• 한국지능시스템학회:학술대회논문집
• /
• 한국지능시스템학회 2007년도 추계학술대회 학술발표 논문집
• /
• pp.366-369
• /
• 2007

In this paper we derived fuzzy entropy that is based on similarity measure. Similarity measure represents the degree of similarity between two informations, those informations characteristics are not important. First we construct similarity measure between two informations, and derived entropy functions with obtained similarity measure. Obtained entropy is verified with proof. With the help of one-to-one similarity is also obtained through distance measure, this similarity measure is also proved in our paper.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권4호
• /
• pp.367-371
• /
• 2005

The similarity measure is derived using fuzzy entropy and distance measure. By the elations of fuzzy entropy, distance measure, and similarity measure, we first obtain the fuzzy entropy. And with both fuzzy entropy and distance measure, similarity measure is obtained., We verify that the proposed measure become the similarity measure.

• 한국지능시스템학회논문지
• /
• 제18권2호
• /
• pp.257-261
• /
• 2008

In this Paper we derived fuzzy entropy that is based on similarity measure. Similarity measure represents the degree of similarity between two informations, those informations characteristics are not important. First we construct similarity measure between two informations, and derived entropy functions with obtained similarity measure. Obtained entropy is verified with proof. With the help of one-to-one similarity is also obtained through distance measure, this similarity measure is also proved in our paper.

• 한국융합학회논문지
• /
• 제5권4호
• /
• pp.155-161
• /
• 2014

We survey the relation of fuzzy entropy measure and similarity measure. Each measure represents features of data uncertainty and certainty between comparative data group. With the help of one-to-one correspondence characteristics, distance measure and similarity measure have been expressed by the complementary characteristics. We construct similarity measure using distance measure, and verification of usefulness is proved. Furthermore analysis of similarity measure from fuzzy entropy measure is also discussed.

• 한국융합학회논문지
• /
• 제4권3호
• /
• pp.35-41
• /
• 2013

Grouping problem with similarities between DNA sequences are studied. The similaritymeasure and the distance measure showed the complementary characteristics. Distance measure can be obtained by complementing similarity measure, and vice versa. Similarity measure is derived and proved. Usefulness of the proposed similarity measure is applied to grouping problem of 25 cockroach DNA sequences. By calculation of DNA similarity, 25 cockroaches are clustered by four groups, and the results are compared with the previous neighbor-joining method.

• 한국지능시스템학회논문지
• /
• 제18권2호
• /
• pp.200-205
• /
• 2008

데이터 분석을 위하여 데이터의 불확실성에 대한 측도로서 퍼지 집합에 대한 엔트로피를 소개하였고, 또한 데이터간의 유사도를 나타내는 유사측도를 구성하였다. 퍼지 소속 함수간의 유사측도는 거리측도를 이용하여 구성하였고, 제안한 유사측도를 증명을 통하여 확인하였다. 제안한 유사측도의 유용성을 확인하기 위하여 신뢰성 있는 데이터추출 예제에 적용하였다. 적용결과를 퍼지 엔트로피와 통계적 지식을 통하여 얻어진 이전의 결과와 비교하였다.

• 대한전자공학회논문지
• /
• 제20권6호
• /
• pp.23-28
• /
• 1983

콘투어 맷칭을 위한 1차 유사도와 2차 유사도가 제안되었다. 그것들은 최소자승법에 의한 것이다. 특히 2차 유사도는 콘투어의 불완전함 혹은 어파인 변환 혹은 이들 특성의 결합같은 왜곡된 변화에 양호한 신뢰도를 가지고 있다는 사실을 비행기 기종의 판별과 인식하는 실험을 통하여 증명하였다. 또한 맷칭 성능에 있어서 2차 유사도가 1차 유사도뿐만 아니라 기존의 맷칭기법들보다 우수함을 보였다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국지능시스템학회 2007년도 추계학술대회 학술발표 논문집
• /
• pp.199-202
• /
• 2007

The similarity measure is constructed for non-convex fuzzy membership function using well known Hamming distance measure. Comparison with convex fuzzy membership function is carried out, furthermore characteristic analysis for non-convex function are also illustrated. Proposed similarity measure is proved and the usefulness is verified through example. In example, usefulness of proposed similarity is pointed out.

• 한국통신학회논문지
• /
• 제30권9C호
• /
• pp.854-859
• /
• 2005

모호함의 측도를 위하여 퍼지 엔트로피와 거리측도 그리고 유사측도와의 관계를 이용하여 새로운 퍼지 측도를 제안하였다. 제안된 퍼지 엔트로피는 거리측도를 이용하여 구성된다. 거리측도는 일반적으로 사용되는 해밍 거리를 이용하였다. 또한 집합사이의 유사성을 측정하기 위한 유사측도를 거리 측도를 이용하여 구성하였고, 제안한 퍼지 엔트로피와 유사측도를 증명을 통하여 타당성을 확인하였다.

• ETRI Journal
• /
• 제35권2호
• /
• pp.311-320
• /
• 2013

In most spectral clustering approaches, the Gaussian kernel-based similarity measure is used to construct the affinity matrix. However, such a similarity measure does not work well on a dataset with a nonlinear and elongated structure. In this paper, we present a new similarity measure to deal with the nonlinearity issue. The maximum flow between data points is computed as the new similarity, which can satisfy the requirement for similarity in the clustering method. Additionally, the new similarity carries the global and local relations between data. We apply it to spectral clustering and compare the proposed similarity measure with other state-of-the-art methods on both synthetic and real-world data. The experiment results show the superiority of the new similarity: 1) The max-flow-based similarity measure can significantly improve the performance of spectral clustering; 2) It is robust and not sensitive to the parameters.