GENERALIZATION OF THE HYBRID LOGISTIC MODEL FOR MORE THAN ONE RARE RISK FACTOR

Mamunur Rashid 한국데이터정보과학회 2008 한국데이터정보과학회지 Vol.19 No.1

Logistic models are commonly used to analyze case-control data. For case-control studies, if there tends be rare disease in the control group with the risk factors, then the estimation procedure using logistic regression model for such factors becomes difficult. To overcomesuch situation, Chen et. al. (2003) proposed a hybrid logistic model in which they first estimate the problematic risk factor assuming that proportions having disease of such risk factor are treated equal for all permissible strata of the other risk factors, and then the residual of the risk factors are modeled by using the logistic regression model. The purpose of this paper is to extend theoretically the hybrid logistic model for case-control studies for more than one rare risk factor.

Fast incremental learning of logistic model tree using least angle regression

Lee, Sudong,Jun, Chi-Hyuck Elsevier 2018 expert systems with applications Vol.97 No.-

<P><B>Abstract</B></P> <P>Expert and intelligent systems understand the underlying information behind the data by relying on a wide range of machine learning techniques. The interpretation of machine learning models is often the key to success in research areas such as business, finance, medical and health science, and bioinformatics; such research areas demand human understanding of the obtained model. The logistic model tree (LMT) algorithm is a popular classification method that combines a decision tree and logistic regression models. The combination of two complementary algorithms produces an accurate and interpretable classifier by combining the advantages of both logistic regression and tree induction. However, LMT has the disadvantage of high computational cost, which makes the algorithm undesirable in practice. In this paper, we propose an efficient method to learn the logistic regression models in the tree. We employ least angle regression to update the regression model in LogitBoost so that the algorithm efficiently learns sparse logistic regression models composed of relevant input variables. We compare the performance of our proposed method with the original LMT algorithm using 14 benchmark datasets and show that the training time dramatically decreases while the accuracy is preserved. Our proposed algorithm is not only accurate and intuitively interpretable but also computationally efficient. It helps users in making the best possible use of the data that are included in expert and intelligent systems.</P> <P><B>Highlights</B></P> <P> <UL> <LI> Computational efficiency of logistic model tree (LMT) algorithm is improved. </LI> <LI> An efficient boosting method for sparse logistic regression learning is proposed. </LI> <LI> The proposed method employs least angle regression to incorporate variable selection into the boosting process. </LI> <LI> Experimental results on 14 datasets to compare the proposed method with the original LMT algorithm are presented. </LI> </UL> </P>

An Agricultural Land Contractual Management Transfer Prediction Model Based on Analytic Hierarchy Process and Logistic Regression

Hai-ying Cao,Ling Wu 보안공학연구지원센터 2016 International Journal of u- and e- Service, Scienc Vol.9 No.10

In the field of application prediction, there are two kinds of common methods to establish the prediction model: prediction model established by artificial data analysis relying on expert experience, and prediction model achieved by statistical model exploiting data analysis. However, the prediction accuracy of the model based on expert is restricted by the experiences, while the model based on statistical analysis is limited by the quality and scale of the training data. In view of the advantages and disadvantages of these two kinds of models, this paper presents a prediction model by integrating Analytic Hierarchy Process and Logistic Regression. The proposed prediction model uses the Analytic Hierarchy Process, which are based on the training data and expert experience to obtain the rank of predominant factor in a specific domain, and exploits the logistical regression model to learn the weights of each influencing factor. Finally, the linear combination of the two models is used to obtain the prediction model. Further, we take agricultural land contractual management transfer prediction as an example to test the proposed hybrid prediction model.

