• 한국음향학회지
• /
• 제33권3호
• /
• pp.216-224
• /
• 2014

본 논문은 Gaussian Mixture Model (GMM) 기반의 화자식별에서 급격한 계산 복잡도 감소를 위한 새로운 방법을 제안한다. 일반적으로 GMM 기반의 화자식별 시스템은 테스트 발성의 길이, 등록 화자의 수, GMM의 크기 등 크게 세 가지 요인에 비례하는 많은 계산 복잡도를 가진다. 이러한 점은 화자식별 시스템이 다양한 응용분야에 적용되는 것을 막는 큰 요인이기에 계산 복잡도와 식별 성능 사이의 trade-off 관계는 실제 적용을 위해 가장 중요한 고려요소이다. 식별 성능을 거의 그대로 유지하면서 최대한 계산 복잡도를 감소시키기 위해 우리는 Universal Background Model (UBM) 클러스터링 접근 방법을 제시하고, 또한 이 방법은 실시간 구조의 화자식별에 적용할 수 있다는 것을 보여준다. 제안한 방법의 실험을 통해 미미한 정도의 식별 성능 저하에서 speed-up factor 6의 결과를 얻을 수 있었다.

• 터널과지하공간
• /
• 제15권1호
• /
• pp.47-54
• /
• 2005

암반 불연속면의 조사 및 분석 과정에서 거쳐야할 필수적인 단계 중 하나는 방대한 불연속면 자료로부터 군을 판별하는 것이다. 불연속면 군 분류는 암반분류, 키블록 해석. 개별요소해석 및 불연속연결망 생성과 같은 암반공학적 업무에 있어서 필수적이다. 일반적으로 등고선도를 이용한 수작업 군 분류가 적용되었으나 이 방법은 수작업에 의존한 주관적인 결과를 제공한다는 단점이 있다. 본 연구에서는 유전자알고리즘을 이용한 불연속면 군 분석기법을 도입하였으며 방향성 자료에 적용하기 위해 기본적인 유전자알고리즘을 변경하였다. 최종적으로 이러한 이론을 적용한 FORTRAN 프로그램 GAC를 개발하였으며 두 가지 형태의 불연속면 자료의 군 분석에 적용하였다. 적용 결과 GAC를 적용한 군 분류는 빠르고 효율적인 군 분석방법임을 확인하였으며 최적의 불연속면 군 수를 결정하는 데 있어서 분산에 근거한 적합도 함수가 Davis-Bouldin 지수에 근거한 적합도 함수보다 효율적인 것으로 나타났다.

• 전기학회논문지
• /
• 제60권1호
• /
• pp.184-192
• /
• 2011

In this study, we introduce a identification methodology for FCM-based fuzzy model. The two underlying design mechanisms of such networks involve Fuzzy C-Means (FCM) clustering method and Particle Swarm Optimization(PSO). The proposed algorithm is based on FCM clustering method for efficient processing of data and the optimization of model was carried out using PSO. The premise part of fuzzy rules does not construct as any fixed membership functions such as triangular, gaussian, ellipsoidal because we build up the premise part of fuzzy rules using FCM. As a result, the proposed model can lead to the compact architecture of network. In this study, as the consequence part of fuzzy rules, we are able to use four types of polynomials such as simplified, linear, quadratic, modified quadratic. In addition, a Weighted Least Square Estimation to estimate the coefficients of polynomials, which are the consequent parts of fuzzy model, can decouple each fuzzy rule from the other fuzzy rules. Therefore, a local learning capability and an interpretability of the proposed fuzzy model are improved. Also, the parameters of the proposed fuzzy model such as a fuzzification coefficient of FCM clustering, the number of clusters of FCM clustering, and the polynomial type of the consequent part of fuzzy rules are adjusted using PSO. The proposed model is illustrated with the use of Automobile Miles per Gallon(MPG) and Boston housing called Machine Learning dataset. A comparative analysis reveals that the proposed FCM-based fuzzy model exhibits higher accuracy and superb predictive capability in comparison to some previous models available in the literature.

