
A New Class Weighted Algorithm for PSVM
Author(s) -
Xiaoqin Zhu,
Qian Zhang,
Shushan Qiao
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1624/4/042006
Subject(s) - support vector machine , class (philosophy) , computer science , artificial intelligence , intrusion detection system , pattern recognition (psychology) , algorithm , machine learning
Focusing on the problem of slow training speed existing in the SVM in intrusion detection, this paper improves the faster PSVM (Proximal SVM) method and proposes a new kind of class - weighted PSVM algorithm. This kind of algorithm not only has a fast classification speed, but also improves the classification accuracy.