• Journal of Intelligence and Information Systems
• /
• v.21 no.1
• /
• pp.29-45
• /
• 2015

Response modeling is a well-known research issue for those who have tried to get more superior performance in the capability of predicting the customers' response for the marketing promotion. The response model for customers would reduce the marketing cost by identifying prospective customers from very large customer database and predicting the purchasing intention of the selected customers while the promotion which is derived from an undifferentiated marketing strategy results in unnecessary cost. In addition, the big data environment has accelerated developing the response model with data mining techniques such as CBR, neural networks and support vector machines. And CBR is one of the most major tools in business because it is known as simple and robust to apply to the response model. However, CBR is an attractive data mining technique for data mining applications in business even though it hasn't shown high performance compared to other machine learning techniques. Thus many studies have tried to improve CBR and utilized in business data mining with the enhanced algorithms or the support of other techniques such as genetic algorithm, decision tree and AHP (Analytic Process Hierarchy). Ahn and Kim(2008) utilized logit, neural networks, CBR to predict that which customers would purchase the items promoted by marketing department and tried to optimized the number of k for k-nearest neighbor with genetic algorithm for the purpose of improving the performance of the integrated model. Hong and Park(2009) noted that the integrated approach with CBR for logit, neural networks, and Support Vector Machine (SVM) showed more improved prediction ability for response of customers to marketing promotion than each data mining models such as logit, neural networks, and SVM. This paper presented an approach to predict customers' response of marketing promotion with Case Based Reasoning. The proposed model was developed by applying different weights to each feature. We deployed logit model with a database including the promotion and the purchasing data of bath soap. After that, the coefficients were used to give different weights of CBR. We analyzed the performance of proposed weighted CBR based model compared to neural networks and pure CBR based model empirically and found that the proposed weighted CBR based model showed more superior performance than pure CBR model. Imbalanced data is a common problem to build data mining model to classify a class with real data such as bankruptcy prediction, intrusion detection, fraud detection, churn management, and response modeling. Imbalanced data means that the number of instance in one class is remarkably small or large compared to the number of instance in other classes. The classification model such as response modeling has a lot of trouble to recognize the pattern from data through learning because the model tends to ignore a small number of classes while classifying a large number of classes correctly. To resolve the problem caused from imbalanced data distribution, sampling method is one of the most representative approach. The sampling method could be categorized to under sampling and over sampling. However, CBR is not sensitive to data distribution because it doesn't learn from data unlike machine learning algorithm. In this study, we investigated the robustness of our proposed model while changing the ratio of response customers and nonresponse customers to the promotion program because the response customers for the suggested promotion is always a small part of nonresponse customers in the real world. We simulated the proposed model 100 times to validate the robustness with different ratio of response customers to response customers under the imbalanced data distribution. Finally, we found that our proposed CBR based model showed superior performance than compared models under the imbalanced data sets. Our study is expected to improve the performance of response model for the promotion program with CBR under imbalanced data distribution in the real world.

• Journal of Intelligence and Information Systems
• /
• v.21 no.3
• /
• pp.37-52
• /
• 2015

The recent global trends display expansion and growing solidity in both cooperative collaboration between industry, education, and research and R&D network systems. A greater support for the network and cooperative research sector would open greater possibilities for the evolution of new scholar and industrial fields and the development of new theories evoked from synergized educational research. Similarly, the national need for a strategy that can most efficiently and effectively support R&D network that are established through the government's R&D project research is on the rise. Despite the growing urgency, due to the habitual dependency on simple individual personal information data regarding R&D industry participants and generalized statistical data references, the policies concerning network system are disappointing and inadequate. Accordingly, analyses of the relationships involved for each subject who is participating in the R&D industry was conducted and on the foundation of an educational-industrial-research network system, possible changes within and of the network that may arise were predicted. To predict the R&D network transitions, Common Neighbor and Jaccard's Coefficient models were designated as the basic foundational models, upon which a new prediction model was proposed to address the limitations of the two aforementioned former models and to increase the accuracy of Link Prediction, with which a comparative analysis was made between the two models. Through the effective predictions regarding R&D network changes and transitions, such study result serves as a stepping-stone for an establishment of a prospective strategy that supports a desirable educational-industrial-research network and proposes a measure to promote the national policy to one that can effectively and efficiently sponsor integrated R&D industries. Though both weighted applications of Common Neighbor and Jaccard's Coefficient models provided positive outcomes, improved accuracy was comparatively more prevalent in the weighted Common Neighbor. An un-weighted Common Neighbor model predicted 650 out of 4,136 whereas a weighted Common Neighbor model predicted 50 more results at a total of 700 predictions. While the Jaccard's model demonstrated slight performance improvements in numeric terms, the differences were found to be insignificant.

