• 대한음성학회지:말소리
• /
• 제50호
• /
• pp.123-137
• /
• 2004

In this paper, we consider the implementation of connected digit recognition system using tree structured lexicon model. To implement efficiently the fixed or variable length digit recognition system, finite state network (FSN) is required. We merge the word network algorithm that implements the FSN with lexical tree search algorithm that is used for general speech recognition system for fast search and large vocabulary systems. To find the efficient modeling of digit recognition system, we investigate some performance changes when the lexical tree search is applied.

• 대한기계학회논문집A
• /
• 제31권6호
• /
• pp.651-658
• /
• 2007

A new tree-structured neural network classifier is proposed for the automatic real-time inspection of cold-rolled steel strip surface defects. The defects are classified into 3 groups such as area type, disk type, area & line type in the first stage of the tree-structured neural network. The defects are classified in more detail into 11 major defect types which are considered as serious defects in the second stage of neural network. The tree-structured neural network classifier consists of 4 different neural networks and optimum features are selected for each neural network classifier by using SFFS algorithm and correlation test. The developed classifier demonstrates very plausible result which is compatible with commercial products having high world-wide market shares.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -1
• /
• pp.184-187
• /
• 2002

Two-dimensional (2-D) X-ray computerized tomography (CT) equipments are widely used in industrial and medical fields, and nowadays studies on reconstruction algorithm for 3-D cone-beam acquisition systems are active for better utilization. The authors recent-By have proposed a fast reconstruction aigorithm using tree-structured filter bank for 2-D C1, and shown the algorithm is applicable to an approximate reconstruction of 3-D CT. For exact 3-D CT reconstruction, however, we have to backproject 1-D signal into 3-D space. This paper proposes a fast implementation method for this back-projection by use of tree-structured filter bank. and shows the proposed method works approximately 700 times faster than the direct one with almost same reconstruction image quality.

• 한국산업정보학회논문지
• /
• 제24권2호
• /
• pp.1-14
• /
• 2019

웹, XML, 오피스 어플리케이션에 사용되는 대부분의 문서들은 트리 구조로 구성되어 있으며 특히 다중 사용자 환경에서의 트리 구조의 문서의 차이 발견, 합병, 버전 제어 등의 연구가 활발하다. 그러나 이들의 기초가 되는 편집스크립트에 대한 연구는 초보적인 상태에 있다. 본 논문에서는 편집연산들의 실행 시 트리구조의 문서의 변화를 이해하기 위한 문서 모델을 제시하고 편집연산들의 실행 효과의 분석을 통해 트리 구조 문서에 대한 인접한 편집연산들의 순서 교환 방법을 제안한다. 트리 구조 문서에 대한 변화탐지의 결과로 생성되는 편집스크립트들은 대부분 기본연산들(갱신, 삽입, 삭제)만으로 구성된다. 그러나 이동, 복사연산을 포함하는 경우, 이들의 복합연산의 특성으로 인해 주로 2단계 패스의 실행을 전제로 하는 편집스크립트를 생성한다. 본 논문에서는 제안한 편집 연산들의 순서 교환 방법을 통해 2단계 패스의 실행을 전제로 하는 X-treeESgen의 편집스크립트를 단일 패스로 변환하는 알고리즘을 제시한다.

• Journal of Information Processing Systems
• /
• 제18권5호
• /
• pp.677-687
• /
• 2022

We propose a detection algorithm based on tree-structured convolutional neural networks (TsCNNs) that finds pornography, propaganda, or other inappropriate content on a social media network. The algorithm sequentially applies the typical convolutional neural network (CNN) algorithm in a tree-like structure to minimize classification errors in similar classes, and thus improves accuracy. We implemented the detection system and conducted experiments on a data set comprised of 6 ordinary classes and 11 inappropriate classes collected from the Korean military social network. Each model of the proposed algorithm was trained, and the performance was then evaluated according to the images and videos identified. Experimental results with 20,005 new images showed that the overall accuracy in image identification achieved a high-performance level of 99.51%, and the effectiveness of the algorithm reduced identification errors by the typical CNN algorithm by 64.87 %. By reducing false alarms in video identification from the domain, the TsCNNs achieved optimal performance of 98.11% when using 10 minutes frame-sampling intervals. This indicates that classification through proper sampling contributes to the reduction of computational burden and false alarms.

