• Journal of the Korean Statistical Society
• /
• 제36권3호
• /
• pp.435-445
• /
• 2007

This paper addresses for the first time the use of complete data information criteria in unsupervised learning of the Naive-Bayes model. A Monte Carlo study sets a large experimental design to assess these criteria, unusual in the Bayesian network literature. The simulation results show that complete data information criteria underperforms the Bayesian information criterion (BIC) for these Bayesian networks.

• 한국IT서비스학회지
• /
• 제16권4호
• /
• pp.161-176
• /
• 2017

Recently, most of the users can easily get access to a variety of information sources about companies, products, and services through online channels. Therefore, the online user evaluations are becoming the most powerful tool to generate word of mouth. The user's evaluation is provided in two forms, quantitative rating and review text. The rating is then divided into an overall rating and a detailed rating according to various evaluation criteria. However, since it is a burden for the reviewer to complete all required ratings for each evaluation criteria, so most of the sites requested only mandatory inputs for overall rating and optional inputs for other evaluation criteria. In fact, many users input only the ratings for some of the evaluation criteria and the percentage of missed ratings for each criteria is about 40%. As these missed ratings are the missing values in each criteria, the simple average calculation by ignoring the average 40% of the missed ratings can sufficiently distort the actual phenomenon. Therefore, in this study, we propose a methodology to predict the rating for the missed values of each criteria by analyzing user's evaluation information included the overall rating and text review for each criteria. The experiments were conducted on 207,968 evaluations collected from the actual hotel evaluation site. As a result, it was confirmed that the prediction accuracy of the detailed criteria ratings by the proposed methodology was much higher than the existing average-based method.

• Journal of Information Processing Systems
• /
• 제7권2호
• /
• pp.363-384
• /
• 2011

As the complexity of software is increasing, generating an effective test data has become a necessity. This necessity has increased the demand for techniques that can generate test data effectively. This paper proposes a test data generation technique based on adequacy based testing criteria. Adequacy based testing criteria uses the concept of mutation analysis to check the adequacy of test data. In general, mutation analysis is applied after the test data is generated. But, in this work, we propose a technique that applies mutation analysis at the time of test data generation only, rather than applying it after the test data has been generated. This saves significant amount of time (required to generate adequate test cases) as compared to the latter case as the total time in the latter case is the sum of the time to generate test data and the time to apply mutation analysis to the generated test data. We also use genetic algorithms that explore the complete domain of the program to provide near-global optimum solution. In this paper, we first define and explain the proposed technique. Then we validate the proposed technique using ten real time programs. The proposed technique is compared with path testing technique (that use reliability based testing criteria) for these ten programs. The results show that the adequacy based proposed technique is better than the reliability based path testing technique and there is a significant reduce in number of generated test cases and time taken to generate test cases.

• 대한산업공학회지
• /
• 제36권2호
• /
• pp.138-145
• /
• 2010

In this paper, we consider the information processing system, which organizes the collected data to meaningful information when the number of data collected from multiple sources reaches to a predetermined number, and performs any action by processing the collected data, or transmits to other devices or systems. We derive an analytical model to calculate the time until it takes to process information after starting to collect data. Therefore, in order to complete the processing data within certain time constraints, we develop some design criteria to control various parameters of the information processing system. Also, we analyze the discrete time model for packet switching networks considering data with no particular arrival nor drop pattern. We analyze the relationship between the number of required packets and average information processing time through numerical examples. By this, we show that the proposed model is able to design the system to be suitable for user's requirements being complementary the quality of information and the information processing time in the system with time constraints.