• Journal of radiological science and technology
• /
• v.28 no.2
• /
• pp.91-95
• /
• 2005

The cervical spine of anteroposterior oblique view is essential to observe the intervertebral foramen(IVF). The morphologic changes of IVFs were proved to be abnormal with nerve roots and peripheral structures. The purpose of this study is to evaluate the effective projection angle for observing the IVFs in the Korean adults. In a prospective clinical study of 100 normal persons, $45^{\circ}$, $50^{\circ}$ and $55^{\circ}$ oblique views were compared by measuring the maximal transverse diameter of all the cervical IVFs. $45^{\circ}$ oblique views provided slightly better visualization of upper cervical level(C2-C3, C3-C4, C4-C5), but the lower cervical level(C5-C6, C6-C7, C7-T1) of IVF transverse diameters were substantially increased on the $55^{\circ}$ AP oblique projection. In the comprasion of mean differences between 8 obese person(BMI > 25) and 58 normal person(18.5 < BMI < 22.9) proved to be statistically not significant. Consequently this study shows that $55^{\circ}$AP oblique(tube angle $15^{\circ}$cephalad) view is optimal for evaluating the lower cervical IVFs.

• Korea journal of population studies
• /
• v.31 no.3
• /
• pp.57-84
• /
• 2008

This study investigates the patterns of leisure in Korea and the effects of social class on the objective and subjective dimension of leisure activities and life satisfaction. A data set of 1376 Korean men and women over 18 years old is analyzed to yield five main results. First, Korean prefers domestic entertainment to outdoor activities as is exemplified by domestic audio-visual entertainment(TV/DVD/VCR) which ranks the highest in the favored leisure activity. Leisure activities are divided into four types; "activity-based", "relationship-based", "alcohol-based", "relaxation". Second, the function of leisure activity is to strengthen relationships. The main purpose of leisure activity is to relax and revitalize, while creating prospective social network ranks next to relax. But the effect of leisure time is often compromised by recurring thoughts related to work. Third, respondents with high educational and economic backgrounds are more likely to engage in "relationship-based," "activity-based", "alcohol-based" leisure type. However, such factors do not influence on "relaxation" type of leisure. While students and housewives rank highest in number of respondents, respondents with managerial/professional or white-collar/semi-professional occupations enjoy more diverse activities. Fourth, the effort to discern the significance of social class with respect to the leisure-activity-index revealed followings; the index scores elevate with higher education, younger age and higher income. Fifth, leisure-activity-index is the most important variable predicting leisure satisfaction. Leisure satisfaction is influenced by gender, age, income and occupation. The younger the age and higher the income, the higher it is the leisure satisfaction. Men are more satisfied with leisure activities than women. Students experience the highest satisfaction with leisure activities while service/sales workers, industrial/technical/blue-collar workers shows the least satisfaction. Also, the number of family members decreases significantly the leisure satisfaction. While "activity-based" leisure induces the highest satisfaction, "alcohol-based" leisure produces the least satisfaction. The frequency and diversity of leisure activities, and "activity-based" leisure incur the most positive effects on the life satisfaction.

• Journal of Chest Surgery
• /
• v.31 no.9
• /
• pp.867-872
• /
• 1998

Background: The fixed dose regimen with activated coagulation time(ACT) is the most commonly employed method for determining the required dosage of heparin and protamine during cardiopulmonary bypass(CPB). Material and Method: We performed a prospective study on a fixed dose regimen for analyzing adequate dosages of heparin and protamine, the incidence of heparin resistance and heparin-induced thrombocyt openia, factors affecting ACT during CPB, and changes of ACT during aprotinin usage. 300 units/kg of heparin were administered to patients, and ACTs were measured after 5 mins. ACTs were checked at 10 mins and 30 mins after the onset of CPB, and then at 30 min intervals thereafter. If the measured ACT was under 400 secs, we added 100 units/kg of heparin. The heparin was reversed with 1 mg of protamine for each 100 units administered. If the measured ACT was longer than 130 secs 30 mins after protamine administration or if there was definitive evidence of a coagulation defect, we administered a further 0.5 mg/kg of protamine. Result: We studied 80 patients(50 adults and 30 children) who underwent open heart surgery(OHS) at Seoul National University Hospital. Preoperative ACT was 114.3${\pm}$19.3 secs in adults, and 119.5${\pm}$18.2 secs in children. There were no differences in preoperative ACT due to age, body weight, body surface area, or sex. The preoperative ACT was not influenced by a positive past history of OHS. Ten adults(20%) and 3 pediatric patients(10%) needed additional doses of heparin to maintain the ACT above 400 secs. Additional protamine administration was needed in 9 adults(18%) and 10 children(33%). Heparin resistance was found in only two adults. Heparin-induced thrombocytopenia was detected in 2 adults and 1 child. During CPB, ACT was prolonged. 12 adult patients received a low dose of aprotinin and showed longer celite activated ACT compared to the control group.The kaolin activated ACT showed a lower tendency than the celite activated ACT in aprotinin users. Conclusion: In conclusion, fixed dose regimen of heparin and protamine can be used without significant problems, but the incidence of need of additional dosage remains unsatisfactory.

