• Knowledge Management Research
• /
• v.24 no.2
• /
• pp.241-262
• /
• 2023

The phenomena and game records that occur in sports matches are being analyzed in the field of sports game analysis, utilizing advanced technologies and various scientific analysis methods. Among these methods, social network analysis is actively employed in analyzing pass networks. As football is a representative sport in which the game unfolds through player interactions, efforts are being made to provide new insights into the game using social network analysis, which were previously unattainable. Consequently, this study aims to analyze the changes in pass networks over time for a specific football team and compare them in different scenarios, including variations in the game's nature (Qatar World Cup games vs. A match games) and alterations in the opposing team (higher FIFA rankers vs. lower FIFA rankers). To elaborate, we selected ten matches from the games of the Korean national football team following Coach Bento's appointment, extracted network indicators for these matches, and applied four indicators (efficiency, cohesion, vulnerability, and activity/leadership) from a football team's performance evaluation model to the extracted data for analysis under different circumstances. The research findings revealed a significant increase in cohesion and a substantial decrease in vulnerability during the analysis of game performance over time. In the comparative analysis based on changes in the game's nature, Qatar World Cup matches exhibited superior performance across all aspects of the evaluation model compared to A matches. Lastly, in the comparative analysis considering the variations in the opposing team, matches against lower FIFA rankers displayed superior performance in all aspects of the evaluation model in comparison to matches against top FIFA rankers. We hope that the outcomes of this study can serve as essential foundational data for the selection of football team coaches and the development of game strategies, thereby contributing to the enhancement of the team's performance.

• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.251-273
• /
• 2022

Recently, research on applying text analysis to deep learning has steadily continued. In particular, researches have been actively conducted to understand the meaning of words and perform tasks such as summarization and sentiment classification through a pre-trained language model that learns large datasets. However, existing pre-trained language models show limitations in that they do not understand specific domains well. Therefore, in recent years, the flow of research has shifted toward creating a language model specialized for a particular domain. Domain-specific pre-trained language models allow the model to understand the knowledge of a particular domain better and reveal performance improvements on various tasks in the field. However, domain-specific further pre-training is expensive to acquire corpus data of the target domain. Furthermore, many cases have reported that performance improvement after further pre-training is insignificant in some domains. As such, it is difficult to decide to develop a domain-specific pre-trained language model, while it is not clear whether the performance will be improved dramatically. In this paper, we present a way to proactively check the expected performance improvement by further pre-training in a domain before actually performing further pre-training. Specifically, after selecting three domains, we measured the increase in classification accuracy through further pre-training in each domain. We also developed and presented new indicators to estimate the specificity of the domain based on the normalized frequency of the keywords used in each domain. Finally, we conducted classification using a pre-trained language model and a domain-specific pre-trained language model of three domains. As a result, we confirmed that the higher the domain specificity index, the higher the performance improvement through further pre-training.

• The Journal of Society for e-Business Studies
• /
• v.21 no.3
• /
• pp.51-78
• /
• 2016

Online-to-offline (O2O) commerce is the new trend that merges online commerce with traditional industries in various fields. The primary purpose of this paper is to find out which factors influence customers' intention to switch from call-based driver-for-hire services to O2O app-based services. This study used variables and factors based on Theory of Switching Intention, and Extended Unified Theory of Acceptance and Use of Technology in order to design research questions. We surveyed 500 users of call-based driver-for-hire services. According to the result of this study, dissatisfaction with the current call-based driver-for-hire services is estimated to be a significant factor that strengthens customers' intention to switch from the call-based driver-for-hire services to the app-based services. Loyalty to the previous call-based driver-for-hire services was not seen as a crucial motivator that causes customers to switch to the new O2O driver service. Switching cost also did not play a key role in explaining the relationship between dissatisfaction with the current call-based service and the intention to use the new app-based service. Performance expectancy, easiness in use, the level of user's knowledge or available assistance in relation to the use of app-based services, and expectancy for reasonable price was found to have meaningful impacts on customers' intention to switch from the call-based driver-for-hire services to the app-based services. Age, gender and user experience on the new service were found incapable of moderating the relationship between aforementioned factors which influence customers' choice of the app-based driver-for-hire service, and customers' intent to switch to the app-based service.

• The Korean Journal of Applied Statistics
• /
• v.22 no.2
• /
• pp.329-340
• /
• 2009

For complex diseases such as diabetes, hypertension, it is believed that model-free methods might work better because they do not require a precise knowledge of the mode of inheritance controlling the disease trait. This is done by estimating the sharing probabilities that a pair shares zero, one, or two alleles identical by descent(IBD) and has some specific branches of test procedure, i.e., the mean test, the proportion test, and the minmax test. Among them, the minmax test is known to be more robust than others regardless of genetic mode of inheritance in current use. In this study, we compared the power of the methods which are based on minmax test and considering weighting schemes for sib-pairs to analyze sibship data. In simulation result, we found that the method based on Suarez' was more powerful than any others without respect to marker allele frequency, genetic mode of inheritance, sibship size. Also, The power of both Suarez- and Hodge-based methods was higher when marker allele frequency and sibship size were higher, and this result was remarkable in dominant mode of inheritance especially.

