• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권4호
• /
• pp.254-268
• /
• 2013

For real-world clustering tasks, the input data is typically not easily separable due to the highly complex data structure or when clusters vary in size, density and shape. Kernel-based clustering has proven to be an effective approach to partition such data. In this paper, we provide an overview of several fuzzy kernel clustering algorithms. We focus on methods that optimize an fuzzy C-mean-type objective function. We highlight the advantages and disadvantages of each method. In addition to the completely unsupervised algorithms, we also provide an overview of some semi-supervised fuzzy kernel clustering algorithms. These algorithms use partial supervision information to guide the optimization process and avoid local minima. We also provide an overview of the different approaches that have been used to extend kernel clustering to handle very large data sets.

• ETRI Journal
• /
• 제36권6호
• /
• pp.1023-1031
• /
• 2014

Failure analysis is necessary to clarify the root cause of a failure, predict the next time a failure may occur, and improve the performance and reliability of a system. However, it is not an easy task to analyze and interpret failure data, especially for complex systems. Usually, these data are represented using many attributes, and sometimes they are inconsistent and ambiguous. In this paper, we present a scalable approach for the analysis and interpretation of failure data of high-performance computing systems. The approach employs rough sets theory (RST) for this task. The application of RST to a large publicly available set of failure data highlights the main attributes responsible for the root cause of a failure. In addition, it is used to analyze other failure characteristics, such as time between failures, repair times, workload running on a failed node, and failure category. Experimental results show the scalability of the presented approach and its ability to reveal dependencies among different failure characteristics.

• Asia pacific journal of information systems
• /
• 제25권1호
• /
• pp.25-60
• /
• 2015

Use cases are commonly used to represent customer requirements during systems development. In a large software development environment, finding relevant use cases from a library of past or related projects is a complex, error-prone and expensive task. This study proposes an ontological methodology to support use case retrieval in an interactive manner. The architecture of a prototype system that implements this methodology is presented. To evaluate whether the proposed approach can provide satisfactory results to users, this study develops a research model and hypotheses based on interaction theory. These hypotheses are empirically tested using a laboratory experiment which controls information filtering and perceived interaction. Our study suggests that a system which interacts with a user intelligently reduces cognitive load and increases self-efficacy and satisfaction.

• International Journal of Contents
• /
• 제15권2호
• /
• pp.38-43
• /
• 2019

Recently, bottom-up information systems (BUIS), developed according to the requirements of individual user departments, have become popular. However, effective management of BUIS projects is not enough, with many organizations having experienced integration challenges with such individual projects. BUIS projects are relatively small and limited in scope, as opposed to the large, complex systems developed through traditional top-down information system development projects. Due to these differences in characteristics, the control modes as well as the aspects to improve development performance in each type of project are also different. Therefore, it is difficult to apply the results of prior research on control in system development projects to improve BUIS project performance. The purpose of this study is to derive a new theory of control to improve BUIS project performance. The results contribute to the improvement of firm performance through effective control of BUIS projects in modern enterprises.

• 국제학술발표논문집
• /
• The 3th International Conference on Construction Engineering and Project Management
• /
• pp.890-896
• /
• 2009

In past projects, the main goal has been to enhance efficiency by reducing cost and time. However, considering the current condition of safety management in many construction companies, it can be confirmed that safety management has not been a top priority for a long time. Current safety management, which is based on safety standards and rules, is very ineffective and only emphasizes management after an incident. As well, although fewer accidents occur compared to the past, because construction projects are increasingly large in scale and complex, these accidents tend to be more serious and involve greater monetary loss. Furthermore, as the severity of these accidents increases, so does the possibility of fatalities. Therefore, improving safety management is essential. This study proposes an effective program for safety management, focusing on the processes to connect studies and systems, and the basic techniques required for program development. To realize this program, technical tools are suggested, including systems such as BIM (Building Information Modeling); additionally, the coordination of other systems such as an RTLS, a server, and an alarm system is proposed.