• 대한영양사협회학술지
• /
• 제8권1호
• /
• pp.1-8
• /
• 2002

The purpose of this study is to clarify success factors for desirable relationship between buyer and supplier in elementary school. Therefore, the survey questionnaire consisted of general background, past success, success difference, buyer-supplier relationship characteristics(trust, supporting status, communication behavior, conflict resolution techniques, supplier selection process). The subjects were 66 dieticians of elementary school in Inchon. The statistical analysis of data was completed using SPSS program. The results were summarized as follows : Average total cost/day per one person ￦1,156. The number of suppliers per one school were 6. 92.3% of the subjects were in favor of private contract, regarding contract methods of purchasing food materials. For the past success, degree of satisfaction about past their supplier showed 3.49 score. The present success difference was shown higher than the past success. There was significant correlation between the past success and the present success difference. Trust about suppliers showed 3.40score. Supplying companies hardly support for buying school. Among information quality(timely, accurate, adequate, complete, credible), timely and complete showed lower score than the other kind of elements. Among the buyer-supplier relationship characteristic elements, only trust correlated with satisfaction about suppliers significantly. Among the conflict resolution techniques, joint problem solving and persuasive attempts were often made use of by subjects. The supplier selection criteria were shown quality(7.47), supplier's capabilities(6.46), management plan(6.00), price(5.73), scale(5.48), assets(5.27), considers delivery(4.76) and technology(2.39). As results, trust was needed for the desirable relationship between buyers and suppliers. This study has some limitations. The data in this study were collected from only buyer. It is more desired to acquire data from suppliers also.

• 디지털산업정보학회논문지
• /
• 제14권1호
• /
• pp.87-101
• /
• 2018

The purpose of this study is to develop the measuring instruments for evaluation criteria for Malcolm Baldrige National Quality Award(MBNQA), suitable for ICT Industries, and to analyze the cause-effect relationship between those criteria through aforementioned instruments. MBNQA is formed with seven categories: Leadership, Strategic planning, Focus on patients, other customers and markets, Measurement, analysis and knowledge management, Human resource focus, Process management and Results. As excluding the Human Resource Focus category, this study empirically examined the cause-effect relationship among six categories. In order to empirically examine the research model, this study calculated Cronbach's alpha and reliability index, thus examined the reliability and executed Exploratory Factor Analysis. Furthermore, Average Variance Extracted(AVE) is used to verify the discriminant validity. Lastly, the hypothesis testing was made complete through significance test on the paths between variables. The result of this study shows that both leadership and social responsibility have direct cause-effect relationship with Measurement, analysis and knowledge management, Human resource focus, Process management and also that this relationship has direct impact on Human resource focus, Measurement, analysis and knowledge management as well, consequently exerting influence on the result through Process management, Finance and Market data.

• 대한공간정보학회지
• /
• 제17권4호
• /
• pp.23-29
• /
• 2009

최근 포털사이트 및 여러 분야에서 제공하는 지도서비스로 쉽게 공간정보를 이용할 수 있어 다양한 기관에서 제공하는 공간정보의 통합에 대한 연구가 주목을 받고 있다. 이에 따라 본 연구에서는 이종의 공간정보를 통합하는 연구를 진행하였다. 공간정보의 통합과정은 다른 제공기관에서의 공간정보 추출과 동일좌표계로의 변환 그리고 대응되는 공간객체쌍을 찾는 단계로 구성되며, 본 연구에서는 마지막 단계인 실세계의 대응되는 공간객체를 매칭하는 연구를 중심으로, 전자지도와 POI와 같은 이종의 공간정보를 효율적으로 통합하는 방법을 개발하였다. 이를 위해 네 개의 매칭 종류와 그에 따른 기준을 제시하였다. 본 연구의 주된 목적은 이종의 공간객체에 대한 매칭을 위해 공간정보 및 부가적인 속성정보를 이용하는 구체적인 방법과 절차를 제시하는 것이다.

• 제어로봇시스템학회논문지
• /
• 제20권6호
• /
• pp.689-694
• /
• 2014

An SPLC (Safety Programmable Logic Controller) must be designed to meet the highest safety standards, IEEE 1E, and should guarantee a level of fault-tolerance and high-reliability that ensures complete error-free operation. In order to satisfy these criteria, I/O modules, communication modules, processor modules and bus modules of the SPLC have been configured in triple or dual modular redundancy. The redundant modules receive the same data to determine the final data by the voting logic. Currently, the processor of each rx module performs the voting by deciding on the final data. It is the intent of this paper to prove the improvement on the current system, and develop a voting system for multiple data on a system bus level. The new system bus protocol is implemented based on a TCN-MVB that is a deterministic network consisting of a master-slave structure. The test result shows that the suggested system is better than the present system in view of its high utilization and improved performance of data exchange and voting.

