Indexed by:
Abstract:
为改进朴素贝叶斯(naive Bayes,NB)算法在识别未知恶意代码过程中学习速度慢的缺点,在分析研究朴素贝叶斯算法、复合贝叶斯(multi-naive Bayes,MNB)算法的基础上,提出了一种改进贝叶斯(half-increment naive Bayes,HNB)算法.算法采用特征集增量学习方式,在保证分类精度不降低的前提下,学习速度提高约30%.实际样本测试表明,分类精度达到了96%,其中对已知恶意代码的分类精度达到99%.
Keyword:
Reprint Author's Address:
Email:
Source :
北京工业大学学报
ISSN: 0254-0037
Year: 2011
Issue: 5
Volume: 37
Page: 766-772
Cited Count:
WoS CC Cited Count: 0
SCOPUS Cited Count:
ESI Highly Cited Papers on the List: 0 Unfold All
WanFang Cited Count: 3
Chinese Cited Count:
30 Days PV: 3
Affiliated Colleges: