摘要
为了提高垃圾邮件过滤的准确率,提出了一种基于自适应本体的垃圾邮件过滤实验。该实验首先对邮件预处理,构建特征向量,采用TF-IDF方法计算权重。在预处理的基础上,将邮件特征向量输入分类器进行分类,通过Jena映射本体,并将待测邮件检测结果返回训练集,进一步自动调整本体,实现待测邮件的智能识别。理论分析和实验结果表明本实验方法的可行性和有效性,与传统的方法相比,具有更高的准确率和适应性。
In order adaptive ontology to improve the accuracy of spare filtering, this paper proposes a spam filtering experiment based on Firstly, a mail is preprocessed, including characteristic vector constructing, and the weight is calculated by TF-IDF. On the basis of pretreatment, characteristic vector is input into a classifier to classify, and the ontology is mapped by Jena. Furthermore, mail results are sent back to the training set, ontology can be automatic adjusted. The experiment realizes the mail intelligent identification. Theoretical analysis and simulation experiments show that the experiment is feasible and effective, and the solution is with better accuracy and fitness compared with the traditional method.
出处
《实验室研究与探索》
CAS
北大核心
2016年第7期139-142,共4页
Research and Exploration In Laboratory
基金
湖北省教育厅重点科研项目(D20145001)
