An Algorithm of Decision-tree Generating Automatically Based on Classification

Decision tree classification techniques have been used for a wide range of classification problems and becoming an increasingly important tool for classification of remotely sensed data. These techniques have substantial advantages for land use classification problems because of the reflexibility, nonparametric nature, and ability to handle non-linear relations between features and classes. In this paper, an algorithm, by which a decision-tree of classification is generated automatically according to the distribution of targets in feature spaces, is proposed. The method of selecting threshold, data structures and flow of the algorithm are introduced. An example is showed and compared with man-made tree, both of them are generated on the same condition.