Webpage Vulnerability Detection with Machine Learning , 03/2023 - 06/2023
- Enhance web security by developing a machine learning model to detect webpage vulnerabilities, with a focus on the frequent and high-threat OWASP top 10 vulnerabilities.
- Phishing attacks, a prevalent and dangerous vulnerability, are a central concern due to their potential for information and monetary loss.
- This project aims to create a machine learning model for real-time webpage vulnerability detection as users access websites.
- The process involves dataset loading, train-test splitting, and training/testing three machine learning algorithms: Random Forest, Support Vector Machine (SVM), and Decision Tree.
- Additionally, users can input URLs for instant validation.
- Key Areas: Machine Learning (Random Forest, SVM, Decision Tree), Data Preprocessing, Web Security, OWASP Top 10, Python, Front-end Tools (HTML, CSS, JavaScript)
Matrix Multiplication optimization using OPENCL , 08/2022 - 11/2022,
- Developed an optimized matrix multiplication solution using OpenCL to accelerate computation on heterogeneous computing platforms, including CPUs and GPUs.
- Key highlights include:
- Leveraged parallel computing techniques, including tiling, memory coalescing, and loop unrolling, to improve GPU performance.
- Conducted a performance analysis comparing CPU and GPU execution times to demonstrate the efficiency of GPU acceleration.
- Successfully managed memory allocation and data transfer between host and device.
- Enhanced proficiency in parallel computing, hardware acceleration, and optimization strategies.
- Key skills and technologies: OpenCL, C99, APIs, CPUs, GPUs, DSPs, FPGAs, Parallel Computing Techniques, Performance Analysis, Memory Management.