• Communications for Statistical Applications and Methods
• /
• v.13 no.1
• /
• pp.151-165
• /
• 2006

It is important to estimate the true misclassification rate of a given classifier when an independent set of test data is not available. Cross-validation and bootstrap are two possible approaches in this case. In related literature bootstrap estimators of the true misclassification rate were asserted to have better performance for small samples than cross-validation estimators. We compare the two estimators empirically when the classification rule is so adaptive to training data that its apparent misclassification rate is close to zero. We confirm that bootstrap estimators have better performance for small samples because of small variance, and we have found a new fact that their bias tends to be significant even for moderate to large samples, in which case cross-validation estimators have better performance with less computation.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.1350-1353
• /
• 1993

A fuzzy control system typically requires“tuning,”or adjuctment of the parameters defining its linguistic variables. Automating this process amounts to applying a second“metacontrol”layer to drive the controller and plant to desired performance levels. Current methods of automated tuning rely on a single crisp numeric functional to evaluate control system performance. A generalization of Box's complex algorithm allows more realistic tuning based on lexicographic aggregation of multiple ordinal scales of performance, such as effectiveness and efficiency. The method is presented and illustrated using a simple inverted pendulum control system.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2911-2913
• /
• 1999

This paper suggest an optimal identification method to complex and nonlinear system modeling that is based on Fuzzy-Neural Network(FNN). The FNN modeling implements parameter identification using HCM algorithm and optimal identification algorithm structure combined with two types of optimization theories for nonlinear systems, we use a HCM Clustering Algorithm to find initial parameters of membership function. The parameters such as parameters of membership functions, learning rates and momentum coefficients are adjusted using optimal identification algorithm. The proposed optimal identification algorithm is carried out using both a genetic algorithm and the improved complex method. Also, an aggregate objective function(performance index) with weighted value is proposed to achieve a sound balance between approximation and generalization abilities of the model. To evaluate the performance of the proposed model, we use the time series data for gas furnace, the data of sewage treatment process and traffic route choice process.

• The Journal of Society for e-Business Studies
• /
• v.25 no.3
• /
• pp.1-13
• /
• 2020

Machine learning refers to a model generation technique that can solve specific problems from the generalization process for given data. In order to generate a high performance model, high quality training data and learning algorithms for generalization process should be prepared. As one way of improving the performance of model to be learned, the Ensemble technique generates multiple models rather than a single model, which includes bagging, boosting, and stacking learning techniques. This paper proposes a new Ensemble technique with multiple stacking that outperforms the conventional stacking technique. The learning structure of multiple stacking ensemble technique is similar to the structure of deep learning, in which each layer is composed of a combination of stacking models, and the number of layers get increased so as to minimize the misclassification rate of each layer. Through experiments using four types of datasets, we have showed that the proposed method outperforms the exiting ones.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.50 no.7
• /
• pp.339-349
• /
• 2001

This paper suggests an optimal identification method for complex and nonlinear system modeling that is based on Fuzzy-Neural Networks(FNN). The proposed Hybrid Identification Algorithm is based on Yamakawa's FNN and uses the simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rule. In this paper, the FNN modeling implements parameter identification using HCM algorithm and hybrid structure combined with two types of optimization theories for nonlinear systems. We use a HCM(Hard C-Means) clustering algorithm to find initial apexes of membership function. The parameters such as apexes of membership functions, learning rates, and momentum coefficients are adjusted using hybrid algorithm. The proposed hybrid identification algorithm is carried out using both a genetic algorithm and the improved complex method. Also, an aggregated objective function(performance index) with weighting factor is introduced to achieve a sound balance between approximation and generalization abilities of the model. According to the selection and adjustment of a weighting factor of an aggregate objective function which depends on the number of data and a certain degree of nonlinearity(distribution of I/O data), we show that it is available and effective to design an optimal FNN model structure with mutual balance and dependency between approximation and generalization abilities. To evaluate the performance of the proposed model, we use the time series data for gas furnace, the data of sewage treatment process and traffic route choice process.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.10
• /
• pp.73-81
• /
• 2015