• 전자통신동향분석
• /
• 제38권6호
• /
• pp.22-30
• /
• 2023

Recently, artificial intelligence has been in the spotlight across various fields. Artificial intelligence uses massive amounts of data to train machine learning models and performs various tasks using the trained models. For model training, large, high-quality data sets are essential, and database systems have provided such data. Driven by advances in artificial intelligence, attempts are being made to improve various components of database systems using artificial intelligence. Replacing traditional complex algorithm-based database components with their artificial-intelligence-based counterparts can lead to substantial savings of resources and computation time, thereby improving the system performance and efficiency. We analyze trends in the application of artificial intelligence to database systems.

• Animal cells and systems
• /
• 제1권3호
• /
• pp.429-434
• /
• 1997

A new genus with a new species of poecilostomatoid copepod, Polyankylis orientalis, is described based on a pair of specimens collected from a tide pool on muddy shore in Korea. It represents a new family of the lichomolgoid complex with the following distinguishing characters: (1) 6-segmented antennule (resulted from the fusion of the third and fourth segments) with additional aesthetascs in male, (2) hook on mediodistal corner of antennal coxobasis (first segment), (3) proximal and distal scales on the outer margin of mandible present, (4) main axis of mandibular gnathobase short and lacking notches at the base of mandibular lash, and (5) inner margin of mandibular lash with denticulate processes. Additionally, maxillar syncoxa carries a large, pointed process on ventral surface, legs 1-4 do not show sexua1 dimorphism, and leg 4 is uniramous with 1-segmented exopod. A detailed discussion is given of its affinities with the ten existing families of the lichomolgoid complex.

• 한국통신학회논문지
• /
• 제21권1호
• /
• pp.291-299
• /
• 1996

이 논문에서는 다층 구조 고차 신경회로망을 이용하여 로봇 메니퓰레이터와 같이 관성이 크고 복잡한 특성을 갖는 시스템을 효과적으로 동정화하는 새로운 방법을 제안한다. 로봇 메니퓰레이터의 위치 및 속도와 미리 정해준 기준 점들 사이의 차이를 나타내는 특정 성능 지수 함수를 최소화하는 방법을 이용하여 동정화 과정에 필요한 신경회로망의 학습에 이용되는 입력 데이터를 설계하는 방법을 설명한다. 사람의 팔과 같이 비교적 큰 관성을 갖는 2축 로봇 매니퓰레이터를 이용한 컴퓨터 시뮬레이션으로부터 제안된 방법이 복잡한 특성을 갖는 시스템의 동정화에 필요한 입력 데이터를 효과적으로 설계할 수 있음을 보인다.

• 대한산업공학회지
• /
• 제24권2호
• /
• pp.185-197
• /
• 1998

Engineering design involves the specification of many variables that define a product, how it is made, and how it behaves. Before some variables can be determined, other variables must first be known or assumed. This fact implies a precedence order of the variables, and of the tasks of determining these variable consequently. Moreover, design of complex systems may involve a large number of design activities. In this paper, the activity-activity incidence matrix is considered as a representation of design activity analysis which mainly focuses on the precedence constraint. In order to analyze the activity-activity incidence matrix, a heuristic algorithm is proposed, which transforms an activity-activity, parameter-formula, and parameter-parameter incidence matrix into a lower triangular form. The analysis of the structured matrices can not only significantly reduce the overall project complexity by reorganizing few critical tasks in practice, but also aims at obtaining shorter times considering the solution structure by exploring concurrency.

• Journal of Advanced Marine Engineering and Technology
• /
• 제38권2호
• /
• pp.175-181
• /
• 2014

Location awareness is one of the key functions to build an U-city. Recently, many of works for the location-aware systems are emerging to be applied to on-going large-scale apartment complex. For example, when dwellers or cars with active tags are move in the apartment complex, the active tags broadcast their own identifiers periodically and receivers such that routers use this information to calculate the location of the active tags. Since the active tags can actually move between routers and coordinators, it requires a hand-off mechanism. In this paper, we investigate the hand-off problem occurred in location awareness system based on ZigBee active tags. We propose and analyze a new approach for handling the hand-off problem. Through a mathematical analysis, we show that our approach diminishes the quantity of packet transfer than ZigBee's approach.