers. Recently, few works hove been done on ATDG in the presence of pointers, but they ore path-oriented techniques which require the specification of on entire program path to be tested or a program to be executed. This paper presents a new technique for generating test data even without specifying a program path completely. The presented technique is a static technique which transforms the test data generation problem into a SAT(SATisfiability) problem and makes advantage of SAT solvers for ATDG. For the ends, we transform a program under test into Alloy which is the first-order relational logic and then produce test data via Alloy analyzer.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1996.10a
• /
• pp.306-309
• /
• 1996

본 논문에서는 유전자 알고리즘을 이용하여 퍼지 제어기를 위한 최적 소속함수와 제어 규칙들을 자동으로 생성하는 방법을 제안한다. 제안한 방법은 효과적인 염색체 암호화 방법을 이용하여 소속함수의 표현 해상도가 증가하여도 소속함수의 언어항의 개수를 일정하게 유지하여 제어 규칙을 표현하는 염색체의 길이가 크게 늘어나지 않도록 한다. 또, 소속함수의 언어항의 개수가 서로 다른 염색체에 대해서도 개선된 교배 및 돌연변이 연산자를 이용하여 효과적으로 유전자 연산을 적용할 수 있게 한다. 본 논문에서는 제안된 방법을 퍼지 제어기의 자동 생성 방법의 평가 문제로 널리 이용되는 트럭 후진 주차 문제에 적용하여 성능을 평가한다.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.419-423
• /
• 2019

질문과 그에 대한 근거가 있는 문서를 읽고 정답을 예측하는 기계 독해 연구가 최근 활발하게 연구되고 있다. 기계 독해 문제를 위해 주로 사용되는 방법은 다층의 신경망으로 구성된 딥러닝 모델로 좋은 성능을 위해서는 양질의 대용량 학습 데이터가 필요하다. 그러나 질과 양을 동시에 만족하는 학습 데이터를 구축하는 작업에는 많은 경제적 비용이 소모된다. 이러한 문제를 해결하기 위해, 본 논문에서는 정답 분리 인코더와 복사 메커니즘을 이용한 단답 기반 한국어 질문 자동 생성 모델을 제안한다.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.1
• /
• pp.253-263
• /
• 2017

As needs for knowledge on Korean history and the attention of the people are rapidly increasing, various smartphone applications for learning the history have appeared during recent years. These applications provide multiple choice questions to users through their own problem banks. But, since these questions are selected from the fixed set of problems that are stored previously, the learning efficiency of users is inevitably decreased when they use the applications repeatedly. In this paper, we present a question generation system named K-History which generates multiple choice questions in an automatic way using the database on the history of Korea. In addition, we also describe the development of the application Korean History Infinite Challenge as a learning application for Korean history. To develop K-History, we classify typical types of learning problems through various problems based on Korean history learning materials, proposing algorithms to generate problems according to the types found. Through the developed techniques, various learning systems can reduce the cost for creating questions, while increasing the learning efficiency of users.

• Korean Journal of Construction Engineering and Management
• /
• v.5 no.4 s.20
• /
• pp.107-114
• /
• 2004

This paper represents the application of the 3D CAD Model data for 4D simulation and quantity estimation. These support the effective and practical use of 4D CAD model. By using and manipulating the 3D CAD model information, scheduling and quantity estimation could be developed more quickly and effectively. So the 3D CAD model information is made use of not only drawing a blueprint but also playing an important part of data integration platform. The scheduling module sets up the schedule generation logic that consists of period, priority of element arrangement, and time lag of floor placement. It sorts the working items as a priority of working process. And the quantity estimation module queries the material quantity of the structural elements according to the scheduling conditions. These two modules are developed using the 3D CAD model information and assist the function of 4D CAD model.