• 대한전자공학회논문지TE
• /
• 제39권4호
• /
• pp.414-422
• /
• 2002

본 논문에서는 적응 제어 문제를 다루기 위해 CFCM 클러스터링과 퍼지 균등화 기법을 이용하여 새로운 적응 뉴로-퍼지 제어기를 설계하고자 한다. 먼저 오프라인에서 CFCM은 입력데이터의 성질과 출력 패턴의 성질까지도 고려한 퍼지 클러스터링 기법으로 적응 뉴로-퍼지 제어기의 구조동정을 수행한다. 파라미터 동정은 역전과 알고리즘과 RLSE(Recursive Least Square Estimate)을 이용한 하이브리드 학습을 수행한다. 온라인 학습에서는 시변특성으로 인해 전제부 및 결론부 파라미터를 실시간으로 계산된다. 시뮬레이션으로 온 라인 적응 뉴로-퍼지 제어 시스템의 성능을 입증하기 위해 목욕물 온도제어 시스템에 대해 다루고 전형적인 퍼지 제어기에 비해 오프 라인과 온 라인 설계 모두 좋은 성능을 보이고자 한다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.207-209
• /
• 2007

In this paper, we introduce a genetic optimization of fuzzy set-fuzzy model using successive tuning method to carry out the model identification of complex and nonlinear systems. To identity we use genetic alrogithrt1 (GA) sand C-Means clustering. GA is used for determination the number of input, the seleced input variables, the number of membership function, and the conclusion inference type. Information Granules (IG) with the aid of C-Means clustering algorithm help determine the initial paramters of fuzzy model such as the initial apexes of the, membership functions in the premise part and the initial values of polyminial functions in the consequence part of the fuzzy rules. The overall design arises as a hybrid structural and parametric optimization. Genetic algorithms and C-Means clustering are used to generate the structurally as well as parametrically optimized fuzzy model. To identify the structure and estimate parameters of the fuzzy model we introduce the successive tuning method with variant generation-based evolution by means of GA. Numerical example is included to evaluate the performance of the proposed model.

• 한국정밀공학회지
• /
• 제28권4호
• /
• pp.391-397
• /
• 2011

Control of interior permanent magnet (IPMSM) is difficult because its nonlinearity and parameter uncertainty. In this paper, a fuzzy c-regression models clustering algorithm which is based on T-S fuzzy is used to model IPMSM with a series linear model and weight them by memberships. Lagrangian of constrained function is built for calculating clustering centers where training output data are considered. Based on these clustering centers, least square method is applied for T-S fuzzy linear model parameters. As a result, IPMSM can be modeled as T-S fuzzy model for T-S fuzzy control of them.

• Computers and Concrete
• /
• 제6권2호
• /
• pp.155-165
• /
• 2009

Splitting tensile strength (STS) of high-performance concrete (HPC) is one of the important mechanical properties for structural design. This property is related to compressive strength (CS), water/binder (W/B) ratio and concrete age. This paper presents a clustering-based fuzzy model for the prediction of STS based on the CS and (W/B) at a fixed age (28 days). The data driven fuzzy model consists of three main steps: fuzzy clustering, inference system, and prediction. The system can be analyzed directly by the model from measured data. The performance evaluations showed that the fuzzy model is more accurate than the other prediction models concerned.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2005년도 학술대회 논문집 정보 및 제어부문
• /
• pp.143-145
• /
• 2005

In this paper, an empirical study result on pattern estimation method is devoted to reveal underlying data patterns with a relatively reduced computational cost. Presented method performs crisp type clustering with given n number of data samples by means of the sequential agglomerative hierarchical nested model (SAHN). Conventional SAHN based clustering requires large computation time in the initial step of algorithm. To deal with this concern, we modified overall process with a partial approach. In the beginning of this method, we divide given data set to several sub groups with uniform sampling and then each divided sub data group is applied to SAHN based method. The advantage of this method reduces computation time of original process and gives similar results. Proposed is applied to several test data set and simulation result with conceptual analysis is presented.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.370-370
• /
• 2000

In this paper, we design a Multi-Fuzzy model by means of HCM clustering and genetic algorithms for a nonlinear system. In order to determine structure of the proposed Multi-Fuzzy model, HCM clustering method is used. The parameters of membership function of the Multi-Fuzzy ate identified by genetic algorithms. A aggregate performance index with a weighting factor is used to achieve a sound balance between approximation and generalization abilities of the model. We use simplified inference and linear inference as inference method of the proposed Multi-Fuzzy mode] and the standard least square method for estimating consequence parameters of the Multi-Fuzzy. Finally, we use some of numerical data to evaluate the proposed Multi-Fuzzy model and discuss about the usefulness.

• Communications for Statistical Applications and Methods
• /
• 제11권3호
• /
• pp.485-494
• /
• 2004

An algorithm is proposed to identify multiple outliers in linear regression. It is based on the clustering of residuals from the least median of squares estimation. A cut-height criterion for the hierarchical cluster tree is suggested, which yields the optimal clustering of the regression outliers. Comparisons of the effectiveness of the procedures are performed on the basis of the classic data and artificial data sets, and it is shown that the proposed algorithm is superior to the one that is based on the least squares estimation. In particular, the algorithm deals very well with the masking and swamping effects while the other does not.