• Communications for Statistical Applications and Methods
• /
• v.18 no.6
• /
• pp.837-849
• /
• 2011

A self-organizing network is designed to estimate parameters of normal mixtures. SOMN achieves fast convergence and low possibility of divergence even when sample sizes are small, while PMLE eliminate unnecessary components. The proposed network effectively combines the good properties of SOMN and PMLE. Simulation verifies that the proposed network eliminates unnecessary components in normal mixtures when sample sizes are relatively small.

• Korean System Dynamics Review
• /
• v.9 no.1
• /
• pp.73-92
• /
• 2008

Internet portal sites have served customers for more than a decade in Korea. There have been several such sites have survived, succeeded and led the internet industry. What is the source of their competitiveness? We assumed that the network effect, which can vary dynamically, is an important factor; so we tried to build and analyze a system dynamics model on portal sites that represent their advertising business model. This model includes not only the network effect, but also customer's' expectations of the quality of portal sites, which also varies dynamically. As a result of our research, we can state that the network effect is important to stabilize and increase portal sites' quality, market share, income and cash flow. And we know that investment at a level above that which is merely required is necessary to preserve and stabilize their business.

• IE interfaces
• /
• v.13 no.4
• /
• pp.627-638
• /
• 2000

During last ten years, logistics network management has become one of the most important sources of competitive advantage regarding logistics cost and customer service in numerous business segments. The key to success in this dynamic and severe business environment is the ability to design and plan the logistics network optimally in an integrated way. Network design and planning involves many issues relating to strategic, tactical, and operational decisions. To assist a logistics designer or planner, many computer-based decision support systems (DSS) have been developed. In this paper, the issues related to design, development, and implementation of the DSS for logistics network design and planning are discussed, and an ideal framework of the DSS is proposed with some ideas on the future development. Finally, DLNET, the DSS developed by author, is briefly introduced.

• International Journal of Computer Science & Network Security
• /
• v.22 no.1
• /
• pp.55-60
• /
• 2022

Churn prediction is a critical long-term problem for many business like music, games, magazines etc. The churn probability can be used to study many aspects of a business including proactive customer marketing, sales prediction, and churn-sensitive pricing models. It is quite challenging to design machine learning model to predict the customer churn accurately due to the large volume of the time-series data and the temporal issues of the data. In this paper, a parallel artificial neural network is proposed to create a highly-accurate customer churn model on a large customer dataset. The proposed model has achieved significant improvement in the accuracy of churn prediction. The scalability and effectiveness of the proposed algorithm is also studied.

• Korean Business Review
• /
• v.19 no.1
• /
• pp.1-24
• /
• 2006

The purpose of this study is to compare social network characteristics of male managers and female managers. To achieve purpose of the research, data were collected in large Korean companies with more than 500 employees and 402 samples were restricted to middle managers. The results of the study are as follows. First, male managers have more strong instrumental network than female mangers while female mangers have more focused on expressive network than male mangers. Second, social network characteristics of male managers are smaller size but stronger intimacy and higher status than that of female mangers. This means that social network of male managers are more qualitative than that of female managers.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2008.10b
• /
• pp.479-491
• /
• 2008

A "task network" is a type of social network that consists of experts who exchange professional help and advice regarding executing tasks. In this study, we investigate the task network used within the IS department of a national bank in Korea. We identify how this network moderates the influence of computer-mediated communication (CMC) technology on an individual s task performance. Size, density, and centrality were measured as the characteristics of a personal task networks. Size equates to the total number of colleagues who work with a specific member for a certain project. Density is the ratio of the number of actual relationships to the total number of available relationships. Centrality defines whether an individual s position is in the exact center of whole network, and is measured by betweenness centrality, meaning the position one member holds between others in a network. Our findings conclude that the conditions - the larger the size of the task network, the smaller its density and the higher its level of centrality - lead to more benefits of using CMC media. Further, this positive effect of CMC is more noticeable when it provides synchronicity.

• Journal of Venture Innovation
• /
• v.1 no.2
• /
• pp.89-107
• /
• 2018

This study examined the success factor of small business CEOs for continuous growth and development of business, and found it competencies of CEOs, and empirically studied the relationship between this factor and business performance. This study also focused on mediating effect of networking activity between competencies of small business CEOs and business performance. As a result of the study, it turned out that leadership competence, external cooperation competence and marketing competence have a significant effects on business performance. Small business CEOs' own efforts for increasing leadership, cooperation and marketing capability are needed as well as government policy and assistance in order to improve business performance. It also tested the mediating effect of networking activity to get information and resources, and proved it plays a role in enhancing business performance, so that it is necessary to promote small business CEOs' access to information and resources. Based on these empirical tests, government and supporting organizations need to refer to that factors affecting business performance are in priority supported in small business policy.

• Journal of Service Research and Studies
• /
• v.1 no.1
• /
• pp.49-60
• /
• 2011

Recently, with the growing of social network sites, people's choice is also getting more and more. So the notion of loyalty has become an important construct within the Social Network framework because of member is easy switching on the social networking sites. Despite the increasing importance of social network sites loyalty question, there's very little research in this area. In electronic commerce, the website loyalty development process is based on both website satisfaction and website trust toward the net-enabled business. But how to target the members with high or low loyalty in the social network sites is still a question. In this paper we propose one improved RFM model to evaluate the member loyalty to find the potential members for improving the service quality of the social network site. In addition, an empirical case study is performed to demonstrate how this procedure works. Moreover, further applications of this research are provided for improved social network sites experiences and how to use the model to practice.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.11
• /
• pp.219-232
• /
• 2011

The purpose of this study is to give implications for cooperative network building for general corporate IT staffs as well as operational improvement of the IT organization which is a specialized unit in general company though analysis of characteristics of communication networks of the IT organization. 39 IT employees of company "A" were surveyed and interviewed, and the data collected were analyzed in a network analysis method. The analysis shows that a variety of networks were built across employees - e.g. strong/weak ties or centralized/decentralized structure - by each communication area. IT competency network was found weak ties. Career-rela ted network showed centralized and hierarchical characteristics. At the center of each network was positioned manager of each team. To address improvement of the findings, I presented the followings: more active networking over IT competency, enhancement of work centralization and collection of research data as to communication network of IT organization.

• Convergence Security Journal
• /
• v.19 no.3
• /
• pp.51-59
• /
• 2019

In this paper, we propose a device identification system for network visibility that can maintain the secure internal network environment in the IoT era. Recently, the area of enterprise network is getting huge and more complicated. Not only desktops and smartphones but also business pads, barcode scanners, APs, Video Surveillance, digital doors, security devices, and lots of Internet of Things (IoT) devices are rapidly pouring into the business network, and there are highly risk of security threats. Therefore, in this paper, we propose the device identification system that includes the process and module-specific functions to identify the exploding device in the IoT era. The proposed system provides in-depth visibility of the devices and their own vulnerabilities to the IT manager in company. These information help to mitigate the risk of the potential cyber security threats in the internal network and offer the unified security management against the business risks.