• Journal of the Korea Concrete Institute
• /
• v.14 no.6
• /
• pp.1022-1031
• /
• 2002

This paper presents efficient models for bridge structures using CART-ANFIS (classification and regression tree-adaptive neuro fuzzy inference system). A fuzzy decision tree partitions the input space of a data set into mutually exclusive regions, each region is assigned a label, a value, or an action to characterize its data points. Fuzzy decision trees used for classification problems are often called fuzzy classification trees, and each terminal node contains a label that indicates the predicted class of a given feature vector. In the same vein, decision trees used for regression problems are often called fuzzy regression trees, and the terminal node labels may be constants or equations that specify the predicted output value of a given input vector. Note that CART can select relevant inputs and do tree partitioning of the input space, while ANFIS refines the regression and makes it continuous and smooth everywhere. Thus it can be seen that CART and ANFIS are complementary and their combination constitutes a solid approach to fuzzy modeling.

• The Journal of the Korea Contents Association
• /
• v.20 no.10
• /
• pp.167-176
• /
• 2020

The automated fish feed system currently used in Korea supplies a certain amounts of feed to water tanks at a certain time. This automated system can reduce the labor cost of managing aqua farms, but it is very difficult to control intelligently and appropriately the amount of expensive feed that is critical to aqua farm productivity. In this paper, we propose the FIIFF Inference Model( Fuzzy Inference-based Intelligent Fish Feeding Model) that can solves the problems of these existing automatic fish feeding devices and maximizes the efficiency of feed supply while properly maintaining the growth rate of fish in aqua farms. The proposed FIIFF inference model has the advantage of being able to control feed amounts appropriately since it computes the amount of feed using the current water environments and fish activity state of the aqua farms. The result of the feed amount yield experiment with the proposed FIIFF Inference Model represents the effect of saving 14.8% over the eight months of actual feed amount in the aqua farm.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.20 no.43
• /
• pp.37-46
• /
• 1997

제품의 정성적 품질평가에서, 제품의 최종품질을 구성하는 다수의 특성에 대한 만족도가 언어로써 표현되어 소비자의 구매행동이란 의사결정으로 표출되는데, 이러한 주관적 평가에는 평가의 애매함(fuzziness)이 수반되므로 품질의 평가구조를 합리적으로 파악하기 위해서는 애매함의 존재를 고려에 넣지 않으면 안된다. 다수의 품질특성이 계층적(hierarchical)인 구조로 연결되어 최상위 품질특성으로 구성되며, 특성간의 중요도(relative importances)가 계층별로 결정되는 경우, 이들 개개의 특성에 대한 만족도의 평가로부터 어떤 구조적인 관계를 통해 그 제품에 대한 종합평가가 이루어지나, 개개의 특성에 대한 평가가 애매한 이상 최종 결과인 종합적 만족도도 애매한 것으로 된다. 즉, 평가모델의 구조도 평가의 패턴도 퍼지화되므로 이러한 평가에서 퍼지이론의 응용에 따른 효과를 가장 크게 기대할 수 있는 퍼지추론모델을 이용하여 계층간, 품질특성간의 퍼지관계와 특성의 중요도 및 언어변수(linguistic variables)의 형태로 주어지는 입력정보로써 품질구조를 명확히 하고, 패턴인식(pattern recognition)의 개념을 이용하여 평가자의 제품에 대한 평가결과를 언어로써 표현한다.

• Proceedings of the KAIS Fall Conference
• /
• 2010.05a
• /
• pp.89-91
• /
• 2010

본 논문에서는 학습에 임하는 학생과 교수자의 성향을 좀더 객관성 있게 검출, 면학 효과를 증진시키고자, 학습자와 교수자 상호 소통에 필요한 PCSI(Personal Coaching Styles Inventory)검사 모델을, 러프-퍼지 추론 기반에 의하여 평가하는 방법을 제안한다.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.9
• /
• pp.2447-2456
• /
• 1998

Edge detection in the presence of noise is a well-known problem. this pper atempts to implement edge detection algorithms using fuzzy reasoning of fuzzy membership models. It examines an application-motived approach for solving the problem. Our approach is divided into three stages; fitering, segmentation and tracing. Filtering removes the noise from the original image and segmentation determines the edges and deects them. Finally, tracing assembles the edges into the related structure. Proposed method can be used effectively on these procedures by using fuzzy reasoning based on fuzzy models. In is compared with the previous edge detectio algorithms with fvorable results. Simulation results of the research are presented and discussed.