• Journal of Chest Surgery
• /
• v.31 no.5
• /
• pp.456-465
• /
• 1998

Cardiopulmonary bypass(CPB) in children is associated with the accumulation of body water after cardiac operation, as a consequence of an inflammatory capillary leak. Following work by Elliott in 1991, modified ultrafiltration(MUF) was introduced after bypass as a means of hemoconcentrating patients and a potential way of removing water from the tissues. We have carried out a prospective randomized study of 20 children undergoing open heart surgery, comparing MUF with nonfiltered controls. MUF was carried out for a mean of 18.9 minutes after completion of CPB to a hematocrit of 37.1%(mean). The mean water volulme removed by the ultrafiltration was 38.4 ml/kg and the mean blood volume retransfused from the oxygenator during the ultrafiltration was 32.1 ml/kg. Fluid balance, hemodynamics, hematocrit, osmolarity and dosage of drug treatment were recorded for 4∼12 hours postoperatively. The results were analyzed using Student t-test and ANOVA, comparing controls(n=10) to MUF(n=10). Blood loss(ml/kg/24hr) was 14.5(mean) in MUF versus 13.7 in controls; blood transfused(ml/kg/24hr) 6.6 in MUF versus 15.2 in controls; plasma transfused(ml/kg/24hr) 65.7 in MUF versus 59.6 in controls. There was rise in arterial blood pressure and hematocrit during MUF. Percent rise of systolic blood pressure was 28.8% in MUF versus 18.7% in controls(p=0.366); percent rise of diastolic blood pressure was 28.8% in MUF versus 8.5% in controls(p=0.135); and percent rise of mean blood pressure was 36.2% in MUF versus 8.2% in controls (p=0.086). Percent rise of hematocrit was 40.0% in MUF versus 23.5% in controls(p=0.002). There was no significant difference in the inotropic requirement and the postoperative serum osmolarity between two groups. The number of days on the ventilator, the duration of stay in the intensive care unit, and the postoperative hospital stay were not significantly different between the two groups.

• Journal of Chest Surgery
• /
• v.40 no.2 s.271
• /
• pp.103-108
• /
• 2007

Background: Surgical resection is an important modality in the treatment of pulmonary metastases from various solid tumors. We analyzed 37 patients who underwent surgical treatments of pulmonary metastases in our hospital from 1996 to 2005. Material and Method: Age, sex, disease free interval, operative procedure, the number of pulmonary metastases, and lymphatic metastasis were investigated with admission and operative records, and pathologic reports. Actuarial survival and comparisons between each survival rate were calculated according to Kaplan-Meier method and log-rank test, respectively, Result: Complete resections were carried out in 34 of 37 patients. The primary tumor was carcinoma in 25 cases, sarcoma in 10, and others in 2. The number of pulmonary metastases was 1 in 25 cases and 2 or more in 12 cases. 3-year and 5-year survival rates after complete resection were 50.5% and 35.9%, respectively. 3-year and 5-year survival rates for carcinoma were 64.5% and 45.0%, respectively, and 3-year survival rate for sarcoma was 17.5%. Otherwise, none of the operative procedures, the number of pulmonary metastases, lymphatic metastasis, adjunctive therapy and the disease free interval in the case of carcinoma significantly affected the survival rates. Conclusion: Complete resection of pulmonary metastasis in well selected patients allows high long term survival rate with low mortality and morbidity. Long-term follow up and randomized prospective studies were necessary to determine the prognostic factors of pulmonary metastases after surgical resection.

