PREDICTIVE LEARNING VIA RULE ENSEMBLES

• 成果类型: Article
• 署名作者: Frieman, Jerome H.; Popescu, Bogdan E.
• 署名单位: Stanford University
• 刊物名称: ANNALS OF APPLIED STATISTICS
• ISSN/ISSBN: 1932-6157
• DOI: 10.1214/07-AOAS148
• 发表日期: 2008
• 页码: 916-954
• 关键词: Shrinkage

摘要

General regression and classification models are constructed as linear combinations of simple rules derived from the data. Each rule consists of a conjunction of a small number of simple statements concerning the values of individual input variables. These rule ensembles are shown to produce predictive accuracy comparable to the best methods. However, their principle advantage lies in interpretation. Because of its simple form, each rule is easy to understand, as it its influence on individual predictions, selected subsets of predictions, or globally over the entire space of joint input variable values. Similarly, the degree of relevance of the respective input variables can be assessed globally, locally in different regions of the input space, or at individual prediction points. Techniques are presented for automatically identifying those variables that are involved in interactions with other variables, the strength and degree of those interactions, as well as the indentities of the other variables with which they interact. Graphical representations are used to visualize both main and interaction effects.