计算机技术与发展Issue(9)：89-93,5.DOI:10.3969/j.issn.1673-629X.2015.09.019
一种改进的贝叶斯算法在短信过滤中的研究
Research on Optimized Naive Bayesian Algorithm in SMS Spam Filtering
摘要
Abstract
With frequent exchanges of information,various harassing messages with mobile phone disturb the normal life for people. For spam filtering technology,research the constructing method and its realization for text classifier based on optimized Naïve Bayesian algo-rithm. The distribution of Naive Bayesian over-reliance on sample space in the short message filtering system and the inherent instability cause an increase in time complexity,propose a spam message structure text classifier based on the improved Bayesian method. The meth-od uses the Bayesian theory and minimum risk decision algorithm to complete the training of bulk SMS. Describe the key technologies of text classification and implement the text classification algorithm. The test results show that the new algorithm can easily train and im-prove the classification accuracy,solving the instability of Naïve Bayesian algorithm,which provides a reference for filtering technology.关键词
贝叶斯/最小风险/文本分类/短信过滤Key words
Bayes/minimum risk/text classification/message filtering分类
信息技术与安全科学引用本文复制引用
裴向杰,唐红昇,陈鹏..一种改进的贝叶斯算法在短信过滤中的研究[J].计算机技术与发展,2015,(9):89-93,5.基金项目
国家自然科学基金资助项目(60802034) （60802034）
