RT Journal Article
T1 Online adaptive decision trees based on concentration inequalities.
A1 Frías-Blanco, Isvani
A1 Del-Campo-Ávila, José
A1 Ramos-Jiménez, Gonzalo Pascual
A1 Carvalho, Andre C.P.L.F.
A1 Ortiz-Díaz, Agustín
A1 Morales-Bueno, Rafael
K1 Aprendizaje automático (Inteligencia artificial)
K1 Datos masivos
AB Classification trees are a powerful tool for mining non-stationary data streams. In these situations, mas sive data are constantly generated at high speed and the underlying target function can change over time.The iadem family of algorithms is based on Hoeffding’s and Chernoff’s bounds and induces online deci sion trees from data streams, but is not able to handle concept drift. This study extends this family todeal with time-changing data streams. The new online algorithm, named iadem-3, performs two mainactions in response to a concept drift. Firstly, it resets the variables affected by the change and main tains unbroken the structure of the tree, which allows for changes in which consecutive target functionsare very similar. Secondly, it creates alternative models that replace parts of the main tree when theysignificantly improve the accuracy of the model, thereby rebuilding the main tree if needed. An onlinechange detector and a non-parametric statistical test based on Hoeffding’s bounds are used to guaran tee this significance. A new pruning method is also incorporated in iadem-3, making sure that all splittests previously installed in decision nodes are useful. The learning model is also viewed as an ensem ble of classifiers, and predictions of the main and alternative models are combined to classify unlabeledexamples. iadem-3 is empirically compared with various well-known decision tree induction algorithmsfor concept drift detection. We empirically show that our new algorithm often reaches higher levels ofaccuracy with smaller decision tree models, maintaining the processing time bounded, irrespective of thenumber of instances processed.
PB Elsevier
YR 2016
FD 2016-04-21
LK https://hdl.handle.net/10630/30225
UL https://hdl.handle.net/10630/30225
LA eng
NO Frías-Blanco, I., Campo-Ávila, J. del, Ramos-Jiménez, G., Carvalho, A. C. P. L. F., Ortiz-Díaz, A., & Morales-Bueno, R. (2016). Online adaptive decision trees based on concentration inequalities. Knowledge-Based Systems, 104, 179–194.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 20 ene 2026