• Journal of The Korean Association For Science Education
• /
• v.40 no.3
• /
• pp.253-269
• /
• 2020

This study aims to develop items to diagnose pre-service physics teachers' understanding of the conceptual knowledge of modern physics, based on the achievement criteria presented in the 2015 revised national science curriculum, and to identify the validity and reliability of the newly developed items. Data were collected from 467 pre-service physics teachers in the Physical Education Department or Science Education Department (Physics Education Major) of 15 universities across the nation. In this study the content validity, substantive validity, the internal structure validity, generalization validity, and the external validity proposed by Messick (1995) were examined by various statistical tests. The results of the MNSQ analysis showed that there was no nonconformity in the 23 items. The internal structure validity was confirmed by the standardized residual variance analysis, which shows that the 22 items was unidimensional. The generalization validity was confirmed by differential item functioning (DIF) analysis about groups lectured or not modern physics/quantum mechanics. In addition, item analysis and test analysis based on classical test theory were performed. The mean item difficulty is 0.66, mean item discrimination is 0.47 and mean point biserial coefficient obtained was 0.41. These results for item parameters satisfied the criteria respectively. The reliability of the internal consistency of the KR-20 is 0.77 and the Ferguson's delta obtained was δ = 0.972. By Rasch model analysis, the item difficulty (item measures) was discussed.

• Korean Journal of Construction Engineering and Management
• /
• v.18 no.2
• /
• pp.38-48
• /
• 2017

Finding causes of workers' unsafe behaviors is important to prevent construction accidents because 80 percent of accidents occur by workers' unsafe behaviors. In this regard, this research aims to investigate possible reasons of workers' unsafe behaviors based on workers' cognitive process model using System dynamics. This study is based on two ways of workers' cognitive process which are in relation to hazard perception and failure of hazard perception. Based on existing literature, causal loops for workers' cognitive process are developed to explain workers' habituation by staying out of accidents, safety learning by experience, failure of hazard perception, and attitude change by accidents. The interactions between the developed loops provide managerial insights to reduce workers' unsafe behaviors from a safety manager's perspective including increasing the probability of workers' hazard perception through knowledge management, maintaining workers' positive attitude toward safety, and controlling first-line supervisors to eliminate workers' unsafe behavior. The research allows us to better understand the causes and solutions of workers' unsafe behaviors in workers' cognitive perspectives.

• Journal of Engineering Education Research
• /
• v.9 no.4
• /
• pp.28-45
• /
• 2006

As society has changed to being more knowledge-based, it is indispensable that Internet usage is incorporated into education. Therefore, the e-learning system is being developed in order to provide a proper environment. However, many LCMS that were developed, currently are not based on SCORM, the world e-learning standard, nor KEM, the Korea Educational Metadata of Korea Education & Research Information Service(KERIS), and hence, it is difficult to share learning contents developed from such varied environments. National Science Museum, a non-educational institution, also provides the educational science exhibits in reality or in cyber space, which cannot be produced by elementary schools, and secondary schools. Consequently, new systems are necessary, whose modules should be divided into four, for example, 'teachers', 'learners', 'managers', and 'instructors', and be associated with each other so that they are able to integrate and manage such systems, and be used in school education as well. Therefore, in this study, more advanced LMS and LCMS, which are the web-portal sites used for a cyber science class at the National Science Museum, were designed and established. These sites were surely based on the KEM, and the SCORM.

• The Journal of the Korea Contents Association
• /
• v.16 no.11
• /
• pp.421-433
• /
• 2016

Recently, although the interest in transfer and transactions of intangible assets increases, there is no valuation model to objectively assess market value of knowledge and information contents such as electronic databases and the necessity of researches associated is brought up. Therefore, the present study proposes valuation models so as to utilize as objective reference information in the contents market of intangible assets, by assessing the market value of science and technology information contents including patents, academic papers and reports. First, we look into application methods of calculating cash flows by content types out of key variables which has been applied to the present technology valuation, and in case of patents we propose valuation methods based on concepts which are applied in the present technology valuation. Next, in case of both papers and reports, in order to reflect the characteristics of these contents we newly propose qualitative valuation methods which are adjustable based on both technology innovation and market demands indices while estimating the economic life cycle of the technology, and also present the input cost-based calculation method as the calculation method of cash flows. Throughout the study, we could establish frameworks by technology fields and business models applicable such as copyright licensing, transactions of individual science and technology information contents, and expect that more objective and reasonable assessment of content values is accessible.

• Information Systems Review
• /
• v.16 no.1
• /
• pp.89-106
• /
• 2014

The war simulation systems in Republic of Korea Army have been getting more and more important because soldiers can effectively and efficiently learn and share their war-related knowledge based on the interactions through the systems. But, up to now, the access to the war simulation systems has been limited to only soldiers. So, little research on them has been conducted. This study explores the effects of the antecedents to the flow experience and the characteristics of the systems on soldiers' intentions to use them. Based on the 118 samples collected from officers in Republic of Korea Army, this study empirically shows the logical reality of the war simulation systems and the flow experience positively influence soldiers' intentions to use the systems and the clarified goals, feedbacks, and the levels of the missions in the systems are significant antecedents to the flow experience. Useful implications are presented and discussed based on the new findings.

• Smart Media Journal
• /
• v.9 no.2
• /
• pp.51-62
• /
• 2020

Starting with the "Smart Grid National Road Map" in 2010, the Smart Grid 2030 was introduced through the basic plan and implementation plan of the intelligent power grid with the goal of building the world's first national smart grid. In this paper, we intend to build a power domain ontology based on the power business platform based on the upper and lower conceptual models of the "Smart Grid Interoperability Standard Framework and Roadmap", the standard of implementation plan. Ontology is suitable for expressing and utilizing the smart grid conceptual model because it considers hierarchical structure as knowledge defines the properties of entities and relationships between entities, but there is no research related to them. Therefore, in this paper, the upper ontology was defined as a major category for smart grid-related fields, and the lower ontology was defined as detailed systems and functions for the upper ontology to construct the ontology. In addition, scenarios in various situations that could occur in the power system were constructed and significant inference results were derived through inference engines and queries.