Software Engineering Commons^™

Llm-Driven Mission Control And Autonomous Planning For Search-And-Rescue Uavs: A Simulation-Based Evaluation, Naser Bader Alsaedi

Theses

Unmanned aerial vehicles (UAVs) are increasingly used in search‑and‑rescue (SAR) missions, yet many systems still rely on fragmented software where mission design, perception, and flight control are configured separately. This thesis examines whether a unified AI‑driven framework can reduce configuration effort and operator workload in UAV‑based SAR operations. The proposed system integrates natural‑language mission specification using a large language model (LLM) (LLaMA 3.1), autonomous coverage planning, YOLOv8‑based victim detection, and PX4/MAVSDK control within a single architecture. Operators describe missions through free‑form text or a graphical interface; the model converts these descriptions into structured mission parameters that are automatically planned and …

Software Engineering In The Age Of Coding Agents: Failure Modes And Rejection Patterns, Mahd Mohd Hindi

Theses

This thesis investigates the real-world behavior of LLM-driven coding agents that generate code changes and submit pull requests (PRs) to public software repositories. As these tools evolve from autocomplete-style assistants into more autonomous agents, their contributions increasingly interact with socio-technical review processes (human reviewers, bots, CI/CD gates, and project norms). The thesis focuses on understanding why agent-generated PRs are accepted or rejected and what these outcomes reveal about current agent limitations in practical development workflows.

The main objective of this thesis is to systematically characterize rejection patterns and failure modes of agent-generated pull requests in real repositories. Specifically, the thesis …

Enhancing Llm Code Generation: A Systematic Evaluation Of Multi-Agent Collaboration And Runtime Debugging For Improving Accuracy, Reliability, And Latency, Nazmus Ashrafi

Theses

The use of large language models (LLMs) for automated code generation has emerged as a significant focus within AI research. As these pretrained models continue to evolve, their ability to understand and generate complex code structures has opened up new possibilities for automating intricate programming tasks with greater accuracy. Although contemporary foundational models demonstrate promising results, researchers continue to explore optimal post-training strategies to enhance code quality. These include supervised fine-tuning, retrieval-augmented generation (RAG), debugging, and many others. In this thesis, I combine two such widely used post training approaches—namely (1) multi agent collaboration and (2) runtime execution of information-based …

A Data-Driven Recommendation System For Selecting The Appropriate Mode Of Learning And Instructional Tools Based On Course Characteristics, Ayisha Manzoor

Theses

The rapid transformation of educational delivery methods during the COVID-19 pandemic required institutions to transition between online, hybrid, and offline learning approaches, creating both challenges and opportunities for educators and students. While online and hybrid learning modes ensured continuity, their effectiveness across different course types remained uncertain. This thesis addresses this gap by developing a datadriven recommendation framework that predicts Course Learning Outcome (CLO) achievement scores using regression, and recommends the most appropriate learning mode (online, hybrid, or offline) along with instructional tools based on course characteristics. This study analyzed 100 undergraduate and postgraduate courses from the College of Information …

A Framework For Mixed Reality Within Healthcare Education, Benyamin Ebadinia

Theses

Understanding complex three-dimensional systems and spatial relationships is a recurring difficulty in healthcare education, where students are often expected to reason about internal structures and multi-system processes from 2D diagrams, textbook figures, and static mannequins. This thesis presents the design, implementation, and mixed-methods evaluation of Systems Simulation, a reusable mixed reality (MR) application intended to help undergraduate nursing students explore human anatomy and pathophysiology using immersive 3D visualization.

