• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2003년도 Proceeding
• /
• pp.178-184
• /
• 2003

Case-based reasoning (CBR) often shows significant promise for improving effectiveness of complex and unstructured decision making. Consequently, it has been applied to various problem-solving areas including manufacturing, finance and marketing. However, the design of appropriate case indexing and retrieval mechanisms to improve the performance of CBR is still challenging issue. Most of previous studies to improve the effectiveness for CBR have focused on the similarity function or optimization of case features and their weights. However, according to some of prior researches, finding the optimal k parameter for k-nearest neighbor (k-NN) is also crucial to improve the performance of CBR system. Nonetheless, there have been few attempts which have tried to optimize the number of neighbors, especially using artificial intelligence (AI) techniques. In this study, we introduce a genetic algorithm (GA) to optimize the number of neighbors to combine. This study applies the new model to the real-world case provided by an online shopping mall in Korea. Experimental results show that a GA-optimized k-NN approach outperforms other AI techniques for purchasing behavior forecasting.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2007년도 한국지능정보시스템학회
• /
• pp.389-398
• /
• 2007

Case-based reasoning (CBR) is one of the most popular prediction techniques for medical diagnosis because it is easy to apply, has no possibility of overfitting, and provides a good explanation for the output. However, it has a critical limitation - its prediction performance is generally lower than other artificial intelligence techniques like artificial neural networks (ANNs). In order to obtain accurate results from CBR, effective retrieval and matching of useful prior cases for the problem is essential, but it is still a controversial issue to design a good matching and retrieval mechanism for CBR systems. In this study, we propose a novel approach to enhance the prediction performance of CBR. Our suggestion is the simultaneous optimization of feature weights, instance selection, and the number of neighbors that combine using genetic algorithms (GAs). Our model improves the prediction performance in three ways - (1) measuring similarity between cases more accurately by considering relative importance of each feature, (2) eliminating redundant or erroneous reference cases, and (3) combining several similar cases represent significant patterns. To validate the usefulness of our model, this study applied it to a real-world case for evaluating cytological features derived directly from a digital scan of breast fine needle aspirate (FNA) slides. Experimental results showed that the prediction accuracy of conventional CBR may be improved significantly by using our model. We also found that our proposed model outperformed all the other optimized models for CBR using GA.

• 한국식품영양과학회지
• /
• 제14권3호
• /
• pp.305-313
• /
• 1985

There is a trend that the total number of cancer cases is steadily increasing as the population grows. It has been estimated that 85% of the cancer rate in the U.S. is attributed to environmental factors. Among the environmental factors, diet and nutrition appear to be related to the largest number of human cancers. Diet and nutrition might be related to cancer by several mechanisms. Food may contain a direct carcinogen or precursors that become carcinogens by spontanous reactions, or by host metabolism, or through the actions of microbial flora. Chemicals that cause cancers generally have reactive electrophilic centers which can combine with electron-rich atoms in nucleic acids and cause cancers by changing the genetic activity of the cells. A variety of factors in foods might be involved in the etiology of carcinogenesis. Chemicals in food that cause cancers include carcinogens of plants and animal origin and also those in drinking water. Other then these, fungal metabolites alcohol, asbestos, heavy metals, pesticides, and food additives might be included as food carcinogenesis. The method of cooking foods also might contribute to carcinogenesis. Some chemicals in foods act as promoters in carcinogenesis. Prevention of cancers by dietary practises have received much interest. Consumption of certain vegetables or cellulose can reduce carcinogenic activity of several compounds. A variety of antioxidants or micronutrients may be effective anticarciongens. Glutathione in the soluble fraction of the cells, is a major defense against oxidative and alkylating carcinogens. Recently anticarcinogenic activity of chlorophyll was demonstrated. Daily consumption of milk appears to effectively reduce stomach cancer.

• 한국초등수학교육학회지
• /
• 제17권1호
• /
• pp.129-142
• /
• 2013

이 연구의 목적은 2007년부터 2011년까지 일본수학교육학회에서 발행하는 초등수학교육에 관한 학술지 '산수교육'을 통하여 학술지의 구성, 게재된 초등수학 관련 논문의 개요를 파악하여 정통적 초등수학교육 학술지인 한국초등수학교육학회지에 시사점을 주기 위함이다.

• 대한수학회지
• /
• 제53권3호
• /
• pp.645-689
• /
• 2016

This article gives upper bounds on the number of Fourier-Jacobi coefficients that determine a paramodular cusp form in degree two. The level N of the paramodular group is completely general throughout. Additionally, spaces of Jacobi cusp forms are spanned by using the theory of theta blocks due to Gritsenko, Skoruppa and Zagier. We combine these two techniques to rigorously compute spaces of paramodular cusp forms and to verify the Paramodular Conjecture of Brumer and Kramer in many cases of low level. The proofs rely on a detailed description of the zero dimensional cusps for the subgroup of integral elements in each paramodular group.

• Korean Journal of Mathematics
• /
• 제30권3호
• /
• pp.525-532
• /
• 2022

Let 𝓟_n denote the space of all complex polynomials $P(z)=\sum\limits_{j=0}^{n}{\alpha}_jz^j$ of degree n. Let P ∈ 𝓟_n, for any complex number α, D_αP(z) = nP(z) + (α - z)P'(z), denote the polar derivative of the polynomial P(z) with respect to α and B_n denote a family of operators that maps 𝓟_n into itself. In this paper, we combine the operators B and D_α and establish certain operator preserving inequalities concerning polynomials, from which a variety of interesting results can be obtained as special cases.

