• Journal of the Institute of Electronics Engineers of Korea TE
• /
• v.39 no.3
• /
• pp.97-104
• /
• 2002

In this paper we describe the method which optimizes the partition of the input space by means of measure of fuzziness for fuzzy neural network. It covers its generation of fuzzy rules for input sub space. It verifies the performance of the system depended on the various time interval of the input. This method divides the input space into several fuzzy regions and assigns a degree of each of the generated rules for the partitioned subspaces from the given data using the Shannon function and fuzzy entropy function generating the optimal knowledge base without the irrelevant rules. In this scheme the basic idea of the fuzzy neural network is to realize the fuzzy rule base and the process of reasoning by neural network and to make the corresponding parameters of the fuzzy control rules be adapted by the steepest descent algorithm. According to the input interval the proposed inference procedure proves that the fast convergence of root mean square error (RMSE) owes to the optimal partition of the input space

• Journal of the Korea Society of Computer and Information
• /
• v.6 no.1
• /
• pp.14-24
• /
• 2001

The huge and complicated plants such as nuclear power stations are likely to cause the operators to make mistakes due to a variety of inexplicable reasons and symptoms in case of emergency. Thats why the prevention system assisting the operators is being developed for. First of all. I suggest an improved fuzzy diagnosis. Secondly. I want to demonstrate that a classification system of nuclear plants accident investigating the causes of accidents foresees possible problems. and maintains the reliability of the diagnostic reports in spite of improper working in part. In the event of emergency in a nuclear plant, a lot of operational steps enable the operators to find out what caused the problems based on an emergent operating plan. Our system is able to classify their types within twenty to thirty seconds. As so, we expect the system to put don the accidents right after the rapid detection of the damage control-method concerned.

• Journal of Internet Computing and Services
• /
• v.16 no.6
• /
• pp.69-78
• /
• 2015

Recently, research on the multi-agent system is developed actively in the various fields, especially on the control of complex system and optimization. In this study, we develop a multi-agent system for NPC simulation in game environment. The purpose of the development is to support quick and precise decision by inferencing the situation of the dynamic discrete domain, and to support an optimization process of the agent system. Our approach employed Petri-net as a basic agent model to simplify structure of the system, and used fuzzy inference engine to support decision making in various situation. Our experimentation describes situation of the virtual battlefield between the NPCs, which are divided two groups, such as fuzzy rule based agent and automata based agent. We calculate the percentage of winning and survival rate from the several simulations, and the result describes that the fuzzy rule based agent showed better performance than the automata based agent.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.1
• /
• pp.240-259
• /
• 2020

Analyzing network traffic is the basis of dealing with network security issues. Most of the network security systems depend on the feature selection of network traffic data and the detection ability of malicious traffic in network can be improved by the correct method of feature selection. An FAFS method, which is short for Fuzzy Association Feature Selection method, is proposed in this paper for network malicious traffic detection. Association rules, which can reflect the relationship among different characteristic attributes of network traffic data, are mined by association analysis. The membership value of association rules are obtained by the calculation of fuzzy reasoning. The data features with the highest correlation intensity in network data sets are calculated by comparing the membership values in association rules. The dimension of data features are reduced and the detection ability of malicious traffic detection algorithm in network is improved by FAFS method. To verify the effect of malicious traffic feature selection by FAFS method, FAFS method is used to select data features of different dataset in this paper. Then, K-Nearest Neighbor algorithm, C4.5 Decision Tree algorithm and Naïve Bayes algorithm are used to test on the dataset above. Moreover, FAFS method is also compared with classical feature selection methods. The analysis of experimental results show that the precision and recall rate of malicious traffic detection in the network can be significantly improved by FAFS method, which provides a valuable reference for the establishment of network security system.

• Journal of Korea Multimedia Society
• /
• v.12 no.10
• /
• pp.1499-1507
• /
• 2009

With the abundance of digital contents, the necessity of precise indexing technology is consistently required. To meet these requirements, the intelligent software entity needs to be the subject of information retrieval and the interoperability among intelligent entities including human must be supported. In this paper, we analyze the unifying framework for multi-modality indexing that Snoek and Worring proposed. Our work investigates the method of improving the authenticity of indexing information in contents-based automated indexing techniques. It supports the creation and control of abstracted high-level indexing information through ontological concepts of Semantic Web skills. Moreover, it attempts to present the fundamental model that allows interoperability between human and machine and between machine and machine. The memory-residence model of processing ontology is inappropriate in order to take-in an enormous amount of indexing information. The use of ontology repository and inference engine is required for consistent retrieval and reasoning of logically expressed knowledge. Our work presents an experiment for storing and retrieving the designed knowledge by using the Minerva ontology repository, which demonstrates satisfied techniques and efficient requirements. At last, the efficient indexing possibility with related research is also considered.

