• Journal of Information Technology Applications and Management
• /
• v.15 no.1
• /
• pp.243-260
• /
• 2008

To prevent low yields in the semiconductor industry is crucial to the success of that industry. However, to prevent low yields is difficult because of too many factors to affect yield variation and their complex relation in the semiconductor manufacturing process. This study presents a new efficient detection methodology for detecting abnormal yields including high and low yields, which can forecast the yield level of a production unit (namely a lot) based on yield-related feature variables' behaviors. In the methodology, we use C5.0 to identify the yield-related feature variables that are the combination of correlated process variables associated with yield, use SOM (Self-Organizing Map) neural networks to extract and classify significant patterns of past abnormal yield lots and finally use C5.0 to generate classification rules for detecting abnormal yield lot. We illustrate the effectiveness of our methodology using a semiconductor manufacturing company's field data.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.7
• /
• pp.633-640
• /
• 2001

In this paper we examine the machine learning issues raised by the domain of the Intrusion Detection Systems(IDS), which have difficulty successfully classifying intruders. There systems also require a significant amount of computational overhead making it difficult to create robust real-time IDS. Machine learning techniques can reduce the human effort required to build these systems and can improve their performance. Genetic algorithms are used to improve the performance of search problems, while data mining has been used for data analysis. Data Mining is the exploration and analysis of large quantities of data to discover meaningful patterns and rules. Among the tasks for data mining, we concentrate the classification task. Since classification is the basic element of human way of thinking, it is a well-studied problem in a wide variety of application. In this paper, we propose a classifier system based on genetic algorithm, and the proposed system is evaluated by applying it to IDS problem related to classification task in data mining. We report our experiments in using these method on KDD audit data.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.6
• /
• pp.494-501
• /
• 2001

In this paper an ANFIS(Adativo Neuro-Fuzzy Inference System)- based fault detection and diagnosis for a closed loop control system is proposed. The proposed diagnostic system contains two ANFIS. One is run as a parallel model within the model in closed loop control(MCL) and the other is run as a series-parallel model within the process in closed loop(PCL) for the generation of relevant symptoms for fault diagnosis. These symptoms are further processed by another classification logic with simple rules and neural network for process and controller fault diagnosis. Experimental results for a DC shunt motor control system illustrate the effectiveness of the proposed diagnostic scheme.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.6
• /
• pp.64-78
• /
• 1998

In this paper, on the basis of factory rationalization, the detection and evaluation system of weak assembly process was developed to analyze the processes for improvement of assembly productivity in the current assembly system. Using this detection and evaluation system of weak assembly process, the weak degrees of assembly process were quantitatively calculated. In this system, the improved design rules were constructed for assemblability and the redesign alternative was Presented for elimination of weak Process. After review of the redesign alternative, it was applied to the actual assembly system.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2002.05a
• /
• pp.278-285
• /
• 2002

본 논문에서는 인지적 긍정/부정 선택 과제의 수행 시 뇌파를 이용하여 피험자의 반응시간 RT를 감지하는 방법에 관하여 기술한다. 실험 Task에서 피험자는 시각적 자극에 대한 반응, 문제의 해석, 손 움직임의 조절, 손동작 등과 같은 작업을 수행한다. 이와 같은 상황에서의 피험자의 정신상태의 변화를 모델링하고, 선택시간 ST를 감지하여 피험자의 반응시간 RT를 예측한다. ST를 감지하기 위하여 측정한 뇌파로부터 $\alpha$, $\beta$, ${\gamma}$파를 분리하고, 4쌍의 전극들로부터 3가지의 특징들을 추출한다. 추출한 특징들을 분석하여 각 피험자별로 나타나는 상세 규칙과 공통적인 특성인 일반 규칙들을 설정하고 이들을 적용한다. 4명의 피험자를 대상으로 평균 81%의 ST 감지 성공률을 보이고, ST 감지 이후 약 0.73초에서 RT가 나타나는 것을 보인다. 본 논문에서 제안한 방법을 기존의 인지적인 정신상태 판별을 위한 방법들이나 왼손/오른손 동작구분 방법들과 결합하여 사용할 경우 BCI를 위한 기반 기술로 활용될 것으로 기대한다.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.9 no.1
• /
• pp.19-26
• /
• 2010

In this paper, we propose an efficient decision rule in order to get better chance to detect the unused spectrum assigned to a licensed user and improve reliability of spectrum sensing performance. Each secondary user receives the signals from the licensed user. And the resulting signals input to an energy detector. Then, each sensing result is combined and used to make a decision whether the primary user is present at the licensed spectrum band or not. In order to make the reliable decision, we apply an efficient decision rule that is called as a majority rule in this paper. The simulation results show that spectrum sensing performance with the proposed decision rule is more reasonable and efficient than that with conventional decision rules.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.3
• /
• pp.329-336
• /
• 2007

This paper considers a fault accommodation problem for inertial navigation systems (INS) that have redundant inertial sensors such as gyroscopes and accelerometers. It is wellknown that the more sensors are used, the smaller the navigation error of INS is, which means that the error covariance of the position estimate becomes less. Thus, when it is decided that double faults occur in the inertial sensors due to fault detection and isolation (FDI), it is necessary to decide whether the faulty sensors should be excluded or not. A new accommodation rule for double faults is proposed based on the error covariance of triad-solution of redundant inertial sensors, which is related to the navigation accuracy of INS. The proposed accommodation rule provides decision rules to determine which sensors should be excluded among faulty sensors. Monte Carlo simulation is performed for dodecahedron configuration, in which case the proposed accommodation rule can be drawn in the decision space of the two-dimensional Cartesian coordinate system.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.34C no.6
• /
• pp.79-88
• /
• 1997

New technologies, systems, and services in telecommunication have increased the need for an efficient and robust control mechanism to protect switching systems from overload. To achieve proper control, it is necessary to find a set of parameters that can describe the system. However, it is difficult to find types of data that can form a suitable basis for control. In this paper, we categorize the load status of a switching system into three classes (i.e., normal state, pre-overload state, and overload state) and formulate the overload detection as a classification problem. We find the relationships between the load classes and a set of monitored switching system parameters by applying a fuzzy rule-generation method. The rules are automatically generated from training data. Simulation results involving a switching system is given.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.159-161
• /
• 2003

Fitting the projected wire-frame model to the detected edge pixels on images by using least-squares approach, called Least-squares Model-image Fitting (LSMIF), is the key of the Model-based Building Extraction (MBBE). It is implemented by iteratively adjusting the model parameters to minimize the squares sum of distances from the extracted edge pixels to the projected wire-frame. This paper describes a series of experiments and studies on various factors affect the fitting results, including the edge detectors, the weighting rules, the initial value of parameters, and the number of overlapped images. The experimental result is not only helpful to clarify the influences of each factor, but is also able to enhance the robustness of the LSMIF algorithm.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.06a
• /
• pp.293-304
• /
• 2001

frauds detection is a difficult problem, requiring huge computer resources and complicated search activities. researchers have struggled with the problem. Even though a flew research approaches have claimed that their solution is much bettor than others, research community has not found 'the best solution'well fitting every fraud. Because of the evolving nature of the frauds, a Revel and self-adapting method should be devised. In this research a new approach is suggested to solving frauds in insurance claims and credit card transaction. Based on evolutionary computing approach, the method is itself self-adjusting and evolving enough to generate a new set of decision-making rules. We believe that this new approach will provide a promising alternative to conventional ones, in terms of computation performance and classification accuracy.