• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.215-229
• /
• 2020

The interest in machine learning is growing in all industries, but it is difficult to apply it to real-world tasks because of inexplicability. This paper introduces a case of developing a financial customer churn prediction model for a securities company, and introduces the research results on an attempt to develop a machine learning model that can be explained using the SHAP Value methodology and derivation of interpretability. In this study, a total of six customer churn models are compared and analyzed, and the cause of customer churn is inferred through the classification and data analysis of SHAP Value and the type of customer asset change. Based on the results of this study, it would be possible to use it as a basis for comprehensive judgment, such as using the Value of the deviation prediction result that can infer the cause of the marketing manager's actual customer marketing in the future and establishing a target marketing strategy for each customer.

• The Journal of the Korea Contents Association
• /
• v.10 no.11
• /
• pp.304-317
• /
• 2010

The study was performed to measure structural relationship among service value, brand image, and royalty of patients in regional public hospitals in Korea. Based on the analysis, the study aims to establish strategic direction of the regional public hospitals. Through a review of related literatures, measurement variables were identified, and a path model was developed for the study. Five regional public hospitals were chosen, and questionnaire has been collected from 387 outpatients and 358 inpatients. The study found that service value has a positive direct effect on brand image, and brand image has a positive direct effect on loyalty. Based on the findings, it is desirable for the regional public hospitals should mange the issues with lower level of perception and those with bigger effect, and establish low pricing strategy with improving the quality of service they provide.

• Communications for Statistical Applications and Methods
• /
• v.18 no.6
• /
• pp.687-696
• /
• 2011

Customer Lifetime or Customer Lifetime Value is a essential metric of differentiated CRM marketing and differentiated marketing strategy as a company core competency. However, customer lifetime used in companies is easily obtained from a confined simple customer attrition rate at some specific time point regardless of customer characteristics. In this study, in order to overcome the constraints of previous simple methods and to make practical use of it in industries, we suggest a method that estimates a customer lifetime using a customer segment based survival analysis with the censored data of customers; in addition, we apply this method to A mobile telecom company data. A method using customer segment based survival analysis is suggested in this study 1) includes all customers having different subscription dates, 2) reduces individual error, 3) can reflect trends after the observed time point and is more realistic.

• The Journal of Information Systems
• /
• v.20 no.3
• /
• pp.257-277
• /
• 2011

The current study categorizes factors of smartphone into three, using KANO model: attractive factors which cause only product satisfaction, must-be factors for dissatisfaction, and one-dimensional factors for both. Based on it, it presents a new model for the effects that smartphone factors have on satisfaction or dissatisfaction. The purpose is to theoretically explain that smartphone factors on which companies and users place a high value can actually affect satisfaction or dissatisfaction. After choosing 15 factors out of 25 which had been selected through literature study, these were divided into attractive, must-be, and one-dimensional ones. 93 out of 109 questionnaires returned were used for analysis. After frequency analysis using SPSS were conducted on the surveys, the factors were grouped, based on KANO table. The grouping results are as follows. Attractive factors include 'expansion slots for external memory, battery desorption, brand awareness, mobile banking and internet telephony'. Must-be ones include 'multi-touch, information security, entertainment, information retrieval, location based service and SNS. Finally, 'screen visibility, size of internal memory, the amount of internal memory, battery life, and response to after-sales service' are classified as one-dimensional factors. A critical finding of this paper is that since the results are different depending on the operating system of smartphones, it must be taken into consideration in studies on smartphones. The wide and rapid spread of smartphones has changed people's lifestyle as well as business environment, which forces companies to compete with each other to adapt to the changed circumstances. In this competitive system, studies on smartphone factors of satisfaction and dissatisfaction are essential for firms to establish a new strategy. From this point of view, the present paper is expected to be a basic material for enterprises not only to develop goods and services that maximize customer satisfaction and minimize dissatisfaction, but also to establish the future business strategy.

