http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Intrusion Detection Using Log Server and Support Vector Machines
Donghai Guan,Donggyu Yeo,Juwan Lee,Dukwhan Oh 한국정보과학회 2003 한국정보과학회 학술발표논문집 Vol.30 No.2Ⅰ
With the explosive rapid expansion of computer using during the past few years, security has become a crucial issue for modem computer systems. Today, there are many intrusion detection systems (IDS) on the Internet. A variety of intrusion detection techniques and tools exist in the computer security community such as enterprise security management system (ESM) and system integrity checking tools. However, there is a potential problem involved with intrusion detection systems that are installed locally on the machines to be monitored. If the system being monitored is compromised, it is quite likely that the intruder will alter the system logs and the intrusion logs while the intrusion remains undetected. In this project KIT-I, we adopt remote togging server (RLS) mechanism, which is used to backup the log files to the server. Taking into account security, we make use of the function of SSL of Java and certificate authority (CA) based key management. Furthermore, Support Vector Machine (SVM) is applied in our project to detect the intrusion activities.
Nearest neighbor editing aided by unlabeled data
Guan, Donghai,Yuan, Weiwei,Lee, Young-Koo,Lee, Sungyoung Elsevier 2009 Information sciences Vol.179 No.13
<P><B>Abstract</B></P><P>This paper proposes a novel method for nearest neighbor editing. Nearest neighbor editing aims to increase the classifier’s generalization ability by removing noisy instances from the training set. Traditionally nearest neighbor editing edits (removes/retains) each instance by the voting of the instances in the training set (labeled instances). However, motivated by semi-supervised learning, we propose a novel editing methodology which edits each training instance by the voting of all the available instances (both labeled and unlabeled instances). We expect that the editing performance could be boosted by appropriately using unlabeled data. Our idea relies on the fact that in many applications, in addition to the training instances, many unlabeled instances are also available since they do not need human annotation effort. Three popular data editing methods, including edited nearest neighbor, repeated edited nearest neighbor and All <I>k</I>-NN are adopted to verify our idea. They are tested on a set of UCI data sets. Experimental results indicate that all the three editing methods can achieve improved performance with the aid of unlabeled data. Moreover, the improvement is more remarkable when the ratio of training data to unlabeled data is small.</P>
Data Clustering Using Hybrid Neural Network
( Donghai Guan ),( Andrey Gavrilov ),( Weiwei Yuan ),( Sungyoung Lee ),( Young-koo Lee ) 한국정보처리학회 2007 한국정보처리학회 학술대회논문집 Vol.14 No.1
Clustering plays an indispensable role for data analysis. Many clustering algorithms have been developed. However, most of them suffer poor performance of learning. To archive good clustering performance, we develop a hybrid neural network model. It is the combination of Multi-Layer Perceptron (MLP) and Adaptive Resonance Theory 2 (ART2). It inherits two distinct advantages of stability and plasticity from ART2. Meanwhile, by combining the merits of MLP, it improves the performance for clustering. Experiment results show that our model can be used for clustering with promising performance.
The Role of Reputation in Ubiquitous Healthcare System
( Weiwei Yuan ),( Donghai Guan ),( Sungyoung Lee ) 한국정보처리학회 2007 한국정보처리학회 학술대회논문집 Vol.14 No.1
In this work, we analyze the role of reputation in ubiquitous healthcare system as well as the relationship of security, trust and reputation in this environment in details. In addition, an example is given to show how to use reputation system in ubiquitous healthcare and how to use reputation system on decision making.