一种基于SVM的多层分类策略  

A Strategy of Multi-level Classification Based on SVM

作　　者：路斌[1] 杨建武[1] 陈晓鸥[1] 

机构地区：[1]北京大学计算机研究所文字信息处理技术国家重点实验室,北京100871

出　　处：《计算机工程》2005年第1期73-75,113,共4页Computer Engineering

摘　　要：提出了一种新的基于反例文档选择的多层分类策略1-vs-brothers。与原策略相比,该策略在训练阶段仅仅选择兄弟节点包含的样例文档作为反例,从而减少了较深层次节点需要学习的文档。实验结果表明,在该文的实验条件下,基于该策略的算法使得训练效率提高了60%,而分类精度却基本上保持不变。该策略还可以用在1-vs-1之上形成1-vs-brother策略,用来减少多层分类情况下节点训练时需要对比学习的节点数目。This paper proposes a new multi-level classification strategy named 1-vs-brothers on the basis of 1-vs-rest strategy which is used to transfer multi-category problems to two category problem. Compared with the original strategy, the new strategy which is based on the selection of negative examples, only selects the example documents of brother nodes as negative examples, that cuts down the documents number needed to learn during the non-first level nodes training period. The experiment shows that on the data of this paper the algorithm based on this strategy improves the training efficiency about 60 percent, and the classification precision yet remains no change on the whole. This strategy can also be used on l-vs-l strategy to form 1-vs-brother strategy, which will cut down the node numbers needed to learning during the training period of multi-level classification.

关 键 词：支持向量机 自动分类 多层分类 I-vs-brothers策略 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] TP391[自动化与计算机技术—控制科学与工程]