• The KIPS Transactions:PartB
• /
• v.12B no.1 s.97
• /
• pp.91-100
• /
• 2005

It has been proposed diverse methods to use web information efficiently as the size of information is increasing. Most of search systems use a keyword-based method that mostly relies on syntactic information. They cannot utilize semantic information of documents and thus they could generate to users. To solve shortcoming in searching documents, a technique using the Semantic Web is suggested. A semantic web can find relevant information to users by employing metadata which are represented using standard ontologies. Each document is annotated with a metadata which can be reasoned by agents. In this paper, we propose a search system using semantic web technologies. Our semantic search system analyzes semantically questions that user input, and get resolution information that user want. To improve efficiency and accuracy of semantic search systems, this paper proposes DQL(DAML Query Language) engine that employs inference engine to execute reasoning and DQL converter that changes keyword form question of the user to DQL.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.1
• /
• pp.41-52
• /
• 2012

According to 2005 Korea National Health and Nutrition Survey, it has been reported that 32.9% men and 31.8% women have Metabolic syndrome among the population of age 30 and over. The importance of prevention and management is being emphasized in Metabolic syndrome which is a complex disease related to various generic and environmental factors like other chronical disease. In this study we suggest an service based on the data using the system architecture, ontology and Jena2.0 inference engine and organizing the disease-related guideline. The study also arrives at the result through proper interpretation and reasoning process using health management service model based on ontology. The accuracy according to the situation was tested and 930 data samples were selected and experimented. We drew a conclusion that the much personalized data is available, the more personalized services are possible. Since the risk factors of Metabolic syndrome are various, it would be effective to suggest customized services based on various personalized data.

• The Journal of Korean Association of Computer Education
• /
• v.23 no.2
• /
• pp.13-20
• /
• 2020

The interest in artificial intelligence is due to an increase in influence on companies, organizations, daily lives and society. The purpose of this study is to analyze the key elements in the teaching subjects of artificial intelligence-related subjects of Korean universities based on the intelligent system area of Computer Science 2013 in terms of human resources development. According to the analysis, there are five out of nine universities that run the required courses. Based on the 12 detailed knowledge domains of intelligent systems, the compulsory subjects of universities are distributed in the field of basic search theory, basic knowledge expression and reasoning, and inference based on uncertainty. The elective courses of each university covered topics in five to eight areas of the total knowledge area of the intelligent system, with 69.9 percent of universities with the highest average ratio of areas involving the subject of teaching subjects and 46.3 percent of universities with the lowest. This study has implications for the fact that prior to entering an artificial intelligence graduate school, we were able to grasp the level of knowledge about artificial intelligence at the undergraduate level.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.24 no.6
• /
• pp.1111-1116
• /
• 2010

We in this paper design and implement a traditional korean medicine diagnosis system based on basic ontology. If doctors put the symptoms or tongues or pulses of a patient in the diagnosis system, they can be recommended for the diagnosis results. To support the doctors decision, the diagnosis system make the inference based on the basic ontology and compute the similarity between symptoms of patient and those of ontology. The diagnosis systems also provide the learning mechanism about diagnosis results which save the results in the ontology and reuse them in the next diagnosis. Thus, doctors can share their knowledge for the diagnosis by exchanging their ontology each other. In future, we will expand the knowledge of the basic ontology continuously so that doctors can get the more accurate diagnosis results. We also implement the prescription function and integrate it to the diagnosis system.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.6
• /
• pp.765-777
• /
• 2002

For the battlefield analysis, it is required to get correct information about the identification and moving status of target enemy units. However, it is difficult for us to collect all of the information perfectly, because of the technology of communications, jamming, and tactics. Therefore, we need a reasoning function that predicts and analyzes future moving status for target units by using collected moving information and domain knowledge. Especially. since the moving units have characteristics of moving objects, which change their position and shape over time, they require functions to manage and predict locations of moving objects. Therefore, in this paper, we propose a location prediction system of moving units for battlefield analysis. The proposed system not only predicts unknown units, unidentified units, and main strike directions to application domain for battlefield analysis, but also estimates the past or future locations of moving objects not stored in a database.