An Improved paralle SVM Algorithm for Chinese Text Classification
SVM is an effective method for learning the classification knowledge from massive data, especially in the situation of high cost in getting labeled classical examples. Based on the analysis of tow parallel proximal support vector machine (PSVM) algorithm-model algorithm and Cascade SVMs algorithm, an improved parallel support vector machine classification algorithm (I PSVM) was proposed, based on the Cascade SVMs. In the IPSVM, the circulating feedback between support vector sets is used to improve classifier performance and is adjusted to save training time of classifier based on the feedback type in Cascade SVMs. The IPSVM is especially suitable to the case of large-scale training set and large number of support vectors. The experimental results indicate that the new method works well in precision and recall rate in the condition that the speeds of classification increase remarkably.
SVM parallel algorithm text classification PSVM
Shengli Zhang
School of computer science Wuyi University Jiangmen, China
国际会议
三峡
英文
1354-1356
2012-05-18(万方平台首次上网日期,不代表论文的发表时间)