作者: Andrés R. Masegosa , Joaquín Abellán
DOI: 10.1007/978-3-540-75256-1_46
关键词: Probability distribution 、 Dirichlet distribution 、 Decision tree 、 Regular polygon 、 Mathematics 、 Data mining 、 Entropy (information theory)
摘要: In this article, we shall present a method for combining classification trees obtained by simple from the imprecise Dirichlet model (IDM) and uncertainty measures on closed convex sets of probability distributions, otherwise known as credal sets. Our combine has principally two characteristics: it obtains high percentage correct classifications using few number can be parallelized to apply very large databases.