基于多分类LSTM的浏览器指纹识别方法
BROWSER FINGERPRINT RECOGNITION METHOD BASED ON MULTI-CLASS LSTM NETWORK
李建伏 1宋国平1
作者信息
- 1. 中国民航大学计算机科学与技术学院 天津 300300
- 折叠
摘要
现有基于机器学习的方法将浏览器指纹的用户识别处理成二分类问题,但该处理方式信息损失较多且识别效率低下.为解决上述问题,提出基于多分类长短期记忆网络(Long Short-Term Memory,LSTM)的浏览器指纹识别方法.其基本思路是将同一用户的浏览器指纹数据处理成时间序列,利用多分类LSTM模型对其进行分类,从而实现用户识别.实验结果表明,该方法比基于二分类的指纹识别方法有更高的准确率和更快的识别速度.
Abstract
The existing methods based on machine learning process the user recognition of browser fingerprint into a binary classification problem,but they have more information loss and low recognition efficiency.In order to solve the above problems,this paper proposes a browser fingerprinting recognition method based on multi-class LSTM.The basic idea of this method was to process the same user's browser fingerprint data into time series,and it used the multi-class LSTM model to classify them,so as to achieve user recognition.The experimental results show that the proposed method has higher accuracy and faster recognition speed than the fingerprinting recognition method based on binary classification.
关键词
浏览器指纹/LSTM/用户识别Key words
Browser fingerprint/LSTM/User recognition引用本文复制引用
基金项目
国家重点研发计划项目(2020YFB1600101)
天津市教委科研计划项目(2020KJ024)
出版年
2024
国家科技期刊平台
NSTL
维普
万方数据