Built in C# with the StereoKit framework for Microsoft HoloLens 2, Systems Simulation organizes nine anatomical body systems within a shared application. Learners can select a system, anchor the model in their …

Ai-Powered Pedagogy: Revolutionizing Students’ Learning Experiences Through Integration Of Ai Technologies, Amna Awad Alsaedi

Theses

This research examines the integration of Artificial Intelligence (AI) within the educational sector with the aim of enhancing student learning outcomes. AI offers tailored learning experiences, interactive educational content, prompt feedback, and access to diverse learning resources. Nonetheless, challenges including addiction, costliness, privacy infringement, bias, ethical dilemmas, market competition, and moral considerations require resolution. Research particularly delves into the utilization of chatbots, and algorithms designed to simulate human interactions and generate text resembling human speech. Educational applications powered by AI have the potential to heighten student engagement, comprehension, and academic performance by substituting traditional assignments with concise, information-rich lessons. Furthermore, …

Effect Of Virtual Reality Technology On Computer Science/Engineering Based Laboratories Education – A Case Study, Mariam A. Al Nuaimi

Theses

Virtual reality (VR) is becoming increasingly popular and essential in education as institutions strive to incorporate technology in education. In Engineering and Computer Science, students face difficulties in comprehending many abstract complex concepts that are typically studied in Labs. This situation worsens with hardware failure and the lack of pedagogical tools. The main goal of this paper is to investigate the impact of VR environments on learning advanced and complex STEM concepts. We also explore the impact of different human-computer interaction (HCI) techniques including gamification on the learning experience of STEM students. To measure such impact, we developed a VR …

Effect Of Virtual Reality Technology On Computer Science/Engineering Based Laboratories Education – A Case Study, Saeed Salem Al Shebli

Theses

The rapid growth in mobile applications raises critical concerns about the security of apps and users' privacy, especially in permission control. Mobile apps access sensitive information of users, and the current cybersecurity landscape faces a huge challenge in ensuring the least required permissions are granted. This research focuses on designing an advanced permission recommendation system that couples the strengths of Natural Language Processing (NLP) and Machine Learning (ML) in solving most of the existing gaps in permission management, thus guiding which permissions are mostly needed by Android applications.

The research thus follows a multi-classification approach, integrating state-of-the-art ML techniques with …

An Edge Platform Streamlining Connectivity Between Modern Edge Devices And Cloud, Anderson Carvalho

Theses

The numerous uses of cloud, fog, and edge computing in a variety of industries are thoroughly examined in this thesis, with a special emphasis on smart factories, smart cities, and smart agriculture. It starts with an introduction to cloud computing, going over its history, importance in contemporary digital infrastructures, and related security concerns. It emphasizes the value of cloud computing for processing and storing data while addressing issues like guaranteeing low latency applications and possible improvements from container technologies. The conversation then shifts to fog computing, going over its history, designs, and uses. It focuses on how fog computing might …

Fish-Eye Camera-Based Real-Time Pedestrian Crossing Predicting System, Mohammed Abdulla Almesafri

Theses

Fisheye cameras are widely used in traffic monitoring for their broad view, yet their distortion challenges deep-learning models in pedestrian detection and tracking. Despite available datasets like FishEye8K, collected in Hsinchu, Taiwan, and the availability of several studies that have delved into pedestrian prediction systems, a notable gap remains: the absence of datasets specifically designed for the cultural context of the UAE. This study aims to address this gap by introducing an in-house fisheye dataset tailored to enhance the prediction and tracking of pedestrians in fisheye footage within the UAE's environment. The present study addresses the development of a Graphical …

An Intelligent Academic Advising System For Course Recommendation Using Large Language Models, Malika Maya Iratni

Theses

Academic advising is an important resource for students, especially in higher education, in order to guide them to make the best possible decisions to improve their overall academic performance and overall academic journey. With the increasing number of students joining these institutions each year, traditional advising becomes a time-consuming and inefficient process that can leave students discouraged, and advisors overwhelmed. Therefore, there is a need to develop intelligent advising systems that make use of the recent advancements in technology, to support the advising process, and increase overall student satisfaction. In this work, an academic advising model that uses Recommender Systems …

Vascular Brain Digital Twins For Medical Training And Education In Metaverse, Shamma Khaled Alghafri

