혼합변수를 갖는 데이터의 분류를 위한 하이브리드 분류기

홍정식(Jung Sik Hong),박관호(Kwan Ho Park),박종률(Jong Ryul Park) 대한산업공학회 2017 대한산업공학회지 Vol.43 No.5

금융 차세대시스템 구축방식의 의사결정에 대한 연구

홍정식 ( Jung Sik Hong ),문희진 ( Hee Jin Moon ),조남욱 ( Nam Wook Cho ) LGCNS 엔트루정보기술연구소 2010 Entrue Journal of Information Technology Vol.9 No.1

최근 금융업계의 차세대시스템 구축이 본격화됨에 따라 구축방식의 의사결정에 대한 체계적인 접근법이 요구되고 있다. 본 논문에서는 금융 차세대시스템 구축 프로젝트 수행에 요구되는 소요시간과 투자비용에 따라 차세대시스템 구축 접근방법을 혁신적(Big Bang) 방식과 단계적(Phase) 방식으로 구분하고, 시스템 구축 단계에서 두 가지 대안을 체계적으로 비교하기 위하여 AHP(Analytical Hierarchy Process) 기법을 적용한 금융권 차세대시스템 구축방식 평가방법론을 제시하였다. 계층적 평가지표 도출을 위해 비용, 성과, 위험을 최상위 기준으로 설정하고 각각의 세부 평가 항목을 3단계까지 도출하였다. 도출된 평가 항목을 토대로 설문조사 대상자를 금융회사의 내부 현업 담당자와 외부 SI업계 전문가를 구분하여 금융 차세대시스템의 구축방식 선호도에 대한 비교분석을 수행하였다. 부석 결과 과거 국내 금융회사에서 많이 이루어진 혁신적 방식보다 단계적 방식에 대한 선호도가 높게 나타났다. 이러한 결과는 아키텍처 중심의 청사진을 통해 단계적으로 시스템을 구축하고자 하는 최근의 경향이 반영된 결과라 할 수 있다. 또한 회사유형에 따라 평가 항목별 가중치에 대한 차이가 나타났다. 즉, 금융회사는 시스템을 운영하는 입장에서 위험을 회피하고 안정성을 강조한 반면에 IT서비스회사는 차세대시스템을 구축하기 위한 비용과 그에 따른 성과를 강조하는 경향이 나타났다. 본 연구는 향후 금융 차세대시스템의 구축방법 결정시 가이드라인으로 활용될 것으로 기대된다. This study defines the Next Generation System(NGS) built by domestic financial businesses and classifies their architecture into two typical types according to the duration of the project and the relative cost of IT investment in the short term: the Big Bang approach and the Phase approach. Herein, we study the two approaches as alternatives in developing the Next Generation System, and derive the factors that are to be considered in the evaluation of the two alternatives for financial businesses. The set of standards for the choice between the two models are grouped into categories that constitute performance evaluation for IT-Cost, Performance and Risk. We drill down further into each category to second and third subordinate levels to derive detailed selection criteria. Based on the criteria drawn from the study, we conduct a survey with information system planners, IT managers and specialists at financial companies who are currently planning, developing or have completed a NGS. Results are analyzed using the AHP methodology to compare and understand the different approach in the implementation of NGS for financial business. According to the research result, survey respondents showed the different evaluation weight between alternatives by industry, but the Phase approach showed higher preference than the Big Bang approach in the building method.

마코프 에러형태 하에서의 Blast 프로토콜의 수율 분석

홍정식(Jung Sik Hong),홍정완(Jung Wan Hong),이창훈(Chang Hoon Lie) 대한산업공학회 1997 대한산업공학회지 Vol.23 No.4

국내 IT 산업 부문 간 효율성 비교 평가: 자료포락분석(DEA) 기법을 중심으로

홍정식 ( Jung Sik Hong ),양창준 ( Chang Joon Yang ),이학연 ( Hak Yeon Lee ) 한국경영공학회 2012 한국경영공학회지 Vol.17 No.1

This study measures and compare efficiency of the eight Korean IT sectors. Data envelopment analysis (DEA) is employed to measure efficiency of about 900 firms with annual sales over 10 billions won in the following eight sectors: telecommunication service, wireless communication service, digital contents, broadcasting service, wireless communication device, semiconductor, display panel, and computer-related service. DEA is a linear programming model for measuring the relative efficiency of decision making units (DMUs) with multiple inputs and outputs. Total number of employees, amount of investment, and cost of sales are considered as inputs, and output variables include value added, net income, and sales. Every firm in every sector is evaluated together by DEA, and then, Kruskal-Wallis test is run to compare the efficiency of the sectors. It is found that wireless communication service and telecommunication service have relatively high efficiency while digital contents, wireless communication device, and semiconductor show low performance. The results of efficiency evaluation and benchmarking way for improving efficiency obtained from this study are expected to provide fruitful implications for IT policy making.

