• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10b
• /
• pp.313-318
• /
• 2007

위치기반서비스에서 사용자의 정확한 위치가 요구되면서 측위와 추적에 대한 연구가 활발히 진행되고 있다. 측위 방법에는 위성기반 방법[1, 2], 로컬네트워크기반 방법[3-6], 센서기반 방법[1, 7, 8, 9]등이 있다. 본 연구에서는 로컬네트워크 중 WLAN (Wireless Local Area Network) 환경의 옥내에서 사용자의 위치를 추적하는 기존의 방법의 정확성을 제고하는 방안을 제안한다. 제안하는 방법은 WLAN 환경에서 RSS를 측정하여 K-NN방식으로 현재 위치를 판단한 다음, 칼만필터를 사용하여 사용자의 위치와 이동경로를 예측한다는 점에서 기존의 방법과 비슷하다. 제안하는 방법의 특징은 도면 정보를 이용하는 것이다. 제안하는 방법은 도면정보로부터 갈림길 영역을 파악하고, 갈림길 영역에서는 측정치에 가중치를 두고 갈림길이 아닌 지역에서는 시스템 모델에 가중치를 두도록 파라메타의 값을 조절한다. 제안하는 방법의 효율성을 실험적으로 증명하기 위한 실험 결과와 분석 내용도 제시한다.

• Journal of KIISE:Software and Applications
• /
• v.28 no.3
• /
• pp.285-297
• /
• 2001

본 논문에서는 특정한 객체의 색 분포 모델링으로부터 얻어지는 가중 색 히스토그램과 지배적인 색의 영상공간 분포특성을 이용한 내용기반 영상 검색 방법을 제안한다. 특정한 객체의 예로 사람 얼굴을 선택했고, 그것의 색 분포를 u*-v* 색도 공간에서 모델링 했으며, 모델의 정규화된 부피를 균등 양자화된 색도 공간의 각 빈(bin)의 히스토그램 값에 대한 가중치로 결정하고, 결정된 가중치를 히스토그램 정합 과정에 적용하였다. 또한 색 히스토그램 값이 큰 특정한 수의 빈으로 정의되는 지배적인 색의 영상 공간 분포를 가중 색 히스토그램과 함께 유사성의 측정기준으로 사용하였다. 제안한 검색 방법을 500여개의 영상에 대해 실험한 결과 제안한 방법이 얼굴을 포함하는 영상을 질의로 주었을 때 얼굴을 포함하는 영상을 우선적으로 찾는데 효과적임을 확인하였다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.7
• /
• pp.610-614
• /
• 1991

This paper proposes a trajectory constrol fo a robot manipulator by using neural network. The inverse dynamic model of manipuator is learned by neural network. The manipulator is controlled by weight values of the learned neural network. The weight valuese is change with a torque of liner vontroller and a acceleration error. Phsically, the totlal torque for a manipualator is a sum of the liner controller torque and the nerural network controller torque. The proposed control effect is estimated by computer simulation.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.308-309
• /
• 2009

Human vision information processing has many characteristics when image information is transmitted from retina to visual cortex. Among them, we analyze the sensibility of the orientation on an image and compare the recognition rates by the response_weight of the vertical, horizontal and diagonal orientation. In statistics analysis, we show that a particular simple cell responds best to a bar with a vertical orientation. After then, we will apply the characteristics to Human visual system.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.592-596
• /
• 2021

레코드 연결의 대표적인 문제 중 하나는 레코드 간 비교 비용이 크다는 것이다. 이러한 문제를 해결하기 위해서는 레코드 연결에 필수적으로 블로킹 단계가 포함되어야 한다. 블로킹이란 같은 레코드일 가능성이 높은 대상들을 그룹화하여 비교연산을 수행할 대상을 선정하는 단계를 말한다. 블로킹의 목적은 최대한 결과의 recall을 희생시키지 않으면서 비교 연산 횟수 최소화하는 것이다. 메타 블로킹은 가중치 그래프를 블로킹에 적용함으로써 전통적인 블로킹 방식의 한계를 극복하고 더 좋은 성능을 나타내는 모델이다. 본 논문에서는 메타블로킹에서 주목하지 않았던 블록 생성방식을 데이터베이스 속성에 따라 블록을 생성하는 방식으로 개선하고 그에 맞는 가중치 계산식을 제안하였다. 또한 키 기반 블로킹, 메타블로킹, 속성인지 메타블로킹으로 생성된 블로킹 결과에 대한 성능을 측정 및 비교하였다.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.8 no.3
• /
• pp.65-73
• /
• 2005

