融合网络安全领域知识的Web指纹智能识别

Web Fingerprint Intelligent Recognition Integrating Knowledge in the Field of Network Security

针对Web指纹识别能力有限的问题,提出了一种融合网络安全领域知识的Web指纹识别方法。通过构建网络安全领域的术语词典和使用预训练的专业领域词向量,将网络知识与Web指纹识别能力相融合,并合理利用了白字符和网页中的特殊关键字对识别结果进行优化。实验结果表明,多种模型在融合网络安全知识的策略以后,平均识别准确率从75.1%提高到82.5%、召回率从62.3%提高到75.7%、F1值从0.683提高到0.790。

Aiming at the limited ability of Web fingerprint identification,this paper puts forward a Web fingerprint identification method which integrates the knowledge of network security field.This method combines network knowledge with Web fingerprint recognition ability by constructing a term dictionary in the field of network security and using a pre-trained professional domain word vector,and optimizes recognition results by using white characters and special keywords in web pages.The experimental results show that the average recognition accuracy of multiple models is increased from 75.1% to 82.5%,recall rate from 62.3% to 75.7%,F1 score from 0.683 to 0.790 after the integration of network security knowledge strategies.