• 한국정보과학회논문지:데이타베이스
• /
• 제34권5호
• /
• pp.465-472
• /
• 2007

XML과 같은 준구조적 데이타가 많이 사용됨에 따라 이를 효과적으로 저장하고 관리하는 것이 중요해지고 있다. XML 데이타는 트리 형태로 모델링이 가능하며, 기본적으로 질의 처리는 트리를 탐색하는 방식으로 이루어진다. 본 논문에서는 XML 데이타를 디스크 블록에 저장하는 알고리즘을 제안한다. 제안하는 알고리즘은 트리의 각 노드마다 아래쪽에서 위쪽으로 숫자를 할당하며 그 숫자를 이용하여 디스크 블록에 노드들을 매핑한다. 제안하는 알고리즘은 접근 패턴 정보를 필요로 하지 않으며 어떠한 접근 패턴에 대해서도 좋은 성능을 보인다. 제안하는 방법의 몇가지 특성을 증명하고, 실험을 통해서 성능을 평가한다.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 1996년도 추계학술대회 학술발표 논문집
• /
• pp.154-157
• /
• 1996

Conventional fuzzy systems have serious problems in dealing with the nonlinear approximations on high-dimensional spaces due to the explosive increase of the number of fuzzy IF-THEN rules. In order to avoid such problems, this paper proposes a tree-structured fuzzy system in which semi-local basis functions form its basic elements, and develops a training algorithm for the proposed system based on the evolution program and LMS rule. The experimental studies demonstrate the effectiveness of the developed methodology.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권12호
• /
• pp.3117-3132
• /
• 2012

The tree topology in multicast systems has high transmission efficiency, low latency, but poor resilience to node failures. In our work, some nodes are selected as backbone nodes to construct a tree-like core overlay. Backbone nodes are reliable enough and have strong upload capacity as well, which is helpful to overcome the shortcomings of tree topology. The core overlay is organized into a spanning tree while the whole overlay is of mesh-like topology. This paper focuses on improving the performance of the application-layer multicast overlay by optimizing the core overlay which is periodically adjusted with the proposed optimization algorithm. Our approach is to construct the overlay tree based on the out-degree weighted reliability where the reliability of a node is weighted by its upload bandwidth (out-degree). There is no illegal solution during the evolution which ensures the evolution efficiency. Simulation results show that the proposed approach greatly enhances the reliability of the tree-like core overlay systems and achieves shorter delay simultaneously. Its reliability performance is better than the reliability-first algorithm and its delay is very close to that of the degree-first algorithm. The complexity of the proposed algorithm is acceptable for application. Therefore the proposed approach is efficient for the topology optimization of a real multicast overlay.

• Communications for Statistical Applications and Methods
• /
• 제19권2호
• /
• pp.293-301
• /
• 2012

Quantile regression proposed by Koenker and Bassett (1978) is a statistical technique that estimates conditional quantiles. The advantage of using quantile regression is the robustness in response to large outliers compared to ordinary least squares(OLS) regression. A regression tree approach has been applied to OLS problems to fit flexible models. Loh (2002) proposed the GUIDE algorithm that has a negligible selection bias and relatively low computational cost. Quantile regression can be regarded as an analogue of OLS, therefore it can also be applied to GUIDE regression tree method. Chaudhuri and Loh (2002) proposed a nonparametric quantile regression method that blends key features of piecewise polynomial quantile regression and tree-structured regression based on adaptive recursive partitioning. Lee and Lee (2006) investigated wage determinants in the Korean labor market using the Korean Labor and Income Panel Study(KLIPS). Following Lee and Lee, we fit three kinds of quantile regression tree models to KLIPS data with respect to the quantiles, 0.05, 0.2, 0.5, 0.8, and 0.95. Among the three models, multiple linear piecewise quantile regression model forms the shortest tree structure, while the piecewise constant quantile regression model has a deeper tree structure with more terminal nodes in general. Age, gender, marriage status, and education seem to be the determinants of the wage level throughout the quantiles; in addition, education experience appears as the important determinant of the wage level in the highly paid group.

• 한국지능시스템학회논문지
• /
• 제6권3호
• /
• pp.25-36
• /
• 1996

기존의 RBF 신경망 및 퍼지 시스템을 고차원 입력 공간 위에서의 비선형 근사에 적용할 경우 은닉 노드의 수혹은 퍼지 IF-THEN 규칙의 수가 기하급수적으로 증가한다. 본 논문에서는 이러한 문제점을 개선하기 위해 반국소 유닛을 기본 요소로 하는 수목구조지능시스템을 제안하고, 이를 효과적으로 학습하기 위하여 수정 유전자 알고리즘 및 LMS 규칙에 기반을 둔 학습 알고리즘을 개발한다. 제안된 시스템에 대한 근사 능력 해석이 수행되고, 실험적 고찰을 통하여 개발된 방법론의 유용성이 입증된다.