Digital Commons Network^™

Enhancing Code Review Automation With Large Language Models Using Qlora Fine-Tuning And Rags, Sumukh Naveen Aradhya

Master's Projects

In this technological era where Artificial Intelligence and Machine Learning are revolutionizing various domains, Large Language Models (LLMs) are emerging as a very powerful tool. In the software development lifecycle, it is extremely important to have reliable code reviews to ensure security and maintain code quality. This project aims to survey various existing methodologies to aid creation of efficient code review automation agents and also research on ways to make this process more efficient. Parameter Efficient Fine-Tuning (PEFT) methodologies such as LoRA and QLoRA have been explored with an additional focus on a hybrid model that combines adaptive QLoRA with …

Real-Time Adaptive Framework For Topic Modeling In Social Engineering Attacks, Manav Bhasin

Master's Projects

Detecting social engineering attempts is crucial for security, as these threats are becoming more frequent and increasingly exploit human vulnerabilities. This research focuses on topic modeling using conversational data from Kevin Mitnick’s ”The Art of Deception” with dialogues that illustrate various social engineering strategies. The dataset comprises manually extracted and synthetically augmented conversations to ensure natural dialogue flow. Two methodologies are presented for utterance-level and global topic extraction: prompt engineering leveraging OpenAI’s GPT-4o-mini, characterized by few-shot learning and chain-of-thought prompting, and Quantized Low Rank Adaptation (QLoRA) utilizing Mistral’s 7B instruct model for efficient fine-tuning. Through experimentation and evaluation, this study …

Exploring The Red Sea Crisis’S Supply Chain Disruption Impacts On The Shipping Industry, Po-Hsing Tseng, Nick Pilcher

Journal of Marine Science and Technology–Taiwan

Since October 7, 2023, the Houthi militia in Yemen has continuously attacked Israel and its nearby ships in the Red Sea. This crisis has impacted on global ship operations, freight rates, and shipping networks, since this conflict seriously affects the international trade between Europe and Asia and brings many unpredictable butterfly effects (e.g. supply chain disruption, and high transportation costs in shipping and air cargo). This papero explores its impact on the supply chain and suggests potential solutions;, and to consider perceptions on who should be responsible for the crisis, and who will actually shoulder the responsibility for the crisis. …

In-Situ And Laboratory Sound Velocities Of Unconsolidated Sediments In The Western South Korea Plateau Of The East Sea, Kiju Park, Gil Young Kim, Yuri Kim, Bo Yeon Yi, Nyeon Keon Kang, Gwang Soo Lee

Journal of Marine Science and Technology–Taiwan

The in-situ velocity of unconsolidated shelf sediments in the western South Korea Plateau of the East Sea was directly measured using the Korea Institute of Geoscience and Mineral Resources (KIGAM) Seafloor Acoustic Probe. Physical properties, including laboratory velocity, were also measured for core samples collected from the same location. The laboratory measurements were conducted at a temperature of 23℃ and under atmospheric pressure. Owing to the in-situ temperature and pressure (0.5–1.5℃ and 100–180 atm, respectively) at the sampling site, the laboratory velocity was adjusted to the in-situ conditions. In addition, a comparison with a theoretical prediction model (Biot-Stoll model) was …

Key Success Factors In Using The Lifting Method To Construct A Steel Bridge Pylon, Ming-Chao Su, Hui-Ping Tserng

Journal of Marine Science and Technology–Taiwan

Constructing the inclined steel pylon of cable-stayed bridges is challenging because of the pylon’s height, spatial limitations, and safety concerns. To address these challenges, this study developed a modular lifting method that can enhance the efficiency and precision of pylon construction. The proposed method involves using rotation jigs to temporarily assemble the pylon, horizontal positioning on temporary supports, and synchronized lifting through computer-controlled hydraulic systems. Field implementation on a large-span bridge project revealed that compared with the traditional method, the proposed method resulted in higher welding quality, lower risks associated with great-height operations, and a shorter on-site assembly time. The …

Deep Learning-Based Model For Automated Prediction Of Coastal Changes: A Robust Approach To Environmental Forecasting, Tsair-Fwu Lee, Chu-Ho Chang, Chin-Shiuh Shieh, Chih-Hsien Wu, Jen-Chung Shao, Chien-Liang Chiu

