Identification of Web Risks in the Application Layer using Machine Learning Techniques
Bhanuteja Yada1, Geraldine Bessie Amali D2, Umadevi KS3
1Bhanuteja Yada, School of Computer Science and Engineering VIT, Vellore.
2Geraldine Bessie Amali D, School of Computer Science and Engineering VIT, Vellore.
3Umadevi KS, School of Computer Science and Engineering VIT, Vellore.
Manuscript received on 13 April 2019 | Revised Manuscript received on 18 May 2019 | Manuscript published on 30 May 2019 | PP: 2305-2311 | Volume-8 Issue-1, May 2019 | Retrieval Number: A1258058119/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Web applications have increasingly become one of the customary platforms for service releases and representing data and information over the worldwide web. And thus, many security susceptibilities have controlled to various types of attacks in web applications. This paper is thus intended to look into the use of supervised-machine learning techniques which include; genetic algorithms and support-vector machines, as they are used in detecting some of the key web application layer threats. Some of the most common application layer web threats include; remote file inclusion attacks, SQL injections, and cross-site scripting. As the internet keeps growing each other day, it has become very important to detect the web threats as well as leveraging the powers of machine learning which is one of the various prospective methods in order to make the detection more effective. We will look into how we would use genetic algorithm and support-vector machines to detect the above-mentioned threats in the application layer due to the millions of data requests send every second. From this information, I will come up with a conclusion where I will be able to state the effectiveness, viability and weaknesses of each of the key techniques from which support-vector machines proved to be more effective compared to genetic algorithm in terms of performance and viability.
Index Terms: SQL Injection, RFI, XSS, GA, SVM, Parser, Gathering Test Data, Parsing the Requests
Scope of the Article: Machine Learning