• Journal of the Korean Institute of Intelligent Systems
• /
• v.23 no.5
• /
• pp.400-405
• /
• 2013

The purpose of this study is to develop a composite platform for knowledge extractions, visualizations, and inference. Generally, the big data sets were frequently used in the healthcare and medical area. To help the knowledge managers/users working in the field, this study is focused on knowledge management (KM) based on Data Mining (DM), Knowledge Distribution Map (KDM), Decision Tree (DT), RDBMS, and SQL-inference. The proposed mechanism is composed of five key processes. Firstly, in Knowledge Parsing, it extracts logical rules from a big data set by using DM technology. Then it transforms the rules into RDB tables. Secondly, through Knowledge Maintenance, it refines and manages the knowledge to be ready for the computing of knowledge distributions. Thirdly, in Knowledge Distribution process, we can see the knowledge distributions by using the DT mechanism.Fourthly, in Knowledge Hierarchy, the platform shows the hierarchy of the knowledge. Finally, in Inference, it deduce the conclusions by using the given facts and data.This approach presents the advantages of diversity in knowledge representations and inference to improve the quality of computer-based medical diagnosis.

• Proceedings of the KOSOMBE Conference
• /
• v.1996 no.11
• /
• pp.225-228
• /
• 1996

ECG data are used for the diagnostic purposes with many clinical situations, especially heart disease. In this paper, an efficient ECG data compression technique by wavelet transform and high-speed vector quantization on PMS-B algorithm is proposed. In general, ECG data compression techniques are divided into two categories: direct and transform methods. The direct data compression techniques are AZTEC, TP, CORTES, FAN and SAPA algorithms, besides the transform methods include K-L, Fourier, Walsh, and wavelet transforms. In this paper, we applied wavelet analysis to the ECG data. In particular, vector quantization on PMS-B algorithm to the wavelet coefficients in the higher frequency regions, but scalar quantized in the lower frequency regions by PCM. Finally, the quantized indices were compressed by LZW lossless entropy encoder. As the result of simulation, it turns out to get sufficient compression ratio while keeping clinically acceptable PRD.

• Journal of the Korean Society for Precision Engineering
• /
• v.8 no.1
• /
• pp.105-115
• /
• 1991

This paper describes an integrated CAD and CAPP system for prismatic parts of injection mold which generates a complete process plan automatically from CAD data of a part without human intervention. This system employs Auto CAD as a CAD model and GS-CAPP as an automatic process planning system for injection mold. The proposed CAD/CAPP system consists of three modules such as CAD data conversion module, manufacturing feature recognition module, and CAD/CAPP interface module. CAD data conversion module transforms design data of AutoCAD into three dimensional part data. Manufacturing feature recognition module extracts specific manufacturing features of a part using feature recognition rule base. Each feature can be recognized by combining geometry, position and size of the feature. CAD/CAPP interface module links manufacturing feature codes and other head data to automatic process planning system. The CAD/CAPP system can improve the efficiency of process planning activities and reduce the time required for process planning. This system can provide a basis for the development of part feature based design by analyzing manufacturing features.

• Smart Media Journal
• /
• v.10 no.3
• /
• pp.9-22
• /
• 2021

Basically, machine learning models use input data to produce results. Sometimes, the input data is too complicated for the models to learn useful patterns. Therefore, feature engineering is a crucial data preprocessing step for constructing a proper feature set to improve the performance of such models. One of the most efficient methods for automating feature engineering is the autoencoder, which transforms the data from its original space into a latent space. However certain factors, including the datasets, the machine learning models, and the number of dimensions of the latent space (denoted by k), should be carefully considered when using the autoencoder. In this study, we design a framework to compare two data preprocessing approaches: with and without autoencoder and to observe the impact of these factors on autoencoder. We then conduct experiments using autoencoders with classifiers on popular datasets. The empirical results provide a perspective regarding the best suited autoencoder for these factors.

• Proceedings of the IEEK Conference
• /
• 2001.06d
• /
• pp.49-52
• /
• 2001

This paper presents a FPGA Implementation of wavelet-based CODEC, which can compress 2-dimensional image. For real-time processing, a scheduling method of input image data is proposed and a new structure of MAC(multiplier-accumulator) is proposed for wavelet transforms. Also this study proposes global pipelining structure of wavelet CODEC and efficient buffering method at interfaces between each module with different clock frequency.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.259-262
• /
• 2000

In this paper, an Adaptive neuro-fuzzy Inference system(ANFIS) using fuzzy min-max network(FMMN) is proposed. Fuzzy min-max network classifier that utilizes fuzzy sets as pattern classes is described. Each fuzzy set is an aggregation of fuzzy set hyperboxes. Here, the proposed method transforms the hyperboxes into gaussian menbership functions, where the transformed membership functions are inserted for generating fuzzy rules of ANFIS. Finally, we applied the proposed method to the classification problem of iris data and obtained a better performance than previous works.

• Electronics and Telecommunications Trends
• /
• v.9 no.4
• /
• pp.9-21
• /
• 1994

Wide variety of the architectural complexity of parallel computer often makes it difficult to develop efficient programs for them. One of approaches to improve this difficulty is to program in familiar sequential languages such as Fortran or C and to parallelize sequential programs into equivalent parallel programs automatically. This paper presents an organization of parallelizing compiler which transforms sequential programs into equivalent parallel programs. The parallelizer consists mainly of syntax analysis, control and data flow analysis, program transformation, and parallel code generation. In particular, the program restructuring in this parallelizer maximizes loop parallelism.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.367-367
• /
• 2000

In this paper, an Adaptive neuro-fuzzy Inference system(ANFIS) using fuzzy min-max network(FMMN) is proposed. Fuzzy min-max network classifier that utilizes fuzzy sets as pattern classes is described. Each fuzzy set is an aggregation of fuzzy set hyperboxes. Here, the proposed method transforms the hyperboxes into gaussian membership functions, where the transformed membership functions are inserted for generating fuzzy rules of ANFIS. Finally, we applied the proposed method to the classification problem of iris data and obtained a better performance than previous works.

• Journal of the Korean Statistical Society
• /
• v.25 no.2
• /
• pp.265-276
• /
• 1996

The nonparametric deconvolution problems are studied to recover an unknown density when the data are contaminated with Gaussian error. We propose the estimator which is a linear combination of kernel type estimates of derivertives of the observed density function. We show that this estimator is consistent and also consider the properties of estimator at small sample by simulation.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.29B no.1
• /
• pp.26-33
• /
• 1992

This paper presents lattice structure of bilinear filter and the conversion equations from lattice parameters to direct-form parameters. Billnear models are attractive for adaptive filtering applications because they can approximate a large class of nonlinear systems adequately, and usually with considerable parsimony in the number of coefficients required. The lattice filter formulation transforms the nonlinear filtering problem into an equivalent multichannel linear filtering problem and then uses multichannel lattice filtering algorithms to solve the nonlinear filtering problem. The lattice filters perform a Gram-Schmidt orthogonalization of the input data and have very good easily extended to more general nonlinear output feedback structures.