TY - CONF
T1 - Multistage modeling for the classification of numerical and categorical datasets
T2 - 2017 IEEE International Conference on Fuzzy Systems, FUZZ 2017
AU - Salgado, C.M.
AU - Fernandes, M.P.
AU - Horta, A.
AU - Xavier, M.
AU - Sousa, J.M.C.
AU - Vieira, S.M.
N1 - Conference code: 130106
Export Date: 18 October 2017
CODEN: PIFSF
References: Sharma, G., Kuo, Y., Freeman, J., Zhang, D.D., Goodwin, J.S., Comanagement of hospitalized surgical patients by medicine physicians in the United States (2010) Archives of Internal Medicine, 170 (4), pp. 363-368; Auerbach, A.D., Wachter, R.M., Katz, P., Showstack, J., Baron, R.B., Goldman, L., Implementation of a voluntary hospitalist service at a community teaching hospital: Improved clinical efficiency and patient outcomes (2002) Annals of Internal Medicine, 137 (11), pp. 859-865; Weiser, T.G., Regenbogen, S.E., Thompson, K.D., Haynes, A.B., Lipsitz, S.R., Berry, W.R., Gawande, A.A., An estimation of the global volume of surgery: A modelling strategy based on available data (2008) The Lancet, 372 (9633), pp. 139-144; Chen, L.M., Wilk, A.S., Thumma, J.R., Birkmeyer, J.D., Baner-Jee, M., Use of medical consultants for hospitalized surgical patients: An observational cohort study (2014) JAMA Internal Medicine, 174 (9), pp. 1470-1477; Keller, J.M., Liu, D., Fogel, D.B., (2016) Fundamentals of Computational Intelligence, Neural Networks, Fuzzy Systems and Evolutionary Computation; Vernieuwe, H., De Baets, B., Verhoest, N.E.C., Comparison of clustering algorithms in the identification of takagi-sugeno models: A hydrological case study (2006) Fuzzy Sets and Systems, 157 (21), pp. 2876-2896; Bezdek, J.C., Keller, J., Krisnapuram, R., Pal, N., (1999) Fuzzy Models and Algorithms for Pattern Recognition and Image Processing, 4. , Springer Science & Business Media; Salgado, C.M., Ferreira, M.C., Vieira, S.M., Mixed fuzzy clustering for misaligned time series (2016) IEEE Transactions on Fuzzy Systems; Freedman, D.A., (2009) Statistical Models: Theory and Practice, , Cambridge university press; Takagi, T., Sugeno, M., (1985) Fuzzy Identification of Systems and its Application to Modeling and Control; Bezdek, J.C., Ehrlich, R., Full, W., FCM: The fuzzy c-means clustering algorithm (1984) Computers & Geosciences, 10 (2), pp. 191-203; Krishnapuram, R., Keller, J.M., A possibilistic approach to clustering (1993) IEEE Transactions on Fuzzy Systems, 1 (2), pp. 98-110; Babuška, R., (1998) Fuzzy Modeling for Control, 12. , Springer Science & Business Media; Krishnapuram, R., Keller, J.M., The possibilistic c-means algorithm: Insights and recommendations (1996) IEEE Transactions on Fuzzy Systems, 4 (3), pp. 385-393; Daabiss, M., American society of anaesthesiologists physical status classification (2011) Indian Journal of Anaesthesia, 55 (2), p. 111; Glance, L.G., Lustik, S.J., Hannan, E.L., Osler, T.M., Mukamel, D.B., Qian, F., Dick, A.W., The surgical mortality probability model: Derivation and validation of a simple risk prediction rule for noncardiac surgery (2012) Annals of Surgery, 255 (4), pp. 696-702; Lichman, M., (2013) UCI Machine Learning Repository
PY - 2017
Y1 - 2017
N2 - Logistic regression and Takagi-Sugeno fuzzy models are sequentially trained with categorical and numerical data in an ensemble-based multistage scheme. In the first stage, a logistic regression model is used to transform the binary feature space into a numerical feature that is used to train a second stage of models consisting of an ensemble of two Takagi-Sugeno fuzzy models. In the ensemble, one model is trained in the space of numerical features and first stage prediction values. The other model is trained only with samples that were classified with a low degree of confidence by the first stage model, in the space of numerical variables. The final output is given by the average of the ensemble predictions at second stage. This scheme was devised under the hypothesis that separating binary from numerical features in the modeling process would increase the performance of a single model using both types of features together. The proposed multistage approach is used to solve a clinical classification problem in a Portuguese hospital. The problem consists of predicting comanagement signalling based on patient clinical data, including diagnosis, procedures, comorbidities and numerical scores, collected before surgery. The multistage performed better in the comanagement dataset, and in 2 out of 5 benchmark datasets. © 2017 IEEE.
AB - Logistic regression and Takagi-Sugeno fuzzy models are sequentially trained with categorical and numerical data in an ensemble-based multistage scheme. In the first stage, a logistic regression model is used to transform the binary feature space into a numerical feature that is used to train a second stage of models consisting of an ensemble of two Takagi-Sugeno fuzzy models. In the ensemble, one model is trained in the space of numerical features and first stage prediction values. The other model is trained only with samples that were classified with a low degree of confidence by the first stage model, in the space of numerical variables. The final output is given by the average of the ensemble predictions at second stage. This scheme was devised under the hypothesis that separating binary from numerical features in the modeling process would increase the performance of a single model using both types of features together. The proposed multistage approach is used to solve a clinical classification problem in a Portuguese hospital. The problem consists of predicting comanagement signalling based on patient clinical data, including diagnosis, procedures, comorbidities and numerical scores, collected before surgery. The multistage performed better in the comanagement dataset, and in 2 out of 5 benchmark datasets. © 2017 IEEE.
KW - Bins
KW - Classification (of information)
KW - Forecasting
KW - Fuzzy systems
KW - Regression analysis
KW - Categorical datasets
KW - Clinical classification
KW - Ensemble prediction
KW - Logistic Regression modeling
KW - Logistic regressions
KW - Multistage approach
KW - Numerical variables
KW - Takagi Sugeno fuzzy models
KW - Numerical models
U2 - 10.1109/FUZZ-IEEE.2017.8015665
DO - 10.1109/FUZZ-IEEE.2017.8015665
M3 - Paper
ER -