Theses

This study addresses the need for innovative educational tools in the field of anatomy, specifically focusing on brain anatomy. The objective is to develop a virtual reality application and a 3D visualization that offer immersive and interactive learning experiences for students. The VR application, developed using Unity and designed for the Oculus Quest 2 headset, creates an immersive virtual laboratory environment. This environment includes interactive elements such as a table with buttons for displaying brain models and a whiteboard for user interaction. Users can manipulate and explore different brain structures, enhancing their understanding of complex anatomical features. Additionally, we integrated …

Impact Of Covid-19 On Security Vulnerabilities Of Learning Management Systems: A Study Towards Security And Sustainability Enhancement, Souheil Abdel-Latif Akacha

Theses

The rapid adoption of Learning Management Systems (LMSs) like Moodle, Chamilo, and Ilias became essential for online education due to the Coronavirus Disease 2019 (COVID-19) pandemic, revolutionizing online learning while exposing security vulnerabilities. This thesis explores security concerns within these LMSs across different pandemic periods. By analyzing existing patches, security measures, and emerging cybersecurity technologies, recommendations are formulated to enhance LMS security against evolving cyber threats, providing actionable insights for educational institutions to ensure secure online education continuity. The numerical findings highlight the increasing need for proactive security measures in Moodle, the fluctuating nature of vulnerabilities in Chamilo, and the …

Dress-Code Violation Detection In Arabic Regions Using Object Detection Machine Learning Model, Maha Sadat Aghaei

Theses

The dress code violation detection system is crucial for assessing clothing appropriateness in public areas. This study aims to improve this system using advanced computer vision and machine learning techniques to more effectively categorize people's attire in images and videos. To enhance adaptability and create a user-friendly graphical interface for system management and deployment, we have generated a unique dataset from various contexts mix of Western and Arabic clothing. This allows users to interact with graphical components, including the ability to upload images or use live video for clothing detection. Moreover, we have taken privacy concerns into account and implemented …

The Combination Approaches For One Class Classifier Ensembles For Software Defect Datasets, Maitha Mohammed Alkalbani

Theses

The classification of imbalanced datasets poses significant challenges, becoming a crucial topic in ML, particularly when standard algorithms struggle with accurate classification. In OCC, classifiers may encounter objects from ensembles of one class, leading to outlier scores generated at different scales. Additionally, there is a lack of a unified combination method, with many experiments resorting to using an average as the combination method. The thesis aims to investigate the effectiveness of normalization and unnormalized outlier scores on OCC ensembles. Furthermore, we conducted a comparative study of different types of combination methods. We used k-means clustering as OCC model, and the …

Enhancing Autism Education: Exploring Interactive Videos And Ai Integration For Effective Teaching, Fatima Ahmed Alraeesi

Theses

This research focuses on enhancing autism education by integrating interactive videos and AI solutions to improve teacher training. As the number of autistic students rises, it becomes crucial for special education teachers to employ effective teaching strategies tailored to individual needs. The most effective teaching methods for autistic students involve understanding the condition and incorporating customized instruction strategies, such as adapting assignments to suit the student's needs, assisting those with difficulty speaking, and employing visual aids for better organization. The proposed solution involves utilizing interactive video technology to train teachers, bridging the gap between research and practical implementation of educational …

Data-Driven Modeling Of Student Performance In The Time Of Distance Learning, Iman Saad Megdadi

Theses

One of the important aspects that all academic institutions work towards improving is Student Performance. It is obviously the primary indicator of success or failure of institutions. Student performance predictions are vital to instructors and educational decision makers to help, across all levels, tailor learning according to the students’ needs. Therefore, it is essential for Higher Education Institutions to predict student performance in distance learning which has been, and remains, the primary method of learning in some countries due to Corona Virus pandemic. For this reason, this research is going to predetermine a fitting definition of student performance in time …

Interactive Emirate Sign Language E-Dictionary Based On Deep Learning Recognition Models, Ahmed Abdelhadi Abdelhadi

Theses

According to the ministry of community development database in the United Arab Emirates (UAE) about 3065 people with disabilities are hearing disabled (Emirates News Agency - Ministry of Community Development). Hearing-impaired people find it difficult to communicate with the rest of society. They usually need Sign Language (SL) interpreters but as the number of hearing-impaired individuals grows the number of Sign Language interpreters can almost be non-existent. In addition, specialized schools lack a unified Sign Language (SL) dictionary, which can be linked to the Arabic language being of a diglossia nature, hence many dialects of the language co-exist. Moreover, there …

A Data Driven Model To Promote Preparedness And Respond Intelligently To Pandemic Outbreaks, Safea Mohammed Al Senani

Theses

The COVID-19 pandemic has had a major effect on various vital sectors of the economy, including education healthcare, and the industry. Governments have imposed strict regulations to reduce the spread of this global disease outbreak. Consequently, working from home, online learning, social distancing and various control measures were enforced. In response, many schools shifted to distance learning, although most of these schools were neither technically ready nor administratively prepared for the online transition. Despite recent progress, countries are still experiencing daunting challenges to control the infection rate and magnitude, stabilize the economy, and relax socialization and public life activities. Decision-makers …

A Reinforcement Learning Approach To Vehicle Path Optimization In Urban Environments, Shamsa Abdulla Al Hassani

Theses

Road traffic management in metropolitan cities and urban areas, in general, is an important component of Intelligent Transportation Systems (ITS). With the increasing number of world population and vehicles, a dramatic increase in road traffic is expected to put pressure on the transportation infrastructure. Therefore, there is a pressing need to devise new ways to optimize the traffic flow in order to accommodate the growing needs of transportation systems. This work proposes to use an Artificial Intelligent (AI) method based on reinforcement learning techniques for computing near-optimal vehicle itineraries applied to Vehicular Ad-hoc Networks (VANETs). These itineraries are optimized based …

Heart Rhythm Classification From Static And Ecg Time-Series Data Using Hybrid Multimodal Deep Learning, Ahmad Abdulrazaq Abdulla Alnajjar

Theses

Cardiovascular arrhythmia diseases are considered as the most common diseases that cause death around the world. Abnormal arrhythmia diseases can be identified by analyzing heart rhythm using an electrocardiogram (ECG). However, this analysis is done manually by cardiologists, which may be subjective and susceptible to different cardiologist observations and experiences, as well as to noise and irregularities in those signals. This can lead to misdiagnosis. Motivated by this challenge, an automated heart rhythm diagnosis approach from ECG signals using Deep Learning has been proposed. In order to achieve this goal, three research problems have been addressed. First, recognize the role …

The Role Of Software Engineering In Bioinformatics, Brendan Sean Lawlor

Theses

This thesis proposes that by applying state-of-the-art software engineering tools, techniques and frameworks to currently recognised challenges in bioinformatics, improved outcomes can be attained in that field. It begins by decomposing software engineering into two categories, namely process and architecture, and choosing two key challenges in the practice of bioinformatics: reproducibility and scalability. The body of the thesis is an exploration of the intersection between these two software engineering categories and these two bioinformatics challenges. The question is asked: Can best practices in professional software engineering be applied to address key issues in the bioinformatics domain, creating positive outcomes? And …

New Methods For Deep Learning Based Real-Valued Inter-Residue Distance Prediction, Jacob Barger

Theses

Background: Much of the recent success in protein structure prediction has been a result of accurate protein contact prediction--a binary classification problem. Dozens of methods, built from various types of machine learning and deep learning algorithms, have been published over the last two decades for predicting contacts. Recently, many groups, including Google DeepMind, have demonstrated that reformulating the problem as a multi-class classification problem is a more promising direction to pursue. As an alternative approach, we recently proposed real-valued distance predictions, formulating the problem as a regression problem. The nuances of protein 3D structures make this formulation appropriate, allowing predictions …

Detecting And Characterizing Self Hiding Behavior In Android Applications, Raina Samuel

Theses

Applications (apps) that conceal their activities are fundamentally deceptive; app marketplaces and end-users should treat such apps as suspicious. However, due to its nature and intent, activity concealing is not disclosed up-front, which puts users at risk. This study focuses on characterization and detection of such techniques, e.g., hiding the app or removing traces, known as 'self hiding' (SH) behavior. SH behavior has not been studied per se - rather it has been reported on only as a byproduct of malware investigations. This gap is addressed via a study and suite of static analyses targeted at SH in Android apps. …

Integration Of Lightweight & Energy Efficient Cipher In Wireless Body Area Network Fore-Health Monitoring, Azza Zayed Aishamsi

Theses

There is an increase in the diseases of the circulatory system in United Arab Emirates, which makes it the first leading cause of death. This led to a high demand for a continuous care that can be achieved by adopting an emerging technology of e- Health monitoring system using Wireless Body Area Network (WBAN) that can collect patient’s data. Since patient’s data is private, securing the communication within WBAN becomes highly essential. In this research thesis, we propose an architecture to secure the data transmission within the Wireless Body Area Network (WBAN) in e-Health monitoring. More specifically, our proposed architecture …

Uav-Cloud: A Platform For Uav Resources And Services On The Cloud, Sara Yousif Mohamed Mahmoud

Theses

UAVs - Unmanned Aerial Vehicles – have gained significant attention recently, due to the increasingly growing range of applications. However, developing collaborative UAV applications using traditional technologies in a tightly coupled design requires a great deal of development effort, time, and budget especially for heterogeneous UAVs. Moreover, monitoring and accessing UAV resources using traditional communication media suffer from several restrictions and limitations. This research aims to simplify the efforts, reduce the time, and lower the costs of developing collaborative applications for distributed heterogeneous UAVs. In addition, the research aims to provide ubiquitous UAV resources access. A platform is proposed for …

Formal Specification And Refinement Of The Navigation Tasks Of Autonomous Robots, Eman Rabiah

Theses

Autonomous robots are hybrid systems whose role in our daily life is becoming increasingly critical. They are tasked with various activities requiring reliability, safety, and correctness of their software-controlled behavior. Formal methods have been proved effective in addressing development issues associated with these software qualities. However, even though autonomous robot navigation is a primordial function, there is no research dealing with enhancing reliability of the navigation algorithms. Thus, our focus is to investigate this type of algorithms, and specifically path planning, a fundamental and critical functionality supporting autonomy. We formally address the issue of enhancing reliability of the widely-used A* …

Reducing The Risk Of Software Cost Estimation, Shixian Yang

Theses

Inaccurate cost estimation is a well-known problem in software development. The common cost estimation models are point estimates that are unable to quantify uncertainties. Furthermore, it is difficult to calibrate the uncertainties in cost estimation due to the lack of information. The purpose of this thesis is to prove that probability techniques could be synthesized into COCOMO (Constructive Cost Model) to quantify uncertainties. Another aim is to find out how to get more insight on reducing the risk of cost estimation. In this thesis, some historical data is presented to show the variance in factors of COCOMO. Monte Carlo simulation …

Amvc: A Loosely-Coupled Architecture For Component-Based Clients, Brendan Lawlor

Theses

There is an architectural deficit in most currently undertaken rich client applications: In n-tier applications the presentation layer is represented as a single layer. This is as true of browser-based rich clients (Rich Internet Applications) as of rich desktop clients since both are designed along component-based, event-driven lines. The result of creating rich clients without sufficient application architecture underpinning is often tightly-coupled code of unnecessary complexity, with the associated high cost of maintenance. No commonly understood pattern is currently evident that would allow rich client components to be combined into a loosely coupled application. This paper describes the theory and …

Model-Based Instrumentation Of Distributed Applications, Jan Schäfer

Theses

Problems such as inconsistent or erroneous instrumentation often plague applications whose source code is manually instrumented during the implementation phase. Integrating performance instrumentation capabilities into the Model Driven Software Development (MDSD) process would greatly assist software engineers who do not have detailed knowledge of source code instrumentation technologies. This thesis presents an approach that offers instrumentation support to software designers and developers. A collection of instrumentation patterns is defined to represent typical instrumentation scenarios for distributed applications. A UML profile derived from these patterns is then used to annotate UML models. Based on suitable code generation templates, the annotated models …