• Journal of the Korean Society for Railway
• /
• v.19 no.2
• /
• pp.145-158
• /
• 2016

This paper introduces a risk graph which is one method for determining the SIL as a measure of the effectiveness of signaling system. The purpose of this research is to make up for the weakness of the qualitative determination, which has input value ambiguity and a boundary problem in the SIL range. The fuzzy input valuable consists of consequence, exposure, avoidance and demand rate. The fuzzy inference produces forty eight fuzzy rule by adapting the calibrated risk graph in the IEC 61511. The Max-min composition is utilized for the fuzzy inference. The result of the fuzzy inference is the fuzzy value. Therefore, using the de-fuzzification method, the result should be converted to a crisp value that can be utilized for real projects. Ultimately, the safety requirement for hazard is identified by proposing a SIL result with a tolerable hazard rate. For the validation the results of the proposed method, the fuzzy risk graph model is compared with the safety analysis of the signaling system in CENELEC SC 9XA WG A10 report.

• Journal of the Korea Society of Computer and Information
• /
• v.3 no.3
• /
• pp.55-61
• /
• 1998

The learning order of teaching materials to be a learning data in CAI is arranged from an easy item to a difficult one A learning in not necessary to be learned arranged this order. Actually the learning is done by the rules of trial and error on the sequences of an arrangement among items. In this papers, the constructed is modelled by the fuzzy inference after leaning the understanding on items by the intelligent CAI through the rile of trial and error of fuzziness. Given the difference of leaning and understanding, the leaning model is quantified by the order relationship among items and by the rules of fuzzy inference. The rule of trial and error of learning is restricted to the treatment of CAL system minimizing the rules of inference.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.235-238
• /
• 2000

대용량의 데이터베이스에서 효율적인 의사결정을 하기 위해서는 불필요한 지식을 제거한 지식베이스의 구축이 필요하다. 사용자의 언어적인 질의에 대해 대용량의 데이터베이스에서 불필요한 규칙을 제거한 최소지식베이스를 구축한다. 또한 불완전한 데이터베이스로부터 규칙들을 일반화한 근사함수에 기반하여 규칙 추출의 중요도를 나타낸다. 그리고 앞에서 생성된 최소지식베이스를 통해 언어적 변수에 대한 퍼지 연산을 수행하여 추론값을 도출할 수 있는 모델을 제안한다.

• Proceedings of KOSOMES biannual meeting
• /
• 2003.05a
• /
• pp.111-116
• /
• 2003

최근에는 전문가의 지식과 경험정보가 데이터베이스로 구축된 전문가 시스템의 정보를 이용하여 처리된 결과를 판단하여 안전하고 효율적인 선박운항이 가능하도록 한 지능형 선박에 관한 연구가 활발하게 진행되고 있다. 본 논문에서는 지능형 선박을 구현하기 위한 연구의 일환으로써, 선박의 조타기를 제어하기 위한 지능형 조타 조작 모델을 구현한다. 지능형 시스템을 구현하기 위해서 자연언어를 사용하는 인간의 학습 방법에 기초한 언어지시기반학습(LIBL)기법을 적용하고. 퍼지이론을 이용하여 승선경력이 풍부한 조타수의 경험을 조사 및 분석하여 그 결과를 바탕으로 퍼지 추론에 의해 타각을 제어하기 위한 퍼지 조타 조작 모델을 구현하여 그 효용성을 살펴보았다.

• The Korean Journal of Food And Nutrition
• /
• v.15 no.3
• /
• pp.228-234
• /
• 2002

In this paper, we proposed a preparation of Jeung- pyun (Korean fermented steamed rice cake with sour taste and spongy texture) using fuzzy theory. Before this preparation was introduced, it thoroughly analyzed the existing data of Jeung-pyun preparation with sensory evaluation and instrumental measurement. It defined a membership auction of Fuzzy set by analyzed three sorts of data on Jeung-pyun. And it established the Fuzzy model using the quantity of materials as input, such as rice, flour, wheat flour and fermentation time, and the sensory test scores as output, such as grain, softness, sourness, chewiness, overall quality, pH value and volume, respectively. We got the results that the Fuzzy model was accord with the conventional method with sensory evaluation. And the validity of this method is shown through the computer simulation of the test data. Therefore, the proposed method by Fuzzy model will apply to make Jeung-pyun without sensory evaluation. This study will contribute to develop standard preparation for korean foods and expert system of preparation using computer system.