• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.413-435
• /
• 2018

Although the accuracy of handwritten character recognition based on deep networks has been shown to be superior to that of the traditional method, the use of an overly deep network significantly increases time consumption during parameter training. For this reason, this paper took the training time and recognition accuracy into consideration and proposed a novel handwritten character recognition algorithm with newly designed network structure, which is based on an extended nonlinear kernel residual network. This network is a non-extremely deep network, and its main design is as follows:(1) Design of an unsupervised apriori algorithm for intra-class clustering, making the subsequent network training more pertinent; (2) presentation of an intermediate convolution model with a pre-processed width level of 2;(3) presentation of a composite residual structure that designs a multi-level quick link; and (4) addition of a Dropout layer after the parameter optimization. The algorithm shows superior results on MNIST and SVHN dataset, which are two character benchmark recognition datasets, and achieves better recognition accuracy and higher recognition efficiency than other deep structures with the same number of layers.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.2
• /
• pp.175-182
• /
• 2008

Context-aware service environment should provide many kinds of services according to users' requests. Users want a great variety of services. In response to their demands the service provider should make a new service every time. But making a new service every time may be inefficient even for a small number of users' requests. So, there are studies on how to efficiently support various and complex requests fFom users. In many researches, service compositions have lately attracted considerable attention. However, existing researches have mainly focused on Web services. So they are not proper to rapidly providing services in response to users' requests, especially In context-aware service environment. This paper proposes a rapid service composition using case-based reasoning. For evaluating the proposed algorithm we implement 'purchasing seTvice agent'. With this system, we compare our algorithm and the existing service composition algorithms.

• Journal of KIISE:Computer Systems and Theory
• /
• v.31 no.12
• /
• pp.716-725
• /
• 2004

It is extremely important to minimize network access time in constructing a high-performance PC cluster system. For PC cluster systems, it is possible to reduce network access time by maintaining network cache in each cluster node. This paper presents a Network Cache Coherent NUMA (NCC-NUMA) system to utilize network cache by locating shared memory on the PCI bus, and the NCC-NUMA card which is core module of the NCC-NUMA system is developed. The NCC-NUMA card is directly plugged into the PCI slot of each node, and contains shared memory, network cache, shared memory control module and network control module. The network cache is maintained for the shared memory on the PCI bus of cluster nodes. The coherency mechanism between the network cache and the shared memory is based on the IEEE SCI standard. According to the SPLASH-2 benchmark experiments, the NCC-NUMA system showed improvements of 56% compared with an SCI-based cluster without network cache.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.433-436
• /
• 2004

In this paper, we introduce a new Fuzzy Polynomial Neural Networks (FPNNS)-like structure whose neuron is based on the Fuzzy Set-based Fuzzy Inference System (FS-FIS) and is different from that of FPNNS based on the Fuzzy relation-based Fuzzy Inference System (FR-FIS) and discuss the ability of the new FPNNS-like structure named Fuzzy Set-based Polynomial Neural Networks (FSPNN). The premise parts of their fuzzy rules are not identical, while the consequent parts of the both Networks (such as FPNN and FSPNN) are identical. This difference results from the angle of a viewpoint of partition of input space of system. In other word, from a point of view of FS-FIS, the input variables are mutually independent under input space of system, while from a viewpoint of FR-FIS they are related each other. The proposed design procedure for networks architecture involves the selection of appropriate nodes with specific local characteristics such as the number of input variables, the order of the polynomial that is constant, linear, quadratic, or modified quadratic functions being viewed as the consequent part of fuzzy rules, and a collection of the specific subset of input variables. On the parameter optimization phase, we adopt Information Granulation (IC) based on HCM clustering algorithm and a standard least square method-based learning. Through the consecutive process of such structural and parametric optimization, an optimized and flexible fuzzy neural network is generated in a dynamic fashion. To evaluate the performance of the genetically optimized FSPNN (gFSPNN), the model is experimented with using the time series dataset of gas furnace process.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.11
• /
• pp.1661-1670
• /
• 2016