로지스틱 회귀 모형을 이용한 연관성 규칙 채택률의 추정

박희창 한국자료분석학회 2015 Journal of the Korean Data Analysis Society Vol.17 No.6

Data mining is to explore useful information or unexpected rules in a big database and to be utilized as a basis for decision making. In this paper we proposed three types of logistic regression models to estimate association rule adoption rate and discussed the most appropriate model selection methods by numerical examples. First, Hosmer-Lemeshow goodness-of-fit statistics of model 2 (model of confidence and lift) and model 3 (model of support and lift) was not significant, but that of model 1 (model of support and confidence) was significant. The accuracy of classification of model 2 was larger than that of model 3 (model of support and lift). Coefficient of lift was larger than that of confidence in the regression equation of model 2, and coefficient of lift was larger than that of support in model 3. The odds of confidence was 1.142, and that of lift was 1.345 in model 2. The odds of support was 1.088, and that of lift was 1.278 in model 3. After all these analysis, model 2 was the best logistic regression model. 데이터 마이닝은 빅 데이터 안에 잠재되어 있는 정보나 예기치 못한 규칙 등을 탐색하여 이를 의사결정을 위한 근거로 활용하고자 하는 것이다. 본 논문에서는 연관성 평가 기준을 이용한 규칙의 채택률을 추정하기 위한 3 종류의 로지스틱 회귀 모형을 제안하고, 예제를 이용하여 가장 적절한 모형의 선정 방안에 대해 토의하였다. 각 모형에 대해 적합도를 검정한 결과, 모형 1(지지도와 신뢰도를 고려한 모형)은 적합하지 않는 것으로 나타났다. 따라서 이를 제외하고 모형 2(신뢰도와 향상도를 고려한 모형)와 모형 3(지지도와 향상도를 고려한 모형)에 대해 분류 결과의 정확도를 비교해본 결과. 모형 3보다는 모형 2가 더 높게 나타났다. 또한 모형 2에서는 향상도의 회귀계수의 값이 신뢰도의 회귀계수 값보다 크며, 모형 3에서는 지지도의 회귀계수에 비해 향상도의 회귀계수의 값이 크게 나타났다. 오즈비를 비교해보면 모형 2에서는 신뢰도가 한 단위 증가하면 상대비가 1.142배 증가하는 반면에 향상도가 한 단위 증가하면 상대비가 1.345배 증가하며, 모형 3에서는 지지도가 한 단위 증가하면 상대비가 1.088배 증가하는 반면에 향상도는 1.278배 증가하는 것으로 나타났다. 이들의 결과를 종합해볼 때 모형 2가 가장 바람직한 것으로 나타났다.

Comparative study on statistical packages for analyzing logistic regression - MINITAB, SAS, SPSS, STATA -

김순귀,정동빈,박영술 한국데이터정보과학회 2004 한국데이터정보과학회지 Vol.15 No.2

Recently logistic regression is popular in a variety of fields so that a number of statistical packages are developed for analyzing the logistic regression. This paper briefly considers the several types of logistic regression models used depending on different types of data. In addition, when four statistical packages (MINTAB, SAS, SPSS and STATA) are used to apply logistic regression models to the real fields respectively, their scope and characteristics are investigated.

공간통합 모델을 적용한 암괴류 및 애추 지형 분포가능지 추출

이성호,장동호 한국지형학회 2019 한국지형학회지 Vol.26 No.2

This study analyzed the relativity between block stream and talus distributions by employing a likelihood ratio approach. Possible distribution sites for each debris slope landform were extracted by applying a spatial integration model, in which we combined fuzzy set model, Bayesian predictive model, and logistic regression model. Moreover, to verify model performance, a success rate curve was prepared by crossvalidation. The results showed that elevation, slope, curvature, topographic wetness index, geology, soil drainage, and soil depth were closely related to the debris slope landform sites. In addition, all spatial integration models displayed an accuracy of over 90%. The accuracy of the distribution potential area map of the block stream was highest in the logistic regression model (93.79%). Eventually, the accuracy of the distribution potential area map of the talus was also highest in the logistic regression model (97.02%). We expect that the present results will provide essential data and propose methodologies to improve the performance of efficient and systematic micro-landform studies. Moreover, our research will potentially help to enhance field research and topographic resource management.

강제환기식 돈사의 환기량 추정을 위한 회귀모델의 비교

조광곤,하태환,윤상후,장유나,정민웅 한국농공학회 2020 한국농공학회논문집 Vol.62 No.1

To estimate the ventilation volume of mechanically ventilated swine farms, various regression models were applied, and errors were compared to selectthe regression model that can best simulate actual data. Linear regression, linear spline, polynomial regression (degrees 2 and 3), logistic curve,generalized additive model (GAM), and gompertz curve were compared. Overfitting models were excluded even when the error rate was small. Theevaluation criteria were root mean square error (RMSE) and mean absolute percentage error (MAPE). The evaluation results indicated that degree 3exhibited the lowest error rate; however, an overestimation contradiction was observed in a certain section. The logistic curve was the most stable andsuperior to all the models. In the estimation of ventilation volume by all of the models, the estimated ventilation volume of the logistic curve was thesmallest except for the model with a large error rate and the overestimated model.