Because the geological features of Imha basin are composed of clay and shale layer, much soil particle flows into reservoir in shape of muddy water when it rains a lot. Therefore, turbidity data can be indirect-index to estimate the soil erosion of Imha basin. This study evaluated annual soil erosion using GIS-based soil erosion model and applied rainfall weighting value method by time-series rainfall data to estimate monthly soil erosion. In view of 2003 turbidity data, monthly soil erosion with rainfall weighting value is more efficient than monthly soil erosion with rainfall data.

• Journal of Korea Multimedia Society
• /
• v.13 no.2
• /
• pp.216-224
• /
• 2010

A RAM-based Neural Network(RNN) has the advantages of processing speed and hardware implementation. In spite of these advantages, it has a saturation problem, weakness of repeated learning and extract of a generalized pattern. To resolve these problems of RNN, the 3DNS model using cumulative multi discriminator was proposed. But that model does not solve the saturation problem yet. In this paper, we proposed a adaptive weight cumulative neural net(AWCNN) using the adaptive weight neuron (AWN) for solving the saturation problem. The proposed nets improved a recognition rate and the saturation problem of 3DNS. We experimented with the MNIST database of NIST without preprocessing. As a result of experimentations, the AWCNN was 1.5% higher than 3DNS in a recognition rate when all input patterns were used. The recognition rate using generalized patterns was similar to that using all input patterns.

• Journal of Digital Convergence
• /
• v.15 no.6
• /
• pp.229-236
• /
• 2017

An efficient frequent pattern algorithm is essential for mining association rules as well as many other mining tasks for convergence with its application spread over a very broad spectrum. Models for mining pattern have been proposed using a FP-tree for storing compressed information about frequent patterns. In this paper, we propose a centroid frequent pattern growth algorithm which we called "CAWFP-Growth" that enhances he FP-Growth algorithm by making the center of weights and frequencies for the itemsets. Because the conventional constraint of maximum weighted support is not necessary to maintain the downward closure property, it is more likely to reduce the search time and the information loss of the frequent patterns. The experimental results show that the proposed algorithm achieves better performance than other algorithms without scarifying the accuracy and increasing the processing time via the centroid of the items. The MapReduce framework model is provided to handle large amounts of data via a pseudo-distributed computing environment. In addition, the modeling of the proposed algorithm is required in the fully distributed mode.

• Journal of KIISE
• /
• v.42 no.11
• /
• pp.1349-1360
• /
• 2015

In order to improve software quality, it is necessary to efficiently and effectively remove software defects in source codes. In the development field, defects are removed according to removal ratio or severity of defects. There are several studies on the removal of defects based on software quality attributes, and several other studies have been done to improve the software quality using classification of the severity of defects, when working on projects. These studies have thus far been insufficient in terms of identifying if there exists relationships between defects or whether any type of defect is more important than others. Therefore, in this study, we collected various types of software defects, standards organization, companies, and researchers. We modeled the defects types using an ANP model, and developed the weighted severities of the defects types, with respect to the general application software, using the ANP model. When general application software is developed, we will be able to use the weight for each severity of defect type, and we expect to be able to remove defects efficiently and effectively.

• Journal of KIISE:Software and Applications
• /
• v.29 no.4
• /
• pp.248-257
• /
• 2002

Clustering is a method for grouping objects with similar properties into a same cluster. SVD(Singular Value Decomposition) is known as an efficient preprocessing method for clustering because of dimension reduction and noise elimination for a high dimensional and sparse data set like E-Commerce data set. However, it is hard to evaluate the worth of original attributes because of information loss of a converted data set by SVD. This research proposes a cluster feature selection method, called ENTROPY-SVD, to find important attributes for each cluster based on entropy weighting and SVD. Using SVD, one can take advantage of the latent structures in the association of attributes with similar objects and, using entropy weighting one can find highly dense attributes for each cluster. This paper also proposes a model-based collaborative filtering recommendation system with ENTROPY-SVD, called CFS-CF and evaluates its efficiency and utilization.