• Journal of Chest Surgery
• /
• v.40 no.6 s.275
• /
• pp.428-434
• /
• 2007

Background: Recent studies have suggested that UFT may be an effective adjuvant therapy for completely resected IB (pT2N0) non-small cell lung cancer (NSCLC). We designed this study to clarify the feasibility of performing adjuvant chemotherapy with UFT for completely resected IB nor-small cell lung cancer, Material and Method: We randomly assigned patients suffering with completely resected IB non-small cell lung cancer to receive either UFT 3g for 2 year or they received no treatment. All patients had to be followed until death or the cut-off date (December 31 2006). Result: From June 2002 through December 2004, 64 patients were enrolled. Thirty five patients were assigned to receive UFT (the UFT group) and 29 patients were assigned to observation (the control group). A follow-up surrey on the 3 year survival rate was successfully completed for all the patients. The median follow-up time for all the patients was 32.8 months. In the UFT group, the median time of administration was 98 weeks (range: $2{\sim}129$ weeks). The rate of compliance was 88.2% at 6 months, 87.5% at 12 months, 80.6% at 18 month and 66.7% at 24 months. Seven recurrences (24.1%) occurred in the control group and six (17.1%) occurred in the UFT group (p=0,489). The three-year disease free survival rate was 71.3% for the control group and 82.0% for the UFT group (p=0.331). On the subgroup analysis, the three-year disease free survival rate for the patients with adenocacinoma was 45.0% for the control group and 75.2% for the UFT group (p=0.121). The three-year disease free survival rate for the patients with non-adenocarcinoma was 88.1% for the control group and 88.9% for the UFT group (p=0.964), Conclusion: Postoperative oral administration of UFT was well-tolerated. Adjuvant chemotherapy with UFT for completely resected pT2N0 adenocarcinoma of the lung could be expected to improve the disease free survival, but this failed to achieve statistical significance. A prospective randomized study for a large number of patients will be necessary.

• Radiation Oncology Journal
• /
• v.19 no.3
• /
• pp.224-229
• /
• 2001

Purpose : The aim of this study is to analyze the treatment failure patterns and the risk factors for locoregional or distant failure of uterine cervical carcinoma treated with radiation therapy. Materials and methods . A retrospective analysis was undertaken of 154 patients treated with curative radiation therapy in Gyeongsang National University Hospital from April 1989 through December 1997. According to FIGO classification, 12 patients were stage IB, 24 were IIA, 98 were IIB, 1 were IIIA, 17 were IIIB, 2 were IVA. Results : Overall treatment failure rate was $42.1\%$ (65/154), and that of complete responder was $31.5\%$ (41/130). Among 65 failures, 25 failed locoregionally, another 25 failed distantly, and 15 failed locoregionally and distantly. Multivariate analysis confirmed tumor size (>4 cm) as risk factor for locoregional failure, and tumor size (>4 cm), pelvic lymph node involvement as risk factors for distant failure. Conclusion : On the basis of results of our study and recent published data of prospective randomized study for locally advanced uterine cervical carcinoma, we concluded that uterine cervical carcinoma with size more than 4 cm or pelvic lymph node involvement should be treated with concurrent chemoradiation.

• Radiation Oncology Journal
• /
• v.29 no.1
• /
• pp.36-43
• /
• 2011

Purpose: To evaluate the effectiveness of radiation therapy (RT) for hepatocellular carcinoma (HCC) with portal vein tumor thrombosis (PVTT) and to analyze the prognostic factors. Materials and Methods: From December 2004 to April 2009, 70 patients who had HCC with PVTT were treated with RT at Keimyung University Dongsan Medical Center. Nineteen patients whose total dose was below 30 Gy and one patient who underwent liver transplantation were excluded. The remaining 50 patients (45 males, 5 females; median age 55 years) were analyzed. According to the LCSGJ TNM stage, there were 27 patients (54.0%) with stage III and 23 (46.0%) with stage IV. Total dose of 30~54 Gy was administered (median 45). Thirty patients (60.0%) were treated with concurrent chemoradiation therapy (CCRT). The median follow-up duration was from 13.5 months (range, 3 to 70 months). Results: The median survival time from the start of RT was 9 months. One-year and 2-year overall survival rates were 24.9% and 11.2%, respectively. At the follow-up time, three patients (6.0%) displayed no evidence of disease. Seven patients (14.0%) were alive with disease, and 40 (80.0%) patients had expired due to disease progression. CCRT was associated with worse survival than RT alone (p=0.034), Response to RT (p=0.037), CLIP stage (p=0.017), and TNM stage (p=0.041) were statistically significant prognostic factors. There was no radiation-induced liver disease. Conclusion: RT is an effective and safe modality for HCC with PVTT. Further studies such as prospective randomized trials are needed to confirm the role of RT for HCC with PVTT.