• With the spread and growth of the technology worldwide, it has led to the increase in the number and types of web attacks which results in various security issues for internet users. Web Application Firewall (WAF) has been introduced to block and filter these web attacks by blocking any malicious actions or requests that the website might be exposed to. In addition, WAF helps in protecting web applications from a variety of application layer attacks such as cross-site scripting (XSS), and SQL injection (SQLi).

• The project aims to enhance the performance of detecting any request sent to web applications and to classify whether the request sent is normal or malicious based on using various Machine Learning and Deep Learning techniques.

The goals of DeepShield are:

• Addresses the limitations of traditional WAFs in detecting attacks such as SQL injection, Cross Site Scripting (XSS), and many more.

• Proposes a layered architecture of WAF to improve the accuracy of threat detection.

• Proposes solutions to reduce the attack detection of overhead time using intelligent algorithms that can predict and prevent web attacks.

• Provides blocking or prevention from different classes of attack types patterns that consistently manage to bypass the WAFs.

• The DeepShield system will use deep learning and advanced machine learning techniques to strengthen online applications against a variety of cyber threats.

• The DeepShield system will be able to quickly and accurately detect threats like SQL injection, and cross-site scripting and many more by analyzing and comparing regular HTTP traffic with potentially malicious traffic to identify attack-indicating parameters and features by incorporating machine learning algorithms.

• The DeepShield system intends to improve security posture and accelerate maintenance procedures while enhancing machine learning and deep learning capabilities to supplement conventional rule-based techniques.