• Journal of Intelligence and Information Systems
• /
• v.13 no.4
• /
• pp.17-34
• /
• 2007

Organizations have realized the importance of CRM. To obtain the maximum possible lifetime value from a customer base, it is critical that customer data is analyzed to understand patterns of customer response. As customer databases assume gigantic proportions due to Internet and e-commerce activity, data-mining-based market segmentation becomes crucial for understanding customers. Here we raise a question and some issues of using single SOM approach for clustering while proposing multiple self-organizing maps approach. This methodology exploits additional themes on the attributes that characterize customers in a typical CRM system. Since this additional theme is usually ignored by traditional market segmentation techniques we here suggest careful application of SOM for market segmentation.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.01a
• /
• pp.431-434
• /
• 2001

Knowledge discovery in databases(KDD) is the process for extracting valid, novel, potentially useful and understandable knowledge form real data. There are many academic and industrial activities with new technologies and application areas. Particularly, data mining is the core step in the KDD process, consisting of many algorithms to perform clustering, pattern recognition and rule induction functions. The main goal of these algorithms is prediction and description. Prediction means the assessment of unknown variables. Description is concerned with providing understandable results in a compatible format to human users. We introduce an efficient data mining algorithm considering predictive and descriptive capability. Reasonable pattern is derived from real world data by a revised neural network model and a proposed fuzzy rule extraction technique is applied to obtain understandable knowledge. The proposed neural network model is a hierarchical self-organizing system. The rule base is compatible to decision makers perception because the generated fuzzy rule set reflects the human information process. Results from real world application are analyzed to evaluate the system\`s performance.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.11
• /
• pp.2061-2065
• /
• 2008

In the conventional operation of ice-storage system based on operator's experience and judgement, the failure in forecast of cooling load occurs frequently due to operator's misjudgement and unskilled operation. This study presents the method of constructing self-organizing fuzzy models which forecast tomorrow temperature, humidity and cooling load periodically for economic and efficient operation of ice-storage system. To check the effectiveness and feasibility of the suggested algorithm, the actual example for forecasting temperature, humidity and cooling load of ice- storage system in KEPCO training institute, Sokcho, is examined. The computer simulation results show that the accuracy of temperature, humidity, cooling load forecast of the suggested algorithm is higher than that of the conventional methods.

• The Journal of Korea Robotics Society
• /
• v.3 no.2
• /
• pp.106-111
• /
• 2008

This paper presents a decentralized coordination for a small-scale mobile robot teams performing a task through cooperation. Robot teams are required to generate and maintain various geometric patterns adapting to an environment and/or a task in many cooperative applications. In particular, all robots must continue to strive toward achieving the team's mission even if some members fail to perform their role. Toward this end, given the number of robots in a team, an effective coordination is investigated for decentralized formation control strategies. Specifically, all members are required first to reach agreement on their coordinate system and have an identifier (ID) for role assignment in a self-organizing way. Then, employing IDs on individual robots within a common coordinate system, a decentralized neighbor-referenced formation control is realized to generate, keep, and switch between different geometric shapes. This approach is verified using an in-house simulator and physical mobile robots. We detail and evaluate the formation control approach, whose common features include self-organization, robustness, and flexibility.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.377-382
• /
• 2005

This paper presents a new method for the attitude control of planar space robots. In order to control highly constrained non-linear system such as a 3D space robot, the analytical formulation for the system with complex dynamics and effective control methodology based on the formulation, are not always obtainable. In the proposed method, correspondingly, a non-analytical but effective self-organizing modeling method for controlling a highly constrained system is proposed based on a polynomial data mining algorithm. In order to control the attitude of a planar space robot, it is well known to require inputs characterized by a special pattern in time series with a non-deterministic length. In order to correspond to this type of control paradigm, we adopt the Model Predictive Control (MPC) scheme where the length of the non-deterministic horizon is determined based on implementation cost and control performance. The optimal solution to finding the size of the input pattern is found by a solving two-stage programming problem.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.928-931
• /
• 2002

Network based intrusion detection system is a computer network security tool. In this paper, we present an intrusion detection system based on Self-Organizing Maps (SOM) and Resilient Propagation Neural Network (RPROP) for visualizing and classifying intrusion and normal patterns. We introduce a cluster matching equation for finding principal associated components in component planes. We apply data from The Third International Knowledge Discovery and Data Mining Tools Competition (KDD cup'99) for training and testing our prototype. From our experimental results with different network data, our scheme archives more than 90 percent detection rate, and less than 5 percent false alarm rate in one SYN flooding and two port scanning attack types.