Bass 확산모형의 이분 확장

홍정식(Jung-Sik Hong),엄석준(Seok-Jun Eom) 한국경영과학회 2009 韓國經營科學會誌 Vol.34 No.4

Bass diffusion model have played a central role in studying the diffusion of the new products since 1969, the year of publication of Bass model. Almost 750 publications based on the Bass diffusion model have explored extensions and applications. Extension models can be divided into two types. One is the model containing marketing-mix variables and the other is the model containing additional parameters. This paper presents another extension model of the latter type. Our model allows the time varying coefficients of innovation and imitation. Two pieces approximation of time varying coefficients is introduced and it's parameters are estimated based on NLS(Non-Linear Mean Square) method. Empirical studies are performed and the results show that our model is superior to the basic Bass model and the NUI(Non-Uniform Influence) model which is the well-known extension of the Bass model. The model developed in this paper is, also, transformed into the Bass model with the ready potential adopters in order to enhance the descriptive power.

무방향 네트워크의 K-터미날 신뢰성 계산을 위한 K4-chain 축소

홍정식(Hong, Jung Sik),이창훈(Lie, Chang Hoon) 대한산업공학회 1989 대한산업공학회 춘계학술대회논문집 Vol.1989 No.5

대표적인 의사결정나무 알고리즘의 해석력 비교

홍정식(Jung-Sik Hong),황근성(Geun-Seong Hwang) 한국산업경영시스템학회 2021 한국산업경영시스템학회지 Vol.44 No.2

Most of the open-source decision tree algorithms are based on three splitting criteria (Entropy, Gini Index, and Gain Ratio). Therefore, the advantages and disadvantages of these three popular algorithms need to be studied more thoroughly. Comparisons of the three algorithms were mainly performed with respect to the predictive performance. In this work, we conducted a comparative experiment on the splitting criteria of three decision trees, focusing on their interpretability. Depth, homogeneity, coverage, lift, and stability were used as indicators for measuring interpretability. To measure the stability of decision trees, we present a measure of the stability of the root node and the stability of the dominating rules based on a measure of the similarity of trees. Based on 10 data collected from UCI and Kaggle, we compare the interpretability of DT (Decision Tree) algorithms based on three splitting criteria. The results show that the GR (Gain Ratio) branch-based DT algorithm performs well in terms of lift and homogeneity, while the GINI (Gini Index) and ENT (Entropy) branch-based DT algorithms performs well in terms of coverage. With respect to stability, considering both the similarity of the dominating rule or the similarity of the root node, the DT algorithm according to the ENT splitting criterion shows the best results.

확산 모형에 의한 고가 의료기기의 수요 확산의 특성분석 및 중장기 수요예측에 관한 연구

홍정식 ( Jung Sik Hong ),김태구 ( Tae Gu Kim ),임달오 ( Dar Oh Lim ) 한국보건행정학회 2008 보건행정학회지 Vol.18 No.4

In this study, we explore the long-term demand forecasting of high-price medical equipments based on logistic and Bass diffusion model. We analyze the specific pattern of each equipment`s diffusion curve by interpreting the parameter estimates of Bass diffusion model. Our findings are that (1) Ultrasonic imaging system, CT are in the stage of maturity and so, the future demands of them are not too large. (2) Medical image processing unit is between growth stage and maturity stage and so, the demand is expected to increase considerably for two or three years. (3) MRI is in the stage of take-off and Mammmography X-ray system is in the stage of maturity but, estimates of the potential number of adopters based on logistic model is considerably different to that based on Bass diffusion model. It means that additional data for these two equipments should be collected and analyzed to obtain the reliable estimates of their demands. (4) Medical image processing unit have the largest q value. It means that the word-of-mouth effect is important in the diffusion of this equipment. (5) For MRI and Ultrasonic system, q/p values have the relatively large value. It means that collective power has an important role in adopting these two equipments.

자동차 부품 수요의 예측 모형 개발

홍정식(Jung-Sik Hong),안재경(Jae-Kyung Ahn),홍석기(Suk-Kee Hong) 대한산업공학회 2001 대한산업공학회지 Vol.27 No.3

신제품 및 서비스에 있어 이분조각 Bass모형에 의한 확산 패턴 분석

홍석기(Seok-Kee Hong),홍정식(Jung-Sik Hong) 대한산업공학회 2010 산업공학 Vol.23 No.4

The Bass model is the most widely used model in research of new product diffusion because it presents a nice explanation on the diffusion process of new products. However, it has a limitation that its performance of fitness is lower as the available data become less and also, the diffusion curve is bell-shape and so, it can not represent the various diffusion patterns. Recently, a two-pieces Bass model is developed and applied to analyze diffusion of 10 products. The results are encouraging in terms of fitness. However, diffusion pattern is not dealt with in the paper. In this paper, analysis of diffusion pattern is in depth addressed in two-pieces Bass model. It is shown that the diffusion curves are divided into 3 types with respect to the peak adoption rate and each type is divided into 2 types further. Takeoff time of a diffusion process is analyzed by using the inflection point and regime-change time where it represents the point that imitation and innovation parameters change. Empirical studies for 68 products(28 domestic products and 40 USA products) are performed to analyze the diffusion pattern. Findings are that diffusion patterns of all products except 1 USA product show typeⅠ and regime-change time becomes shorter as the introduction time of the product is later in domestic products and regime-change time can be regarded as a takeoff time in 47% of total 68 products.