In this paper, we present a risk prediction system and customized evacuation pathfinding algorithm in fire scenarios. For the risk prediction, we apply a multi-level clustering mechanism using collected temperature at sensor nodes throughout the network in order to predict the temperature at the time that users actually evacuate. Based on the predicted temperature and its reliability, we suggest an evacuation pathfinding algorithm that finds a suitable evacuation path from a user's current location to the safest exit. Simulation results based on FDS(Fire Dynamics Simulator) of NIST for a wireless sensor network consisting of 47 stationary nodes for 1436.41 seconds show that our proposed prediction system achieves a higher accuracy by a factor of 1.48. Particularly for nodes in the most reliable group, it improves the accuracy by a factor of up to 4.21. Also, the customized evacuation pathfinding based on our prediction algorithm performs closely with that of the ground-truth temperature in terms of the ratio of safe nodes on the selected path, while outperforming the shortest-path evacuation with a factor of up to 12% in terms of a safety measure.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.2
• /
• pp.227-242
• /
• 2019

This paper presents a method to predict ground types ahead of a tunnel face utilizing operational data of the earth pressure-balanced (EPB) shield tunnel boring machine (TBM) when running through soil ground. The time series analysis model which was applicable to predict the mixed ground composed of soils and rocks was modified to be applicable to soil tunnels. Using the modified model, the feasibility on the choice of the soil conditioning materials dependent upon soil types was studied. To do this, a self-organizing map (SOM) clustering was performed. Firstly, it was confirmed that the ground types should be classified based on the percentage of 35% passing through the #200 sieve. Then, the possibility of predicting the ground types by employing the modified model, in which the TBM operational data were analyzed, was studied. The efficacy of the modified model is demonstrated by its 98% accuracy in predicting ground types ten rings ahead of the tunnel face. Especially, the average prediction accuracy was approximately 93% in areas where ground type variations occur.

• Journal of the Korean GEO-environmental Society
• /
• v.16 no.4
• /
• pp.5-12
• /
• 2015

Civil engineers have taken the numerous slope monitoring data for an engineering project subjected to hazard potential of slide. However, the topics on how to deal with and draw out proper information from the data related to the slope behavior have not been widely discussed. Recently, several researchers had installed the real-time monitoring system to cope with slope failure; however they are mainly focused on the hardware system installation. Therefore, this study tries to show how the measured data could be grouped and connected each other. The basic idea of analyzing method studied in this paper came from the clustering, which is the part of data mining analysis. Therefore, at the base of classification of time series data, the authors suggest three mathematical data analyzing methods; Average Index of different displacement ($AD_{i,j}$), Difference of average relative displacement ($\overline{RD}_{i,j}$) and Coordinate system of average and relative displacement ($\overline{RD}$, AD). These analyzing methods are based on the statistical method and failure mechanism of slope. Therefore they showed clustering relationships of the similar parts of the slope which makes the same sliding mechanism.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2002.11a
• /
• pp.345-352
• /
• 2002