• Asia pacific journal of information systems
• /
• 제16권1호
• /
• pp.71-84
• /
• 2006

Case-based reasoning (CBR) is a reasoning technique that reuses past cases to find a solution to the new problem. It often shows significant promise for improving effectiveness of complex and unstructured decision making. It has been applied to various problem-solving areas including manufacturing, finance and marketing for the reason. However, the design of appropriate case indexing and retrieval mechanisms to improve the performance of CBR is still a challenging issue. Most of the previous studies on CBR have focused on the similarity function or optimization of case features and their weights. According to some of the prior research, however, finding the optimal k parameter for the k-nearest neighbor (k-NN) is also crucial for improving the performance of the CBR system. In spite of the fact, there have been few attempts to optimize the number of neighbors, especially using artificial intelligence (AI) techniques. In this study, we introduce a genetic algorithm (GA) to optimize the number of neighbors to combine. This study applies the novel approach to Korean stock market. Experimental results show that the GA-optimized k-NN approach outperforms other AI techniques for stock market prediction.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제2권8호
• /
• pp.535-542
• /
• 2013

사례기반추론(CBR)은 많은 장점으로 인해 지금까지 의료진단, 생산계획, 고객분류 등 다양한 분야의 의사결정 지원에 적용되어 왔다. 그러나, 효과적인 CBR 시스템을 설계, 구축하기 위해서는 연구자가 직관적으로 설정해야 할 많은 설계요소들이 존재한다. 본 연구에서는 이러한 CBR의 여러 설계요소들 중 사례 검색 단계에서 결합할 이웃 사례들을 보다 효과적으로 선정할 수 있는 새로운 모형을 제시한다. 기존 연구에서는 결합할 이웃 사례를 선정하는 방법으로 사전에 정해진 이웃사례의 수(k-NN의 k)를 적용하든가, 혹은 최대 유사도의 상대적 비율을 임계치로 사용하는 방식을 적용해 왔다. 하지만, 본 연구에서는 결합할 유사사례를 선택하는 새로운 기준으로 0에서 1사이의 값을 갖는 절대적 유사 임계치를 사용할 것을 제안한다. 이 경우, 임계치 값이 과도하게 작아지게 되면, 예측결과의 생성이 잘 이루어지지 않을 수 있는 문제가 발생할 수 있다. 이에, 전체 학습사례들 중에서 예측결과가 생성된 사례의 비중을 커버리지(coverage)로 정의하고, 이를 유사 임계치 최적화 시 제약조건으로 설정함으로서, 사용자가 원하는 수준의 커버리지는 유지한 상태에서 가장 효과적인 유사 사례를 찾아 추론할 수 있도록 모형을 설계하였다. 제안 모형의 유용성을 검증하기 위해, 본 연구에서는 이 모형을 실존하는 국내 한 온라인 쇼핑몰의 표적 마케팅 사례에 적용하였다. 그 결과, 제안 모형이 CBR의 예측 성과를 유의미하게 개선시킬 수 있음을 확인할 수 있었다.

• 품질경영학회지
• /
• 제51권4호
• /
• pp.735-749
• /
• 2023

Purpose: The purpose of this study is to derive solutions and prevent similar cases from occurring by analyzing the causes of cracks found in temperature cycling tests of rocket motor. Methods: By combining the results of the current state confirmation test, non-destructive test, domestic and foreign rocket motor comparison test, cutting test, and adhesion test according to the number of times to apply mold release agent, a Cause and Effect Diagram analysis was performed to derive the cause of cracks. Results: Through this study, 26 factors that could cause cracking in rocket motors during temperature cycling tests were identified. Through various additional test results, a total of five causes were identified, including chemical and structural design of the joint between the propellant and stress relief insert, omission of procedure in the manufacturing procedures, natural aging due to temperature, and load accumulation due to temperature changes. The fundamental cause was confirmed to be insufficient consideration of the release properties of the propellant and stress relief insert. Conclusion: During the design process, it was confirmed that this could be solved by structurally or chemically designing the insert so that it does not combine with the propellant, or by applying a mold release agent during the manufacturing process.

• Structural Engineering and Mechanics
• /
• 제56권2호
• /
• pp.169-188
• /
• 2015

There are two types of nonlinear analysis methods for building frameworks depending on the method of modeling the plastification of members including lumped plasticity and distributed plasticity. The lumped plasticity method assumes that plasticity is concentrated at a zero-length plastic hinge section at the ends of the elements. The distributed plasticity method discretizes the structural members into many line segments, and further subdivides the cross-section of each segment into a number of finite elements. When a reinforced concrete member experiences inelastic deformations, cracks tend to spread form the joint interface resulting in a curvature distribution. The program IDARC includes a spread plasticity formulation to capture the variation of the section flexibility, and combine them to determine the element stiffness matrix. In this formulation, the flexibility distribution in the structural elements is assumed to be the linear. The main objective of this study is to evaluate the accuracy of linear flexibility distribution assumed in the spread inelasticity model. For this purpose, nonlinear analysis of two reinforced concrete frames is carried out and the linear flexibility models used in the elements are compared with the real ones. It is shown that the linear flexibility distribution is incorrect assumption in cases of significant gravity load effects and can be lead to incorrect nonlinear responses in some situations.