Title :
Research on incremental decision tree algorithm
Author_Institution :
Dept. of Comput., Zaozhuang Univ., Zaozhuang, China
Abstract :
For data analysis of increase rapidly customer behavior, Web log analysis, network intrusion detection systems and other online classification system, how to quickly adapt to new samples is the key to ensure proper classification and sustainable operation. This paper presents a new adaptation data incremental decision tree algorithm, which combines RAINFOREST structure. It combines with the traditional SPRINT decision tree algorithm, and uses new samples quickly train a new decision tree based on the original decision tree. The improved algorithm deal with new samples at any time to produce a decision tree related, and the tree has been optimized with real-time.
Keywords :
Internet; consumer behaviour; data analysis; decision trees; learning (artificial intelligence); pattern classification; security of data; RAINFOREST structure; SPRINT decision tree algorithm; Web log analysis; customer behavior; data analysis; incremental decision tree algorithm; network intrusion detection system; online classification system; sustainable operation; Algorithm design and analysis; Classification algorithms; Data mining; Decision trees; Indexes; Remuneration; Training; Data mining; Gini-index; Incremental learning; decision tree;
Conference_Titel :
Electronic and Mechanical Engineering and Information Technology (EMEIT), 2011 International Conference on
Conference_Location :
Harbin, Heilongjiang, China
Print_ISBN :
978-1-61284-087-1
DOI :
10.1109/EMEIT.2011.6022930