Journal of Marine Science and Technology–Taiwan

The coastline stands as a critical domain encompassing industry and the environment. The escalating global warming, leading to elevated sea levels and intensified wave-current interactions, has given rise to substantial coastal erosion. This predicament, in conjunction with excessive human development and Taiwan's coastal areas' extreme climatic impact, accentuates the perceptibility of coastal beach alterations. As a result, coastal erosion has emerged as a pressing issue necessitating resolution. Traditional methodologies for assessing coastline changes have conventionally relied on manual measurements. However, owing to the fluctuating distance of coastlines, influenced by tidal patterns, extended measurement processes over several months are susceptible to …

Multimodal Imaging Of Protein-Based Biomaterials For Delivering Chemo-Agent Cargo For Glioblastoma Treatment In A Murine Model, Orin Mishkit

Dissertations and Theses

Protein-based self-assembling biomaterials, known as Thermo-Responsive Assembled Proteins (TRAP), present meaningful potential as drug delivery carriers. These materials enable the controlled, slow release of poorly soluble chemotherapeutic agents, such as doxorubicin (Dox), while potentially reducing systemic off-target effects. In collaboration with multiple NYU labs, this study evaluated the efficacy of two TRAP variants—TRAP and F-TRAP—as drug delivery carriers in a xenograft mouse model of glioblastoma multiforme (GBM).

The primary objective was to compare the effectiveness of TRAP-loaded Dox (TRAP-DOX) to free Dox in achieving tumor extravasation and accumulation, facilitating sustained drug release. We hypothesized that the leaky vasculature of GBM …

A Study Exploring The Use Of Artificial Ligaments To Enhance The Human-Robot Interface Via Force Distribution Manipulation In Biomimetic Cuff Systems, Adifet Cejovic

Dissertations and Theses

Comfort is among the main priorities when designing a wearable assistive device that needs to be used for long sessions. There is a growing need for wearable systems that provide aid to patients with an injured knee. In addition, there has been an effort to incorporate cable-driven designs rather than rigid attachments. This change in design decreases the overall weight of the system and can allow the cable to adapt to a user’s change in gait patterns. This thesis investigates a cuff system centered around the knee and the design of a testing rig. This system incorporates elements from pre-existing …

Adaptation Of The Biodent For Spherical Microindentation To Determine Elastic Modulus In Small Rodent Bones, Mixame E. Jerome

Dissertations and Theses

Mechanical and material properties of bone are essential for understanding both normal skeletal function and the progression of bone-related diseases. Microindentation has become a widely used method for evaluating tissue mechanical properties of bone, as it samples the contributions of microarchitecture and composition. Microindentation using a standard pointed tip inherently introduces both elastic and plastic deformation into the material. In contrast, indentation with a spherical tip can be operated entirely in the elastic range, providing for a non-damaging test that has significant advantages over sharp indentation techniques. Moreover, a key advantage of using spherical indentation lies in its operation under …

Explorations Of Dna-Single-Walled Carbon Nanotube Interactions To Develop Multiplexed Molecularly Specific Biosensors For Inflammation, Amelia K. Ryan

Dissertations and Theses

Inflammatory cytokines such as interleukin-6 (IL-6) and interleukin-12 (IL-12) are central regulators of immune signaling and key biomarkers of disease, yet existing assays for their detection remain slow, invasive, and lack multiplexing ability. This dissertation advances the development of single-walled carbon nanotube (SWCNT) optical nanosensors capable of real-time, multiplexed, and molecularly specific cytokine detection through innovative use of single-stranded DNA (ssDNA) interfaces.

First, an IL-6-specific DNA aptamer was employed as both a dispersing agent and recognition probe for SWCNT fluorescence sensing. Sequence modifications, including anchor domains, truncations, and thermally induced refolding, were systematically tested to optimize sensitivity and selectivity. The …

Advancing Electrical Stimulation: Full-Head Mri Segmentation For Abnormal Brain Anatomy With Tdcs, Andrew Birnbaum

Dissertations and Theses

Evaluating the effectiveness of transcranial direct current stimulation (tDCS) is essential for guiding its integration into therapeutic and performance-enhancement applications. In our laboratory, we investigate the efficacy of tDCS across multiple experimental models, including both animal and human studies. I have contributed significantly to the execution and analysis of these experiments, which include studies in rats and healthy human participants aimed at evaluating whether electrical stimulation of the motor cortex can enhance motor learning. These studies assess improvements in fine motor performance resulting from tDCS. In stroke patients, I contribute to our investigation of tDCS as a rehabilitative intervention, particularly …