• The Journal of the Korea Contents Association
• /
• v.10 no.3
• /
• pp.72-79
• /
• 2010

A EAARS(Electronic Attendance-Absence Recording System) is the important LSS(Learning Support System) for blending a on-line learning in the face-to-face classroom. However, the EAARS based on the smart card can not identify a real owner of the checked card. Therefore, we develop the CS(Client-Sever) system that manages the attendance-absence checking automatically, which is used the self-organizing neural network for the face recognition. A client system creates the ID file by extracting the face feature, a server system analyzes the ID file sent from client system, and performs a student identification by using the Recognized weight file saved in Database. As a result, The proposed CS EAARS shows the 92% efficiency in the CS environment that includes the various face image database of the real classroom.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2004.11a
• /
• pp.139-148
• /
• 2004

사용자의 지리적 위치에 따른 서비스를 제공하는 위치기반서비스는 유비쿼터스 컴퓨팅의 중요한 응용으로 여러 위치 감지기술과 다양한 시험 및 상용 서비스들이 개발되어 왔다. 하지만 기존의 위치기반서비스는 단순히 위치와 서비스를 정적으로 연결하는 기법에 그치고 있어 서비스의 유연성이 떨어지는 한계가 있다. 이를 개선하기 위해 위치 정보로부터 고수준 정보를 추론하여 보다 지능적인 서비스를 제공하려는 연구들이 이루어지고 있다. 본 논문에서는 사용자의 위치이동 데이터를 학습하여 미래의 위치 이동 경로를 예측하는 기법을 제안한다. GPS(Global Positioning System)를 사용하여 수집된 시퀸스 데이터를 시퀸스 데이터 처리에 특화된 RSOM (Recurrent Self Organizing Map)을 사용하여 클러스터링하고 이를 마르코브 모델을 사용하여 학습하여 각 위치 이동 패턴 모델을 구축한다. 현재의 위치이동 패턴을 구축된 각 이동패턴 모델들과 비교하여 가장 유사한 위치 이동패턴으로 미래의 사용자이동을 예측한다. 제안한 위치이동 예측 기법을 평가하기 위해 실제 대학생의 생활을 기반으로 하여 GPS 데이터를 대학 캠퍼스 상에서 수집하고 이를 이용하여 제안한 방법의 학습 및 예측 성능을 평가한다. 그 결과 제안한 방법을 사용하여 사용자의 미래의 위치이동경로를 예측하는 것이 가능하고 불확실한 상황에서도 유연하게 예측을 수행함을 확인하였다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.145-152
• /
• 1995

This paper presents a novel SOFLIC(self organizing fuzzy logic intelligent controller) for reactor rod control system in nuclear power plant. The output of fuzzy controller is gener ated by using two signal : the error between reference and average temperature, and the error between reference and neutron flux-converted temperatures. Flexibility of the controller is enhanced by using self-organizing feature and the controller respond to variation of system parameter with more precision. performances of the SOFLIC and PID are simulated with the model developed for a nuclear power plant. The SOFLIC is superior to PID : SOFLIC provides more rapid load following capability. more robustiness for variation in process dynamics and minimization of engineer's mistakes in controller design.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2003.05a
• /
• pp.400-409
• /
• 2003

Collaborative filtering (CF) recommendation is a knowledge sharing technology for distribution of opinions and facilitating contacts in network society between people with similar interests. The main concerns of the CF algorithm are about prediction accuracy, speed of response time, problem of data sparsity, and scalability. In general, the efforts of improving prediction algorithms and lessening response time are decoupled. We propose a three-step CF recommendation model which is composed of profiling, inferring, and predicting steps while considering prediction accuracy and computing speed simultaneously. This model combines a CF algorithm with two machine learning processes, SOM (Self-Organizing Map) and CBR (Case Based Reasoning) by changing an unsupervised clustering problem into a supervised user preference reasoning problem, which is a novel approach for the CF recommendation field. This paper demonstrates the utility of the CF recommendation based on SOM cluster-indexing CBR with validation against control algorithms through an open dataset of user preference.