• International Journal of Contents
• /
• v.7 no.4
• /
• pp.1-5
• /
• 2011

Physics based fluid animation schemes need large computation cost due to tremendous degree of freedom. Many researchers tried to reduce the cost for solving the large linear system that is involved in grid-based schemes. GPU based algorithms and advanced numerical analysis methods are used to efficiently solve the system. Other groups studied local operation methods such as SPH (Smoothed Particle Hydrodynamics) and LBM (Lattice Boltzmann Method) for enhancing the efficiency. Our method investigates this efficiency problem thoroughly, and suggests novel paradigm in fluid animation field. Rather than physics based simulation, we propose a robust boundary handling technique for procedural fluid animation. Our method can be applied to arbitrary shaped objects and potential fields. Since only local operations are involved in our method, parallel computing can be easily implemented.

• Proceedings of the Korea Database Society Conference
• /
• 1999.06a
• /
• pp.353-359
• /
• 1999

We present a matrix-based inference algorithm suitable for electronic commerce applications. For this purpose, an Extended AND-OR Graph (EAOG) was developed with the intention that fast inference process is enabled within the electronic commerce situations. The proposed EAOG inference mechanism has the following three characteristics. 1. Real-time inference: The EAOG inference mechanism is suitable for the real-time inference because its computational mechanism is based on matrix computation. 2. Matrix operation: All the subjective knowledge is delineated in a matrix form. so that inference process can proceed based on the matrix operation which is computationally efficient. 3. Bi-directional inference: Traditional inference method of expert systems is based on either forward chaining or backward chaining which is mutually exclusive in terms of logical process and computational efficiency. However, the proposed EAOG inference mechanism is generically bi-directional without loss of both speed and efficiency. We have proved the validity of our approach with several propositions and an illustrative EC example.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1993.10a
• /
• pp.195-199
• /
• 1993

In the analysis of metal forming processes by the finite element method, there are many numerical instabilities such as element locking, hourglass mode, shear locking. These instabilities may have a bad effect upon accuracy and convergence. The present work is concerned with improvement of stability and efficiency in two dimensional rigid-plastic finite element method using various type of elements and numerical integration schemes. AS metal forming examples, upsetting and backward extrusion are taken for comparison among the methods : various element types and numerical integration schemes. comparison is made in terms of stability and efficiency. As a result, it has been shown that the finite element computation is stabilized from the viewpoint of computational time, convergency, and numerical instability.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.1
• /
• pp.189-196
• /
• 2004

A modified generalized particle algorithm, MGPA, was suggested to improve the calculation efficiency of standard SPH Method in numerical analysis of high speed impact behavior. MGPA had a new weight function to reduce computation time. The efficiency of this method was proven through calculation for the sample problems of one dimensional rod impact problem and two dimensional plate impact problem. The MGPA method reduced the calculation error and stress oscillation near the boundaries. The validity of this approach was shown by the comparison with ABAQUS results in two dimensional plate impact problem.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.05a
• /
• pp.551-554
• /
• 2002

The paper discusses new approach for machining operation simulation using enhanced Z map algorithm. To extract the required geometric information from NC code, suggested algorithm uses supersampling method to enhance the efficiency of a simulation process. By executing redundant Boolean operations in a grid cell and averaging down calculated data, presented algorithm can accurately represent material removal volume though tool swept volume is negligibly small. Supersampling method is the most common form of antialiasing and usually used with polygon mesh rendering in computer graphics. The key advantage of enhanced Z map model is that the data structure is same with conventional Z map model, though it can acquire higher accuracy and reliability with same or lower computation time. By simulating machining operation efficiently, this system can be used to improve the reliability and efficiency of NC machining process as well as the quality of the final product.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 1999.03a
• /
• pp.353-359
• /
• 1999

We present a matrix-based inference alorithm suitable for electronic commerce applications. For this purpose, an Extended AND-OR Graph (EAOG) was developed with the intention that fast inference process is enabled within the electronic commerce situations. The proposed EAOG inference mechanism has the following three characteristics. 1. Real-time inference: The EAOG inference mechanism is suitable for the real-time inference because its computational mechanism is based on matric computation.2. Matrix operation: All the subjective knowledge is delineated in a matrix form, so that inference process can proceed based on the matrix operation which is computationally efficient.3. Bi-directional inference: Traditional inference method of expert systems is based on either forward chaining or backward chaining which is mutually exclusive in terms of logical process and exclusive in terms of logical process and computational efficiency. However, the proposed EAOG inference mechanism is generically bi-directional without loss of both speed and efficiency. We have proved the validity of our approach with several propositions and an illustrative EC example.

• Journal of Information Processing Systems
• /
• v.14 no.2
• /
• pp.418-434
• /
• 2018

With the increasing use of the Internet and electronic documents, automatic text categorization becomes imperative. Several machine learning algorithms have been proposed for text categorization. The k-nearest neighbor algorithm (kNN) is known to be one of the best state of the art classifiers when used for text categorization. However, kNN suffers from limitations such as high computation when classifying new instances. Instance selection techniques have emerged as highly competitive methods to improve kNN through data reduction. However previous works have evaluated those approaches only on structured datasets. In addition, their performance has not been examined over the text categorization domain where the dimensionality and size of the dataset is very high. Motivated by these observations, this paper investigates and analyzes the impact of instance selection on kNN-based text categorization in terms of various aspects such as classification accuracy, classification efficiency, and data reduction.

• Korean Management Science Review
• /
• v.5 no.1
• /
• pp.56-70
• /
• 1988

In this study, We reviewed the solution methods (for the heuristic and optimization method) for the single-item dynamic lot-sizing problem, and improved the efficiency (speed and optimality) of the conventional heuristic method by utilizing the inventory decomposition property. The iventory decomposition property decomposes the given original problem into several independent subproblems without violating the optimality conditions. Then we solve each decomposed subproblems by using the conventional heuristics such as LTC, LUC, Silver-Meal etc. For testing the efficiency of the proposed decomposition method, we adopted the data sets given in Kaimann, Berry and Silver-Meal. The computational results show that the suggested problem solving framework results in some promising effects on the computation time and the degree of optimality.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.594-597
• /
• 2006

Image co-registration is the process of overlaying two images of the same scene. One of which is a reference image, while the other (sensed image) is geometrically transformed to the one. Numerous methods were developed for the automated image co-registration and it is known as a time-consuming and/or computation-intensive procedure. In order to improve efficiency and effectiveness of the co-registration of satellite imagery, this paper proposes a pre-qualified area matching, which is composed of feature extraction with Laplacian filter and area matching algorithm using correlation coefficient. Moreover, to improve the accuracy of co-registration, the outliers in the initial matching point should be removed. For this, two outlier detection techniques of studentized residual and modified RANSAC algorithm are used in this study. Three pairs of Landsat images were used for performance test, and the results were compared and evaluated in terms of robustness and efficiency.

• Journal of Mechanical Science and Technology
• /
• v.20 no.7
• /
• pp.929-940
• /
• 2006

Early detection and diagnosis of incipient induction machine faults increases machinery availability, reduces consequential damage, and improves operational efficiency. However, fault detection using analytical methods is not always possible because it requires perfect knowledge of a process model. This paper proposes a neural network based expert system for diagnosing problems with induction motors using vibration analysis. The short-time Fourier transform (STFT) is used to process the quasi-steady vibration signals, and the neural network is trained and tested using the vibration spectra. The efficiency of the developed neural network expert system is evaluated. The results show that a neural network expert system can be developed based on vibration measurements acquired on-line from the machine.