• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.95-101
• /
• 1999

In this paper, a comparison between different methods to combine predictions from neural networks will be given. These methods are bagging, bumping, and balancing. Those are based on the analysis of the ensemble generalization error into an ambiguity term and a term incorporating generalization performances of individual networks. Neural Networks and AI machine learning models are prone to overfitting. A strategy to prevent a neural network from overfitting, is to stop training in early stage of the learning process. The complete data set is spilt up into a training set and a validation set. Training is stopped when the error on the validation set starts increasing. The stability of the networks is highly dependent on the division in training and validation set, and also on the random initial weights and the chosen minimization procedure. This causes early stopped networks to be rather unstable: a small change in the data or different initial conditions can produce large changes in the prediction. Therefore, it is advisable to apply the same procedure several times starting from different initial weights. This technique is often referred to as training ensembles of neural networks. In this paper, we presented a comparison of three statistical methods to prevent overfitting of neural network.

• Journal of Intelligence and Information Systems
• /
• v.15 no.4
• /
• pp.37-51
• /
• 2009

One of the critical issues in the management of manufacturing companies is the efficient process of planning and operating service resources such as human, parts, and facilities, and it begins with the accurate service demand forecasting. In this research, service and sales data from the LCD monitor manufacturer is considered for an empirical study on Product Life Cycle (PLC) based service demand forecasting. The proposed PLC forecasting approach consists of four steps : understanding the basic statistics of data, clustering models using a self-organizing map, developing respective forecasting models for each segment, comparing the accuracy performance. Empirical experiments show that the PLC approach outperformed the traditional approaches in terms of root mean square error and mean absolute percentage error.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.4
• /
• pp.59-70
• /
• 2020

The purpose of this study is to construct the ACD model for the block trading volume duration. The ACD model based on the block trading volume duration is referred to as Volume ACD (VACD) in this study. By integrating with GARCH-type models, the VACD based GARCH type models, which include VACD-GARCH, VACD-IGARCH and VACD-FIGARCH models, are set up. This study selects Chunghwa Telecom (CHT) Inc., offering the America Depository Receipt (ADR) in NYSE, to investigate the block trading volume duration in Taiwanese equity market. The empirical results indicate that the long memory in volume duration series increases dependence at level of volatility clustering by VACD (2,1)-FIGARCH (3,d,1) model. Moreover, the VACD (2,1)-IGARCH (1,1) exhibits relatively better performance of prediction on capturing block trading volume duration. This volatility model is more appropriate in this study to portray the change of the CHT Inc. prices and provides more information about the volatility process for investment strategy, which can be a reference indicator of financial asset pricing, hedging strategy and risk management.

• Proceedings of the Korean System Dynamics Society
• /
• 2002.11a
• /
• pp.59-100
• /
• 2002

This study examines the effect that the prepaid trust has in saving the troubled business through the development of a formal system dynamics model. Building on existing case studies of corporate turnaround. we develop a mathematical model on the relationship between the prepaid trust and turnaround. With it. we build a system dynamics model in order to show how the prepaid trust can shift an organizational system from fragile condition of a firm to resilient condition of a firm. The model, which is oriented by attribution theory and problem-sharing by crisis communication. is developed in the aspect of the process from 'clouds' to 'evaporating clouds' This study offers a new characterization of the conditions under which organizations may be recovered the performance by prepaid trust through simulations of the model, and provides both theoretical insights into the corporate turnaround and practical suggestions for those charged with corporate turnaround.

• 한국IT서비스학회:학술대회논문집
• /
• 2009.05a
• /
• pp.415-427
• /
• 2009

Diffusion of information technology such as internet has brought a great influence on modern society and industrial development. Currently wide-spread RFID/USN technology has become a central momentum of this rapid changes for the establishment of ubiquitous society in the future. Historically agricultural products has drawn research interest due to its importance of human life. A standard monitoring system throughout of life cycle of the agricultural products should be developed and careful monitoring process should be introduced. In this paper, by way of case study, current lifetime tracing systems are evaluated and critical drawbacks and issues are also developed. In addition, some implications and guidelines to improve the tracing performance of the products are also suggested.

• Korea Journal of Hospital Management
• /
• v.22 no.3
• /
• pp.88-104
• /
• 2017

The purpose of this study explains how service characteristics such as service competence, service customization, interpersonal communication and image affect service trust and affective commitments and how service trust, affective commitment play a role in affecting risk-taking and service loyalty in specialty hospital. and then confirm the mediating role of relationship quality components such as service trust and affective commitment in the above path relation. For this purpose, the research hypothesis was verified by structural equation model analysis(SEM) using SPSS 21.0 and AMOS 20.0 statistical package. The results of the study are as follows. First, interpersonal communication influenced service trust, image influenced affective commitment, service ability influenced service trust and emotional commitment, while service customization did not affect any factor. Second, service trust and affective commitment did not effect each other. Third, service trust and affective commitment were more powerful in risk-taking than the influence on service loyalty. Fourth, Risk takings affected service loyalty. Therefore, marketers of special hospitals not only medical services by providing medical service with superior service ability, but also by always strengthening interpersonal communication at the time of providing services. This will enable consumers to overcome the avoidance in the service decision process and implement a specific service marketing strategy that can continuously use the hospital service.

• Journal of the Korea Society of Computer and Information
• /
• v.8 no.3
• /
• pp.16-25
• /
• 2003

Modeller is defined as the office work procedures(business processes), and that is systemically based upon the real-time collaborative operations by a set of actors, which is called group. This paper describes the design of the ICN editor that is operable under the real-time collaborative computing environment. We use the database agent that enables the ICN editor to operate among multiple actors(group) through the event-driven collaboration platform. Consequently, a set of workflow and business processes defined through this editor is not only stored onto database but also transformed into the format of the workflow process definition language(WPDL) that is a standardized worflow description and specification language proposed by the workflow management coalition(WfMC). This method can improve the performance of workflow processing by minimizing the workflow execution cost occurred during workflow definition.

• International Journal of Quality Innovation
• /
• v.8 no.1
• /
• pp.99-119
• /
• 2007

This research probes into the execution of small and medium-sized enterprises' value creativities by a difference analysis with different classifications, different capital, different turnover, different employees, and different established years. This study develop a questionnaire about value creativity with five dimensions and thirty-five items according to "Valuation" by McKinsey and Company, Inc. and Copeland et al., such as: "Aspiration and target," "Portfolio management," "Organization design," "Process management," and "Business and individual performance management." The results are as follows: (1) Most small and medium-sized enterprises (SMEs) have executed value creativities; (2) There is a difference in the execution of value creativities between the livelihood industry and the chemical industry; the execution of value creativities by livelihood industry is better than the chemical industry; (3) For value creativities of the execution of different capital and turnover for SMEs, bigger entities are better than smaller ones; (4) For the value creativities of the execution of different numbers of staff in SMEs, those with more staff are better than those with fewer staff; (5) For the value creativities of the execution of different established years for SMEs, those established longer are better than those established shorter.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.41 no.4
• /
• pp.101-110
• /
• 2018

This study aims to objectively measure the efficiency of nanotechnology R&D programs by systematically evaluating the inputs and outputs of nanotechnology R&D activities and to find implications for improving the efficiency of nanotechnology R&D programs. Data on input factors such as R&D investment, R&D manpower, R&D period, and output factors such as paper, patent, and commercialization for R&D projects which started from 2008 or afterwards and ended by 2011 are gathered through National Science and Technology Knowledge Information Service, which are used for efficiency evaluation. In this study, we analyzed R&D efficiency in detailed technology units in depth. The process taken in this study is as follows. First, the basic statistics of input and output factors to compare and analyze R&D investment, R&D manpower, R&D period, paper, patent, and commercialization status by technology unit are analyzed. Next, DEA models are utilized to derive the overall efficiency, pure technology efficiency, and scale efficiency by conducting the efficiency evaluation for each technology unit, from which implications for strategic budget allocation are derived. In addition, partial efficiency evaluation is conducted to identify advantages and disadvantages of each technology unit. In turn, cluster analysis is performed to identify similar technology units, from which implications for efficiency improvement are derived.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.3
• /
• pp.7-14
• /
• 2023

Recently, the importance of impact-based forecasting has increased along with the socio-economic impact of severe weather have emerged. As news articles contain unconstructed information closely related to the people's life, this study developed and evaluated a binary classification algorithm about snowfall damage information by using media articles text mining. We collected news articles during 2009 to 2021 which containing 'heavy snow' in its body context and labelled whether each article correspond to specific damage fields such as car accident. To develop a classifier, we proposed a probability-based classifier based on the ratio of the two conditional probabilities, which is defined as I/O Ratio in this study. During the construction process, we also adopted the n-gram approach to consider contextual meaning of each keyword. The accuracy of the classifier was 75%, supporting the possibility of application of news big data to the impact-based forecasting. We expect the performance of the classifier will be improve in the further research as the various training data is accumulated. The result of this study can be readily expanded by applying the same methodology to other disasters in the future. Furthermore, the result of this study can reduce social and economic damage of high impact weather by supporting the establishment of an integrated meteorological decision support system.