유통업체의 부실예측모형 개선에 관한 연구

김정욱 한국유통과학회 2014 유통과학연구 Vol.12 No.11

Purpose – The National Agricultural Cooperative Federation of Korea and National Fisheries Cooperative Federation of Korea have prosecuted both financial and retail businesses. As cooperatives are public institutions and receive government support, their sound management is required by the Financial Supervisory Service in Korea. This is mainly managed by CAEL, which is changed by CAMEL. However, NFFC’s business section, managing the finance and retail businesses, is unified and evaluated; the CAEL model has an insufficient classification to evaluate the retail industry. First, there is discrimination power as regards CAEL. Although the retail business sector union can receive a higher rating on a CAEL model, defaults have often been reported. Therefore, a default prediction model is needed to support a CAEL model. As we have the default prediction model using a subdivision of indexes and statistical methods, it can be useful to have a prevention function through the estimation of the retail sector’s default probability. Second, separating the difference between the finance and retail business sectors is necessary. Their businesses have different characteristics. Based on various management indexes that have been systematically managed by the National Fisheries Cooperative Federation of Korea, our model predicts retail default, and is better than the CAEL model in its failure prediction because it has various discriminative financial ratios reflecting the retail industry situation. Research design, data, and methodology – The model to predict retail default was presented using logistic analysis. To develop the predictive model, we use the retail financial statements of the NFCF. We consider 93 unions each year from 2006 to 2012 to select confident management indexes. We also adapted the statistical power analysis that is a t-test, logit analysis, AR (accuracy ratio), and AUROC (Area Under Receiver Operating Characteristic) analysis. Finally, through the multivariate logistic model, we show that it is excellent in its discrimination power and higher in its hit ratio for default prediction. We also evaluate its usefulness. Results – The statistical power analysis using the AR (AUROC) method on the short term model shows that the logistic model has excellent discrimination power, with 84.6%. Further, it is higher in its hit ratio for failure (prediction) of total model, at 94%, indicating that it is temporally stable and useful for evaluating the management status of retail institutions. Conclusions – This model is useful for evaluating the management status of retail union institutions. First, subdividing CAEL evaluation is required. The existing CAEL evaluation is underdeveloped, and discrimination power falls. Second, efforts to develop a varied and rational management index are continuously required. An index reflecting retail industry characteristics needs to be developed. However, extending this study will need the following. First, it will require a complementary default model reflecting size differences. Second, in the case of small and medium retail, it will need non-financial information. Therefore, it will be a hybrid default model reflecting financial and non-financial information.

데이터마이닝 기법을 이용한 교차판매 스코어링 모형

최종후,권혁민 한국자료분석학회 2009 Journal of the Korean Data Analysis Society Vol.11 No.3

교차판매는 효과적인 CRM 활동의 하나로 어떤 제품이나 서비스를 구매하려는 고객에게 관련되는 제품을 추가로 판매하는 활동을 말한다. 통상적으로 기존고객을 대상으로 하기 때문에 신규고객유치에 큰 비용과 노력이 소요되는 오늘날의 마케팅 환경에서는 매우 효과적인 마케팅 방법이 되고 있다. 본 연구는 국내 A금융사의 고객 데이터베이스에 기반하여 교차판매를 위한 스코어링 모형을 개발을 위한 방법론을 탐구하는데 목적을 두고 있다. 모형개발을 위한 데이터마이닝 기법으로는 로지스틱 회귀모형, 의사결정나무모형, 신경망모형과 복합모형 등을 활용했는데, 모형평가 결과 로지스틱 회귀모형이 상대적으로 우수하게 나타나고 있어 이에 기반하여 교차판매 스코어링 모형을 개발하였다. 최종 모형에 의한 교차판매에 대한 이익을 살펴보면 전체고객의 반응률은 4.9%에 불과하였으나 개발된 모형을 적용하여 재가입 가능성이 높은 상위 10% 고객에게 캠페인을 실행하였을 경우 예상 반응률은 약 30%로 매우 향상된 결과를 나타내었다. Cross-Sell, which is one of effective tools for CRM, is a marketing term for the practice of suggesting related products or services to a customer who is considering buying something. Usually Cross-Sell is applied to existing customers. Therefore it is very cost effective tool because customer acquisition brings more cost and effort than customer retention. This study is about developing of scoring model for cross-sell based on customer database of A capital company. We use several data mining tools for developing scoring model, those are logistic regression, decision tree, neural network and ensemble model. According to model assesment results, we chose logistic regression model as the final scoring model for cross selling. It gave us 30% response rate for the highest 10% potential customers, on the other hand 4.9% for the data mart.

Prediction of fecal coliform using logistic regression and tree-based classification models in the North Han River, South Korea

Choi, Soo Yeon,Seo, Il Won Elsevier 2018 JOURNAL OF HYDRO-ENVIRONMENT RESEARCH Vol.21 No.-

<P><B>Abstract</B></P> <P>In this study, data-based classification models were developed for real-time prediction of the exceedance of the safety level on fecal coliform in Daesung-ri site of North Han River. The prediction models were developed using the logistic regression model (LRM) and the tree-based models such as classification and regression model (CART), bagging model (BGM), and random forest model (RFM). For model development, rainfall, water quality, and dam discharge data from 2010 to 2015 were collected from the study site. Clustering methods were applied to reduce the sampling bias of training and test datasets and to improve the prediction accuracy. The developed four models were compared with each other in terms of prediction accuracy and applicability. The test results of developed models showed that the total correct classification rate of the four models ranged from 83.7% to 93.0%. Each classification model showed its own strengths; LRM offered flexibility by tuning cutoff values, while RFM showed the highest accuracy among the four models. The hydro-ecological process on fecal coliform could be explained by analyzing important variables in the prediction models and identifying the impacting factors through the field monitoring. The important factors both in the models and field monitoring were revealed as the rainfall-related variables, dam discharge and total phosphorus, which imply that the fecal pollution in North Han River came mainly from the rainfall events and runoff including nutrients from farmland and livestock farming in the upstream basin of Guwoon Creek and Chungpyung Dam.</P>