• 유통과학연구
• /
• 제14권1호
• /
• pp.85-91
• /
• 2016

Purpose - Recently, the importance of rapid change in business models is more and more increasing as the change of information technology environment. Therefore, a variety of business models have emerged. On the other hand, there is no company that can generate revenue. Many enterprises are still maintained while they are changing only their appearance of the business model. Business model is important in e-commerce. However, a lot of researches are targeted only in Web sites. Thus, e-commerce companies do not have the infrastructure for measuring and business models. The purpose of paper is to evaluate factors which are related with the structuring of the e-commerce success. And it proposed a financial items and non-financial items. From the perspectives of administrators and managers, the paper researches the possibility for E-Commerce Evaluation Model as a valuable criteria in measuring business model. Research design, data and methodology - The methods are taken by the classification for the type of business-to-business transactions, transactions subject, and the degree of integration and innovation capabilities. Financial and Non-financial value is used to build E-Commerce Evaluation Model. Evaluation items in Administration's perspective are composed with enhance the effectiveness of the mission, improving efficiency of the administration, and control of costs. Evaluation items in the customer's perspective were measured by customer participation and cooperation with customer Satisfaction. In the case of researching the information system's perspective, three criteria are used such as adequacy of the development process, improvement of the quality of service, and maintenance of standardized information technology. In researching for the ICT competence's perspective, evaluation items were composed of enhanced user capabilities, utilizing new technologies, and empowerment of information workers. Results - In this paper, E-Commerce Evaluation Model with financial and non-financial perspectives shows the possibility to be criteria in the case of measuring business model. Moreover, it gives the positive expectation to be successful criteria. But the research may have ambiguity in its essential concept because it cannot avoid the limitation in selecting evaluation tools from merely the model. It is impossible to exclude the possibility in omitting specific properties which may take place in actual case study. Therefore, In hereafter research, it is necessary to include actual case study research in selecting evaluation tools in order to improve the limit point. Actual measurement items which are derived from actual case study should be subdivided, and it would be more effective to complete the research. Conclusions - In rapid change in business models, there are various kinds of business models. But it is general situation that companies which adopted business models have not brought in revenue. For this reason, E-Commerce Evaluation Model is needed as an important factor for the structuring of the e-commerce success. Although it has the limitation in selecting evaluation tools from model, E-Commerce Evaluation Model proposes the implication for measuring business models as a valuable criteria.

• The Journal of Advanced Prosthodontics
• /
• 제16권2호
• /
• pp.67-76
• /
• 2024

PURPOSE. This study aims to assess and predict lifespan of dental prostheses using newly developed Korean Association of Prosthodontics (KAP) criteria through a large-scale, multi-institutional survey. MATERIALS AND METHODS. Survey was conducted including 16 institutions. Cox proportional hazards model and principal component analysis (PCA) were used to find out relevant factors and predict life expectancy. RESULTS. 1,703 fixed and 815 removable prostheses data were collected and evaluated. Statistically significant factors in fixed prosthesis failure were plaque index and material type, with a median survival of 10 to 18 years and 14 to 20 years each. In removable prosthesis, factors were national health insurance coverage, antagonist type, and prosthesis type (complete or partial denture), with median survival of 10 to 13 years, 11 to 14 years, and 10 to 15 years each. For still-usable prostheses, PCA analysis predicted an additional 3 years in fixed and 4.8 years in removable prosthesis. CONCLUSION. Life expectancy of a prosthesis differed significantly by factors mostly controllable either by dentist or a patient. Overall life expectancy was shown to be longer than previous research.