무선 인터넷을 이용하는 사용자는 정보의 양의 따른 시간적 통신비용의 증가 문제로 개인화 에이전트가 사용자의 관심에 따라 서비스를 제공하는 기능과 맞춤화된 정보를 제공하는 기능, 지식 기반 방식으로 정보를 예측하는 기능을 가지기를 바라고 있다. 본 논문에서는 이와 같이 무선 인터넷을 사용하는 사용자를 위한 PDA 개인화 에이전트 시스템을 구축하고자 한다. PDA 개인화 에이전트 시스템 구축을 위해 프로파일 기반의 에이전트 엔진과 사용자 프로파일을 이용한 지식기반 방식을 사용한다. 사용자가 웹페이지에서 행하는 행위들을 모니터링하여 사용자가 관심 가지는 문서를 파악하고 정보 검색을 통해 얻어진 문서를 분석하여 사용자 각각의 관심 문서로 나누어 서비스하게 된다. 모니터링 되어진 문서를 효과적으로 분석하기 위해 unsupervised clustering 기계학습 방식인 Cobweb을 이용한다. unsupervised 기계 학습은 conceptual 방식을 이용하여 검색되어진 정보를 사용자의 관심 분야별로 clustering한다. 클러스터링을 통해 얻어진 결과를 다시 기계학습을 통해 사용자 관심문서에 대한 프로파일을 생성하게 된다. 이렇게 만들어진 프로파일을 룰(Rule)로 만들어 이를 기반으로 사용자에게 서비스하게 된다. 이러한 룰은 사용자의 모니터링 결과로 얻어지기 때문에 주기적으로 업데이트하게 된다. 제안하는 시스템은 인터넷신문이나 웹진 등에서 사용자들에게 뉴스를 전달하기 위한 목적으로 생성하는 뉴스문서를 특정 대상으로 선정하였고 사용자 정보를 이용한 검색을 실시하고 결과로 얻어진 정보를 정보 분류를 통해 PDA나 휴대폰을 통해 사용자에게 제공한다. 상품을 검색하기 위한 검색노력을 줄이고, 검색된 대안들로부터 구매자와 시스템이 웹상에서 서로 상호작용(interactivity) 하여 해를 찾고, 제약조건과 규칙들에 의해 적합한 해를 찾아가는 방법을 제시한다. 본 논문은 구성기반 예로서 컴퓨터 부품조립을 사용해서 Template-based reasoning 예를 보인다 본 방법론은 검색노력을 줄이고, 검색에 있어 Feasibility와 Admissibility를 보장한다.매김할 수 있는 중요한 계기가 될 것이다.재무/비재무적 지표를 고려한 인공신경망기법의 예측적중률이 높은 것으로 나타났다. 즉, 로지스틱회귀 분석의 재무적 지표모형은 훈련, 시험용이 84.45%, 85.10%인 반면, 재무/비재무적 지표모형은 84.45%, 85.08%로서 거의 동일한 예측적중률을 가졌으나 인공신경망기법 분석에서는 재무적 지표모형이 92.23%, 85.10%인 반면, 재무/비재무적 지표모형에서는 91.12%, 88.06%로서 향상된 예측적중률을 나타내었다.ting LMS according to increasing the step-size parameter $\mu$ in the experimentally computed. learning curve. Also we find that convergence speed of proposed algorithm is increased by (B+1) time proportional to B which B is the number of recycled data buffer without complexity of compu

• Journal of KIISE:Computing Practices and Letters
• /
• v.9 no.1
• /
• pp.1-12
• /
• 2003

Recently, many researches into distributed processing on Internet, and multimedia data processing have been performed. Rapid and convenient multimedia services supplied with high quality and high speed are to be needed. In this paper, we design and implement clip-based video retrieval system on the Web enviroment in real-time. Our system consists of the content-based indexing system supporting convenient services for video content providers, and the Web-based retrieval system in order to make it easy and various information retrieval for users in the Web. Three important methods are used in the content-based indexing system, key frame extracting method by dividing video data, clip file creation method by clustering related information, and video database construction method by using clip unit. In Web-based retrieval system, retrieval method ny using a key word, two dimension browsing method of key frame, and real-time display method of the clip are used. In this paper, we design and implement the system that supports real-time display method of the clip are used. In this paper, we design and implement the system that supports real-time retrieval for video clips on Web environment and provides the multimedia service in stability. The proposed methods show a usefulness of video content providing, and provide an easy method for serching intented video content.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.8
• /
• pp.11-18
• /
• 2011

The increasing of web users load to the excessive transmission traffic and system overload problems. To solve these problems, cluster systems are studied. In conventional cluster systems, when the request size is large owing to such types as multimedia and CGI, the particular server load and response time tend to increase even if the overall loads are distributed evenly. In this paper, we consider the methodology of efficient resource usage, specially distributed web cluster system. We develope an algorithm that distributes the load on the web cluster system to use the system resources, such as system memory equally. The response time is chosen as a performance measure on the various clustering models. And based on the concurrent user to the web cluster system, the response time is also examined as the number of users increases. Simulation experience with this algorithm shows that the response time and average throughput seems to have a good results compare to those with the other algorithm.