• Communications for Statistical Applications and Methods
• /
• v.12 no.3
• /
• pp.659-672
• /
• 2005

This paper evaluates discretization of continuous variables to select relevant variables for supervised learning using mutual information. Three discretization methods, MDL, Histogram and 4-Intervals are considered. The process of discretization and variable subset selection is evaluated according to the classification accuracies with the 6 real data sets of UCI databases. Results show that 4-Interval discretization method based on quantiles, is robust and efficient for variable selection process. We also visually evaluate the appropriateness of the selected subset of variables.

• Proceedings of the KIEE Conference
• /
• 2007.04a
• /
• pp.152-154
• /
• 2007

Anew discretization method for the input-driven nonlinear continuous-time system with time delay is proposed. It is based on the combination of Taylor series expansion and first-order hold assumption. The mathematical structure of the new discretization scheme is explored. The performance of the proposed discretization procedure is evaluated by case studies. The results demonstrate that the proposed discretization scheme can assure the system requirements even though under a large sampling period. A comparison between first order hold and zero-order hold is simulated also.

• Journal of Korean Institute of Industrial Engineers
• /
• v.27 no.4
• /
• pp.394-405
• /
• 2001

Since most real-world application data involve continuous-valued attributes, properly addressing the discretization process for constructing a decision tree is an important problem. A continuous-valued attribute is typically discretized during decision tree generation by partitioning its range into two intervals recursively. In this paper, by removing the restriction to the binary discretization, we present a hybrid multi-interval discretization algorithm for discretizing the range of continuous-valued attribute into multiple intervals. On the basis of experiment using semiconductor etching machine, it has been verified that our discretization algorithm constructs a more efficient incremental decision tree compared to previously proposed discretization algorithms.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.2
• /
• pp.374-388
• /
• 2011

Biometric discretization is a process of transforming continuous biometric features of an identity into a binary bit string. This paper mainly focuses on improving the global discretization method - a discretization method that does not base on information specific to each user in bitstring extraction, which appears to be important in applications that prioritize strong security provision and strong privacy protection. In particular, we demonstrate how the actual performance of a global discretization could further be improved by embedding a global discriminative feature selection method and a Linearly Separable Subcode-based encoding technique. In addition, we examine a number of discriminative feature selection measures that can reliably be used for such discretization. Lastly, encouraging empirical results vindicate the feasibility of our approach.

• Bulletin of the Korean Mathematical Society
• /
• v.48 no.4
• /
• pp.873-896
• /
• 2011

The semi-discrete central scheme and central upwind scheme use Runge-Kutta (RK) time discretization. We do the Lax-Wendroff (LW) type time discretization for both schemes. We perform numerical experiments for various problems including two dimensional Riemann problems for Burgers' equation and Euler equations. The results show that the LW time discretization is more efficient in CPU time than the RK time discretization while maintaining the same order of accuracy.

• Structural Engineering and Mechanics
• /
• v.6 no.3
• /
• pp.339-345
• /
• 1998

The computation size and accuracy in the boundary element method are mutually coupled and strongly influenced by the formulations in boundary discretization and integration. This aspect is studied numerically for two-dimensional elastodynamic problems in the frequency-domain. The localized nature of error is observed in the computed results. A boundary discretization criterion is examined. The number of integration points in the boundary integration is studied to find the optimum number for accuracy. Useful information is obtained concerning the optimization in boundary discretization and integration.

• Database Research
• /
• v.34 no.3
• /
• pp.159-169
• /
• 2018

Discretization of continuous variables intended to improve the performance of various algorithms such as data mining by transforming quantitative variables into qualitative variables. If we use appropriate discretization techniques for data, we can expect not only better performance of classification algorithms, but also accurate and concise interpretation of results and speed improvements. Various discretization techniques have been studied up to now, and however there is still demand of research on discretization studies. In this paper, we propose a new discretization technique to set the cut-point using Wasserstein distance with considering the distribution of continuous variable values with classes of data. We show the superiority of the proposed method through the performance comparison between the proposed method and the existing proven methods.

• Structural Engineering and Mechanics
• /
• v.3 no.4
• /
• pp.391-410
• /
• 1995

The application of adaptive finite element method to dynamic problems is investigated. Both the kinetic and strain energy errors induced by space and time discretization were estimated in a consistent manner and controlled by the simultaneous use of the adaptive mesh generation and the automatic time stepping. Also an optimal ratio of spatial discretization error to temporal discretization error was discussed. In this study it was found that the best performance can be obtained when the specified spatial and temporal discretization errors have the same value. Numerical examples are carried out to verify the performance of the procedure.

• Proceedings of the KIEE Conference
• /
• 2005.05a
• /
• pp.125-127
• /
• 2005

In this paper, we propose a new scheme for the discretization of nonlinear systems using Taylor series expansion and the zero-order hold assumption. This scheme is applied to the sample-data representation of a nonlinear system with constant state tine-delay. The mathematical expressions of the discretization scheme are presented and the effect of the time-discretization method on key properties of nonlinear control system with state tine-delay, such as equilibrium properties and asymptotic ability, is examined. The proposed scheme provides a finite-dimensional representation for nonlinear systems with state time-delay enabling existing controller design techniques to be applied to then. The performance of the proposed discretization procedure is evaluated using a nonlinear system. For this nonlinear system, various sampling rates and time-delay values are considered.

• Communications for Statistical Applications and Methods
• /
• v.10 no.3
• /
• pp.879-894
• /
• 2003

We evaluated the efficiencies of applying attribute selection methods and prior discretization to supervised learning, modelled by C4.5 and Naive Bayes. Three databases were obtained from UCI data archive, which consisted of continuous attributes except for one decision attribute. Four methods were used for attribute selection : MDI, ReliefF, Gain Ratio and Consistency-based method. MDI and ReliefF can be used for both continuous and discrete attributes, but the other two methods can be used only for discrete attributes. Discretization was performed using the Fayyad and Irani method. To investigate the effect of noise included in the database, noises were introduced into the data sets up to the extents of 10 or 20%, and then the data, including those either containing the noises or not, were processed through the steps of attribute selection, discretization and classification. The results of this study indicate that classification of the data based on selected attributes yields higher accuracy than in the case of classifying the full data set, and prior discretization does not lower the accuracy.