The paper proposes a prototype selection method and evaluates the generalization performance of standard algorithms and prototype based classification learning. The proposed prototype classifier defines multidimensional spheres with variable radii within class areas and generates a small set of training data. The nearest-neighbor classifier uses the new training set for predicting the class of test data. By decomposing bias and variance of the mean expected error value, we compare the generalization errors of k-nearest neighbor, Bayesian classifier, prototype selection using fixed radius and the proposed prototype selection method. In experiments, the bias-variance changing trends of the proposed prototype classifier are similar to those of nearest neighbor classifiers with all training data and the prototype selection rates are under 27.0% on average.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.3
• /
• pp.55-65
• /
• 2004

When a neural network is used to solve a given problem it is necessary to match the complexity of the network to that of the problem because the complexity of the network significantly affects its learning capability and generalization performance. Thus, it is desirable to have an algorithm that can find appropriate network structures in a self-organizing way. This paper proposes algorithms which automatically organize feed forward multi-layer neural networks with sigmoid hidden neurons for given problems. Using both constructive procedures and pruning procedures, the proposed algorithms try to find the near optimal network, which is compact and shows good generalization performance. The performances of the proposed algorithms are tested on four function regression problems. The results demonstrate that our algorithms successfully generate near-optimal networks in comparison with the previous method and the neural networks of fixed topology.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.12
• /
• pp.2946-2952
• /
• 2014

Among the Example based Super Resolution(SR) techniques, Neighbor embedding(NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the poor generalization of NE decreases the performance of such algorithm. The sizes of local training sets are always too small to improve the performance of NE. We propose the Learning Sparse-Neighbor Image Representation baesd on SVR having an excellent generalization ability to solve this problem. Given a low resolution image, we first use bicubic interpolation to synthesize its high resolution version. We extract the patches from this synthesized image and determine whether each patch corresponds to regions with high or low spatial frequencies. After the weight of each patch is obtained by our method, we used to learn separate SVR models. Finally, we update the pixel values using the previously learned SVRs. Through experimental results, we quantitatively and qualitatively confirm the improved results of the proposed algorithm when comparing with conventional interpolation methods and NE.

• Journal of KIISE:Databases
• /
• v.28 no.3
• /
• pp.357-372
• /
• 2001

In this paper, we present the concept of generalization in constructing windows for subsequence matching and propose a new subsequence matching method. GeneralMatch, based on the generalization. The earlier work of Faloutsos et al.(FRM in short) causes a lot of false alarms due to lack of the point-filtering effect. DualMatch, which has been proposed by the authors, improves performance significantly over FRM by exploiting the point filtering effect, but it has the problem of having a smaller maximum window size (half that FRM) given the minimum query length. GeneralMatch, an improvement of DualMatch, offers advantages of both methods: it can use large windows like FRM and, at the same time, can exploit the point-filtering effect like DualMatch. GeneralMatch divides data sequences into J-sliding windows (generalized sliding windows) and the query sequence into J-disjoint windows (generalized disjoint windows). We formally prove that our GeneralMatch is correct, i.e., it incurs no false dismissal. We also prove that, given the minimum query length, there is a maximum bound of the window size to guarantee correctness of GeneralMatch. We then propose a method of determining the value of J that minimizes the number of page accesses, Experimental results for real stock data show that, for low selectivities ($10^{-6}~10^{-4}$), GeneralMatch improves performance by 114% over DualMatch and by 998% iver FRM on the average; for high selectivities ($10^{-6}~10^{-4}$), by 46% over DualMatch and by 65% over FRM on the average.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.6
• /
• pp.836-844
• /
• 2008

This paper presents an approach to order reduction of linear parameter varying controller for polytopic model. Feasible solutions which satisfy relevant linear matrix inequalities for constructing full-order parameter varying controller evaluated at each polytopic vertices are first found. Next, sufficient conditions are derived for the existence of a right coprime factorization of parameter varying controller. Furthermore, a singular perturbation approximation for time invariant systems is generalized to reduce full-order parameter varying controller via parameter varying right coprime factorization. This generalization is based on solutions of the parameter varying Lyapunov inequalities. The closed loop performance caused by using the reduced order controller is developed. To examine the performance of the reduced-order parameter varying controller, the proposed method is applied to reduce vibration of flexible structures having the transverse-torsional coupled vibration modes.