Machine Learning Based Malicious URL Detection

Machine Learning Based Malicious URL Detection
复制标题

基于机器学习的恶意URL检测

DOI:
10.35940/ijeat.d1006.0484s19
复制
发表时间:
2020
期刊:
International Journal of Engineering and Advanced Technology
影响因子:
--
通讯作者:
Aditya Joshi
Aditya Joshi
中科院分区:
--
文献类型:
--
作者:
Divya Kapil;Atika Bansal;Anupriya;Nidhi Mehra;Aditya Joshi

文献摘要

被引文献

相似文献

今天,互联网技术已经成为一个重要的 教育、娱乐、游戏、银行业 和沟通。在这个现代数字时代, 点击一下就能获得任何信息。但所有的一切 利弊,因为我们有任何信息在我们的提示,但 互联网也是一个攻击平台。当我们利用互联网 我们的工作很容易,同时许多攻击者试图窃取信息 从我们的系统。攻击的手段很多,恶意的 URL其中之一当用户访问一个网站时, 恶意的,那么它会触发一个恶意的活动, 预先设计的因此,有各种方法可以找到 互联网上的危险URL在本文中,我们使用 机器学习方法来检测恶意URL。我们使用 ISCXURL2016数据集,使用J48,随机森林,懒惰 算法和贝叶斯网络分类器。作为性能指标,我们 计算准确率、TPR、FPR、精确率和召回率。
Today Internet technology has become an essential part of our life for education, entertainment, gaming, banking and communication. In this modern digital era, it is very easy to have any information by one click. But everything which has pros and cons, as we have any information at our tips but Internet is an attack platform also. When we use Internet to make our work easy same time many attacker try to steal information from our system. There are many means for attacking, malicious URL one of them. When a user visits a website, which is malicious then it triggers a malicious activity which is predesigned. Hence, there are various approaches to find dangerous URL on the Internet. In this paper, we are using machine learning approach to detect malicious URLs. We used ISCXURL2016 dataset and used J48, Random forest, Lazy algorithm and Bayes net classifiers. As performance metrics, we calculate accuracy, TPR, FPR, precision and recall.