Supervision Directory
Explore running research projects, completed student capstones, and future MS/PhD research topics led by Dr. Junaid Akram at SSA Lab.
A supervised final-year engineering development project combining a full-stack freelancing marketplace with AI decision assistance and blockchain payment escrow.
Ghulam Mujtaba (FA22-BSE-199)
Muhammad Waqar Ul Mulk (FA22-BSE-153)A running research project leveraging Large Language Models (LLMs) to detect complex software vulnerabilities in source code through fine-tuning and AST-based prompt engineering.
Muhammad Muneeb Azhar(SP23-BSE-125)A running software security project building a scalable token-based semantic code clone detection platform to capture Type-3 and Type-4 clones across large open-source repositories.
Hafiz Muhammad Shams ul Huda(SP23-BSE-067)A running research project developing an automated patch optimization engine powered by LLMs that synthesizes and verifies security patch suggestions for vulnerable code snippets.
An intelligent code smell detection tool targeting Feature Envy through AST structural analysis, Graph Neural Networks, and automated refactoring recommendations with interactive visual analytics.
A multi-modal web-based AI screening platform combining vision analytics, behavioral questionnaires, and machine learning models for early screening of autism spectrum and psychological disorders.
Completed supervised project implementing static disassembly analysis, APK opcode feature vectors, and ensemble machine learning classifiers for resilient Android malware detection.
Completed biomedical machine learning project deploying deep convolutional neural networks to classify blood cell microscopic images and detect diagnostic anomalies automatically.
Completed web security capstone project delivering an automated Cross-Site Scripting (XSS) static code auditor and dynamic browser defense extension.
Completed sports analytics platform combining computer vision for ball-tracking, player biomechanics assessment, and statistical match prediction.
Completed biomedical signal processing system utilizing audio sensor processing and machine learning to detect nocturnal sleep apnea events in real time.
Completed software vulnerability management platform aggregating CVE repositories, automated severity scoring, and code vulnerability tracking across enterprise projects.
Completed static code analysis suite detecting structural code smells (God Class, Long Method, Feature Envy) and generating safe automated refactoring patches.
Completed research project capturing hierarchical Abstract Syntax Tree and Control Flow Graph structures to recommend software architectural improvements.
Completed computer vision healthcare system utilizing deep convolutional networks for automated skin lesion segmentation and dermatological risk classification.
Supervised student project implementing machine learning algorithms to detect synthetic financial fraud signals generated by the PaySim simulator.
Static and dynamic security analysis project auditing OpenSSL memory safety vulnerabilities and buffer overflow defenses.
Malware analysis project extracting assembly opcode instruction frequency patterns to classify binary malware variants.
Supervised research evaluating feature selection techniques for automated malware detection across PE executable headers.
Cryptographic analysis project evaluating fast correlation attack algorithms against linear feedback shift registers (LFSR).
Mobile software engineering project building a user-centered Android application for medication reminders and emergency alerts.
Extract vulnerable code from patch files at function, file, and component levels, analyse repair patterns, and build a reusable benchmark for vulnerability research.
Measure cloned-code and dependency ratios across related software systems and multiple versions of the same system, using Linux kernel releases as an empirical case study.
Develop a tool that propagates an approved security patch to matching vulnerable fragments across repositories that reuse the affected source code.
Learn repair patterns from vulnerability patches and predict the existence or likely number of vulnerable fragments in open-source software.
Generate valid and faulty test inputs for complex data processing systems using constraints and mutation operators, reducing the cost of manual input-space exploration.
How to apply
BS Software Engineering and BS Computer Science students interested in FYP supervision at COMSATS University Islamabad, Lahore Campus should email Dr Junaid Akram with the following details:
