Web Attack Detection Using Deep Learning

EasyChair Preprint 10029

Abstract

The number of unsecured internet programs has recently increased significantly. A detection mechanism that uses a triangle module operator and deep learning methods is proposed to battle web attacks like SQL injection attacks. To handle these types of attacks on data-based websites such as SQL injection or other similar ones, we advocate Data Collection, Data preprocessing, and Model Training through deep learning algorithms followed by Evaluation techniques. For us to protect against SQL injections from happening at all times parallelly while running the website online, therefore it improves information security if we conduct penetration testing tests along with source code vulnerability tests alongside configuration verification beforehand making sure our Web System passes every vulnerability test before going live. When dealing with Cross-Site Scripting (XSS), verifying input values become crucial so only allow-listed inputs are accepted whereas converting the variable output into encoded versions before showing it back onto your page helps prevent XSS-type malicious activities take place inadvertently taking over control off-hand which may lead toward damage beyond imagination!

Keyphrases: Cross Site Scripting, Neural Network., SQL Injection, deep learning

@booklet{EasyChair:10029,
  author    = {B Logesh and Perasani Bhargav and Appikonda Jeevan Kumar and Genji Yaswanth and Chintala Uday Kiran and Jyoti Godara},
  title     = {Web Attack Detection Using Deep Learning},
  howpublished = {EasyChair Preprint 10029},
  year      = {EasyChair, 2023}}