## LOTUS: An algorithm for building accurate and comprehensible logistic regression trees

Citations: | 17 - 5 self |

### BibTeX

@MISC{Chan_lotus:an,

author = {Kin-Yee Chan and et al.},

title = {LOTUS: An algorithm for building accurate and comprehensible logistic regression trees},

year = {}

}

### OpenURL

### Abstract

### Citations

3910 | Classification and Regression Trees - Breiman - 1984 |

2868 | UCI Repository of machine learning databases - Blake, Merz - 1998 |

1840 |
A New Look at the Statistical Model Identification
- Akaike
- 1974
(Show Context)
Citation Context ...mber of categories. Furthermore, because there is no provision for ordinal variables in glm, they are treated as quantitative in LOGIST. The step function uses the Akaike information criterion (AIC) (=-=Akaike 1974-=-), AIC = −2(maximized log-likelihood)+ 2(number of parameters). Starting with a full model, variables are dropped or added sequentially until AIC is minimized.LOTUS: AN ALGORITHM FOR BUILDING LOGISTI... |

749 | Applied logistic regression - Hosmer, Lemeshow - 1989 |

222 | Learning with continuous classes - Quinlan - 1992 |

168 | A comparison of prediction accuracy, complexity, and training time of thirtythree old and new classification algorithms - Lim, Loh, et al. |

119 | Generalized Linear Models. 2nd ed - McCullagh, Nelder - 1989 |

96 | Logistic model trees - Landwehr, Hall, et al. - 2005 |

90 | Applied Regression Including Computing and Graphics - Cook, Weisberg - 1999 |

87 | Tests for linear trends in proportions and frequencies - Armitage - 1955 |

76 | Split selection methods for classification trees - Loh, Shih - 1997 |

75 | Logistic Regression Using the SAS System: Theory and Application - Allison - 1999 |

63 | Tree induction vs. logistic regression: a learning-curve analysis - Perlich, Provost, et al. - 2003 |

56 | Regression trees with unbiased variable selection and interaction detection. Statistica Sinica - Loh |

44 | Classification trees with unbiased multiway splits - Kim, Loh - 2001 |

35 | Some methods of strengthening the common χ2 tests - Cochran - 1954 |

31 | Quantile regression trees - Chaudhuri, Loh - 1998 |

9 | Tree-based Models. Statistical Models - Clark, Pregibon - 1992 |

6 | Ethnic Wage Differences in Malaysia: Parametric and Semiparametric Estimation of the Chinese-Malay Wage Gap - Schafgans - 1998 |

6 | The hybrid CART-Logit model in classification and data mining - Steinberg, Cardell - 1998 |

4 | CoIL Challenge 2000 Tasks and Results: Predicting and Explaining Caravan Policy Ownership.” Coil Challenge 2000 - Putten, Ruiter, et al. - 2000 |

3 | Applied Categorical Data Analysis - Le - 1998 |

2 | Predictors of male residence patterns in groups of black howler monkeys - Jones, Milanov, et al. - 2008 |

2 | Logistic regression using covariates obtained by product-unit neural network models. Pattern Recognition 40:1 - HERVASMARTINEZ, MARTINEZESTUDILLO - 2007 |

2 | Genetic tuning of fuzzy inference within fuzzy classifier systems. Expert Systems 23:2 - Crockett, Bandar, et al. - 2006 |

1 |
2011. Predictors of adherence with self-care guidelines among persons with type 2 diabetes: results from a logistic regression tree analysis
- Yamashita, Kart, et al.
(Show Context)
Citation Context ...p) = log{p/(1 − p)}. The unknown regression parameters β0, β1, ..., βK are usually estimated by maximum likelihood. Although the model can provide accurate estimates of p,ithastwo serious weaknesses: =-=(1)-=- it is hard to determine when a satisfactory model is found, because there are few diagnostic procedures to guide the selection of variable transformations and no true lack-of-fit test, and (2) it is ... |

1 |
Does segmentation always improve model performance in credit scoring?. Expert Systems with Applications . [CrossRef
- Bijak, Thomas
- 2011
(Show Context)
Citation Context ...esses: (1) it is hard to determine when a satisfactory model is found, because there are few diagnostic procedures to guide the selection of variable transformations and no true lack-of-fit test, and =-=(2)-=- it is difficult to interpret the coefficients of the fitted model, except in very simple situations. A good example of the difficulties is provided by the low birth weight dataset of Hosmer and Lemes... |

1 |
Data Mining Methods for Hydroclimatic Forecasting
- Wei, Watkins
- 2011
(Show Context)
Citation Context ...s. Three dependence structures are studied: (1) the “independence” case where the variables are mutually independent, (2) a “weak dependence” case where some of the variables are not independent, and =-=(3)-=- a “strong dependence” case where the correlation between X2 and X3 is .995. The distributions of the variables are given in Tables 4 and 5 and the joint distribution of X4 and X5 is given in Table 6.... |

1 | The Genome-Wide Association Study—A New Era for Common Polygenic Disorders - Roberts, Wells, et al. - 2010 |

1 | Hyoung Lee. 2010. Analysis of rigid pavement distresses on interstate highway using decision tree algorithms - Kang, Kim, et al. |

1 | Torsten Hothorn , Kurt Hornik . 2008. Model-Based Recursive PartitioningModel-Based Recursive Partitioning - Zeileis |

1 | Data mining, neural nets - Ziegler, DeStefano, et al. - 2007 |