• Journal of Korea Multimedia Society
• /
• v.9 no.9
• /
• pp.1222-1230
• /
• 2006

Location-Based Service (LBS) is a service provided to the user based on the user's current geographic location. Since LBS provides a higher value-added service, LBS has been applied on various businesses, industries and even on personal lives. Positioning users is the essential technology in building an LBS system. Thanks to GPS (Global Positioning System), Positioning outdoor is successfully used in practice. However, there is not a general solution for indoor positioning yet, even though many strategies for indoor positioning have been introduced. One of the reasons for the lack of successful indoor positioning is that most of the existing indoor positioning strategies require special equipments dedicated for positioning. This paper introduces an indoor positioning strategy that does not require any additional equipments. Integrating our indoor positioning strategy with GPS-based outdoor positioning, we have implemented an indoor-outdoor positioning system. Experimental results of the system is also introduced.

• Journal of Korean Society of Archives and Records Management
• /
• v.18 no.4
• /
• pp.205-229
• /
• 2018

The purpose of this study is to present a plan to collect records for the post-management of the Pyeongchang Winter Olympics, which was held in February 2018. To achieve the research objectives, the value and types of records of the 2018 Winter Olympics follow-up records were identified, and the cases of mega sports event follow-up were selected and analyzed. Based on this, the following three measures were presented to collect records related to the Pyeongchang Winter Olympics. First, records were collected based on collection methods over time. Second, records were collected based on the object and subject of recording. Third, records were collected based on subjects related to the Pyeongchang Winter Olympics. The study's result can be used to establish a Pyeongchang Winter Olympics memorial hall and to prepare a collection plan for follow-up management after holding a domestic mega sports event.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.14 no.2
• /
• pp.73-83
• /
• 2014

Many recent theoretical developments in the field of machine learning and control have rapidly expanded its relevance to a wide variety of applications. In particular, a variety of portfolio optimization problems have recently been considered as a promising application domain for machine learning and control methods. In highly uncertain and stochastic environments, portfolio optimization can be formulated as optimal decision-making problems, and for these types of problems, approaches based on probabilistic machine learning and control methods are particularly pertinent. In this paper, we consider probabilistic machine learning and control based solutions to a couple of portfolio optimization problems. Simulation results show that these solutions work well when applied to real financial market data.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.3
• /
• pp.649-657
• /
• 2013

DTN flooding based routing protocol replicate the message copy to increase the delivery like hood that overloads the network resources. The probabilistic routing protocols reduce replication cost by forwarding the message to a node that holds high predictability value to meet its destination. However, the network traffic converges to high probable nodes and produce congestion that triggers the drop of previously stored messages. In this paper, we have proposed a routing protocol called as Connection frequency Buffer Aware Routing Protocol (CFBARP) that uses an adaptive method to maintain the information about the available buffer space at the receiver before message transmission. Furthermore, a frequency based method has been employed to determine the connection recurrence among nodes. The proposed strategy has performed well in terms of reducing message drop, message relay while increases the delivery probability.

• International Commerce and Information Review
• /
• v.9 no.2
• /
• pp.399-422
• /
• 2007

Like in other trade negotiations covering comprehensive sectors, it is reported that KOREA-US FTA tried so-called 'package deal' at the last minutes, when telecom services sector was positioned at the forefront and expected to partly play a role in buffering US' market-opening pressure on other sectors. Before everything else, Korean negotiators had to enhance the value of foreign ownership deregulation in telecom services sector as a leverage in the course of KOREA-US FTA negotiations. In addition, since foreign ownership change, if any, is highly sensitive issue either politically or policy-wise, it seems very difficult to find a breakthrough. Focusing on foreign ownership regulation in telecom services sector, this paper seeks how Korea has developed its strategic reasoning for negotiations and evaluates the outcomes of KOREA-US FTA negotiations.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.19 no.1
• /
• pp.153-168
• /
• 1994

This paper is concerned with the development of tree-search algorithm for the exact solution to the vehicle problem (VRP), where set of vehicles of known capacity based at depot, have to be routed in order to supply customers with known requirements. When is required is to design routes, so that the total cost (i. e. total route length or time duration, ect.) is minimized. For obtianing the exact solution, the most important factors are the value of bound and branching strategy. Using the bound based on with bound ascent procedures from subgradient and state-space ascents, the incorporation of bounds into tree search algorithm to solve the problem is shown. Computational results of the corresponding algorithm show that VRPs with up to 40 customers can be solved optimally with this algorithm.