Wearable Disposable Elecetrotherapy, Mohamad Fallahrad

Dissertations and Theses

Non-invasive electrotherapies offer promising treatments for a wide range of conditions such as pain, headaches, depression, addiction, cognitive decline, wound healing, and drug delivery. However, patient compliance remains low due to the high initial cost, cumbersomeness and inconvenience of existing technology compared to pharmaceuticals. Current electrotherapy devices are often bulky and require users to connect disposable electrodes, initiate programs, and manage charging cycles, which hinders widespread adoption.

We address these challenges by designing and validating a novel electrotherapy platform that is single-use, disposable, low-cost, and wearable. Activated automatically upon skin contact, the device delivers a preset dose of controlled electrical …

An Environmental And Durability Study Of Bonded Structural Patch Repairs With Finite Element And Machine Learning Enabled Predictions, Yesim Kokner

Dissertations and Theses

Aircraft and vehicles experience substantial structural and aerodynamic stresses over their operational lifetimes. These stresses can cause damage or structural weakening, especially in aging military and civilian aircraft and vehicles. Depending on the severity of the damage, three options are available to alleviate the damage in these structures or parts: complete replacement of the aircraft or vehicle, replacement of the damaged part, or temporary repair of the damage.

The goal of this study is to evaluate damaged specimens repaired with an adhesively bonded patch subjected to quasi-static and fatigue loads under various environmental conditions. The patch repair technique is widely …

Development Of An Quadcopter Unmanned Aerial Vehicle For Atmospheric Remote Sensing, Omar J. Addasi

Dissertations and Theses

This paper explores the development of a quadcopter unmanned aerial vehicle (UAV, a.k.a. drone) for atmospheric remote sensing of temperature, pressure, humidity, and PM2.5 particulate matter quantities. A 3-D printed drone body is designed and flight tuning is performed. An optimal length for an upward extending mast for the drone body is determined. Low cost, lightweight sensors are compared against higher precision sensors in both static and dynamic conditions. In addition, an initial investigation into the design of a pressure sensor based anemometer is performed.

From Tpu To Qpu: Bridging Fidelity Gaps Across Next-Generation Computing Systems, Mohammad Walid M. Charrwi

Dissertations and Theses

Modern computing systems continue to advance rapidly, enabling unprecedented computational capabilities across diverse domains. Among these powerful architectures are AI accelerators such as Tensor Processing Units (TPUs) and quantum computing systems utilizing Quantum Processing Units (QPUs) stand out as trans-formative technologies. Despite their potential, these systems share significant challenges related to hardware infidelity, manifested as error-induced degradation in computational accuracy and reliability. This thesis explicitly addresses cost-effective methods to screen-out defective AI accelerators across next-generation computing architectures and boost quantum circuit output fidelity through hierarchical circuit optimization using advanced machine learning methodologies.

This thesis first proposes a cost-effective testing methodology …

Renewable Energy Integration In Nyc Subway: Solar Panels (Under Railroads And At The Subway's Edge) & Vertical Axis Wind Turbines (Vawts) Near Tracks, Pengdwinde Olivier Saba

Dissertations and Theses

As urban populations continue to expand, the demands on public transit systems in densely populated regions like New York City are escalating. Multiple cities face the dual challenge of increasing ridership and the imperative to minimize their environmental footprints.According to Law 97 in NYC, the goal is to reduce emissions by 40% and to reach net zero by 2050. An innovative solution will need to be integrated into the NYC environment, such as the subway system, where there are a potential unuse space, building rooftop, since we know that spaces in New York City are very critical, such as these …

Design And Lab Implementation Of An Air Source Transcritical Co2 Heat Pump, Geoffrey Turbeville

Dissertations and Theses

A laboratory-scale air-source transcritical CO2 heat pump (TCHP) was developed to evaluate the performance of low-global-warming-potential (GWP) refrigerants for cold climate applications. A custom climate chamber was also constructed to simulate winter weather conditions typical of the northeastern United States. Both the TCHP and the chamber were fully instrumented to collect temperature, pressure, and mass flow data, which were used to calculate the co- efficient of performance (COP) in accordance with ASHRAE standard refrigerant enthalpy methods. Prior to testing the CO2 system, the experimental setup was benchmarked using a commercially available R410A heat pump manufactured by LG.

The climate chamber …

A Novel Dual-Imaging Approach For Visualizing Multiphase And Droplet Interactions Including Shock-Driven Flows, Tad Jerzy Misztal

Dissertations and Theses

Shock-droplet/particle interactions have garnered substantial interest in recent years, fueled by advancements in aerospace and defense technologies, including supersonic and hypersonic systems in highly humid conditions, as well as the ongoing space exploration efforts from global space centers as well as private space companies. Understanding the interactions between shockwaves and droplets that impact on solid surfaces is critical due to the potential to cause significant damage to materials and structures in high-speed flow environments. Studies like these have applications in areas ranging from aerospace engineering to blast mitigation systems, where these interactions play a pivotal role in designing the relevant …

Experimental Characterization Of Heat Transfer And Fluid Flow In Tcr Fuel Elements Under Varying Operational Conditions, El Mokhtar Majdoub Em

Dissertations and Theses

ABSTRACT

The Transformational Challenge Reactor (TCR) marks a significant leap forward in nuclear reactor innovation, focusing on modular construction, high efficiency, and improved safety. This research explores the coupled behavior of heat transfer and fluid flow within the TCR's 3D-printed ceramic core under varying operational conditions. Nitrogen was utilized as the cooling medium, with experiments carried out at pressures ranging from 300 to 600 psi and temperatures spanning 100°C to 300°C. Critical parameters, including inlet and outlet gas temperatures, wall temperatures at both ends of the core, and coolant flow rates were monitored to evaluate thermal performance and system reliability. …

Virtual Environment Creation And Camera Calibration For Soft Target Identification And Assistance In Crowded Spaces With A Sensor Network And A Robotic Dog, Eltan Samoylov

Dissertations and Theses

Crowded places are increasingly targets of violence due to the increased accessibility and covertness of weapons, explosives, and other technology like drones. Addressing the challenges of protecting crowded places and assisting vulnerable individuals requires a multidisciplinary approach, taking inspiration from many different perspectives. Video surveillance of these crowded public facilities, such as train and bus stations, airports, shopping malls, and sports arenas, is very important to public safety, both for identifying threats/terrorist attacks and implementing evacuation plans.

The work of this thesis is part of a larger project aiming to explore the potential of using real-time computer vision and deep …

Cfd Modeling And Simulation Of Natural Circulation In Helium- Cooled Very High Temperature Reactors, Mohammad Sakib U. Abrar

Dissertations and Theses

This thesis investigates the thermal-hydraulic behavior of natural circulation in helium-cooled Very High Temperature Reactor (VHTR) systems, focusing on the performance and validation of passive safety features under Pressurized or Depressurized Loss of Forced Cooling (P-LOFC or D-LOFC) events. An experimental facility inspired by the General Atomics 350 MWt Modular High Temperature Gas-Cooled Reactor (MHTGR) was developed at The City College of New York. The setup consists of a plenum-to-plenum flow loop with four riser and four downcomer tubes, each equipped with heating coils, thermocouples, and pressure sensors. Natural circulation is driven by density gradients induced by asymmetric heating of …

Characterization Of The Intrinsic Neural Timescale Indices As A Reliable Fingerprint Of Brain Dynamics, Helya Honarpisheh

Dissertations and Theses

Intrinsic neural timescales (INTs) characterize the persistence of neural activity over time and offer insight into the brain’s temporal hierarchy. This thesis evaluates and compares six autocorrelation-based INT metrics—AR(1) Coefficient (Autoregressive Coefficient at 1), First Zero Lag, Lag at ACF = 1/e, Sum of Positive ACFs, ACF at Half Lag, Exponential Decay Constant (τ)— using resting-state fMRI data from 400 cortical regions across a large, lifespan sample (ages 6– 85) from the NKI-Rockland dataset. A discriminability analysis revealed that the AR(1) coefficient provided the highest test-retest reliability and individual specificity across repeated scans, establishing it as the most reliable INT …

Metasurfaces And Their Applications In Photonic Devices, