GEOBIA (Geographic Object-Based Image Analysis) allows the simulation from the view of a human interpreter using knowledge models expressed by semantic networks. The construction of knowledge models is a complex task which demands an extensive time. Data mining techniques have been widely used as a support tool for the construction of the semantic network. In this sense, the aim of this study is to analyze the performance of the CART and C4.5 algorithms, which use decision trees, to classify urban land cover. It uses both cognitive approaches and data mining. A WorldView-2 image from São José dos Campos – SP (Brazil) was used for this analysis. Both algorithms presented a good accuracy. The C4.5 algorithm accuracy presented average values slightly higher than the CART algorithm. Regarding the tree models obtained in the experiments, the C4.5 algorithm showed a better generalization capacity in the formulation of the attribute rules. The C4.5 algorithm was supported by other software for the execution of the analyses. This posed a challenge to the researchers for data integration, data format conversion, knowledge of the utilized software and also file replication. Albeit this study covered a reduced geographic area, it presented a high number of objects. From the findings one concludes that, in a large scene, the data volume may represent a big barrier. On the other hand, the CART algorithm tested is part of an integrated GEOBIA platform, which benefits the user reducing the time spent to execute all the image analysis steps.