Vulnerability Detection Using Machine Learning Thesis Ideas
We are a team of professionals who are well versed in machine learning concept. Our thesis work will motivate the readers with our unique writing style and quality. Your involvement will be only selection of thesis topic further everything we will guide you. The best thesis topics that we have worked under Vulnerability Detection are listed go through our work contact us for more support.
1. Cross Site Scripting (XSS) vulnerability detection using Machine Learning and Statistical Analysis
Keywords:
Attack detection, machine learning, Web Application Security, XSS attacks
Our paper debates Cross Site Scripting (XSS) vulnerability detection using various ML and DL techniques. The Open Web Application Security Project (OWASP) has XSS attacks as a top three risks to web applications. We used various methods to detect XSS attack DL methods like LSTM, CNN and boosting methods like Ada Boost, Gradient Boost algorithm, and classification methods such as LR, SVM, KNN, RF, NB, DT to detect XSS attack. Our paper verifies ML model has only able to accurately find XSS attacks.
2. Cross-Site Request Forgery as an Example of Machine Learning for Web Vulnerability Detection
Keywords:
Cross-Site Request Forgery’s (C.S.R. Fs), ML classifier
Our paper uses give a method for discovering flaws in web application over ML. We might merge cognitive knowledge of web app terminology with automated software methods based on verbally reported data. We used Mitch tool a ML method to black-box investigation for Cross Site Request Forgery (CSRF) issues can be constructed by utilizing these principles.
3. An empirical study of text-based machine learning models for vulnerability detection
Keywords:
Vulnerability detection, Text-based analysis
Our paper presents empirical study text-based ML methods to detect vulnerability. Our paper uses seven ML methods, five natural language processing methods and three data processing methods. At first we offer results on full context function and next we propose condensed function and handle statistical analytics to regulate it with certain method, technique and or models.
- Detection of Vulnerabilities by Incorrect Use of Variable Using Machine Learning
Keywords:
Software fault detection, variable vulnerability
The Common Weakness Enumeration (CWE) defines the list of errors from hardware or software. We propose a method to decrease false alarms and to detect vulnerabilities by performing static and dynamic verification using ML. We also implemented our method VVDUM (Variable Vulnerability Detector Using Machine learning).We also manage the comparison with existing static/dynamic analysis tool.
- Effective Industrial Internet of Things Vulnerability Detection Using Machine Learning
Keywords:
Detection, IIoT, Smart Factory, Vulnerability assessment.
Our work uses ML methods to detect and improve vulnerability accuracy. Our work offers an IIoT protocols and their vulnerabilities. The performance of ML based detection system was improved using the dataset WUSTLIIoT for industrial control system (SCADA) cyber security research. The method was validated using ICS-SCADA and CICDDoS2019 datasets a current dataset that take new dimension of Distributed Denial of Service (DDoS) attack on network.
- A Systematic Literature Review on Software Vulnerability Detection Using Machine Learning Approaches
Keywords:
Software vulnerabilities, vulnerability detection, deep learning, program analysis.
Our paper analyses and reviews the current state-of the art research implementing ML and DL methods to detect vulnerability. Our aim is to examine the neural techniques for learning and to understand code semantics to make possible vulnerability detection. We used NVD and SARD datasets for detection and the outcome verifies ML and DL methods give better outcome in vulnerability detection.
- Machine-Learning-Based Vulnerability Detection and Classification in Internet of Things Device Security
Keywords:
IoT security; cyber-attacks; device security
Our work examines a review to explore the techniques and tools utilized in vulnerability detection in IoT environments that uses ML methods on different datasets. Our study uses the general possible vulnerabilities of IoT architectures are recognize on each layer. We proposed ML that can be used to detect IoT vulnerabilities.
- Machine learning techniques for software vulnerability prediction: a comparative study
Keywords:
Prediction models
We have proposed Vulnerability discovery model (VDM) to present accurate outcome for each vulnerability dataset. Our paper uses ML methods and we present an empirical study by relating some ML methods than statistical techniques to predict the software vulnerability. We have to estimate the following ML methods like SVM, MLP, M5Rrule, M5P, reduced error pruning tree etc. We also used statistical methods like Alhazmi-Malaiya model, LR and logistic regression model
- A Machine Learning Approach for Web Application Vulnerability Detection Using Random Forest
Keywords:
URL, Attacks, Extract, Web portals, Vulnerabilities, Random Forest Model
The goal of the attacker is to get important data’s from users through URL links. We have to fill the gap of existing method to stop the attacks but that does not perform well and the attackers will get that from web applications. Now we are finding steady and trustworthy web application attack detection software. Our model uses ML method to secure web application and getting high accuracy while using random forest method.
- Detecting Unknown Vulnerabilities in Smart Contracts with Binary Classification Model Using Machine Learning
Keywords:
Smart contract, Unknown vulnerability, Opcode sequences, N-gram, Blockchain
Our paper proposes a ML based unknown vulnerability detection by utilizing opcode sequences. Our scheme initially obtains opcode sequence of execution path of contract transaction in the Ethereum virtual machine (EVM) by repeat them in Ethereum. Finally we validate them by three ML methods like K-NN, SVM and LR. The SVM gives the higher accuracy rate.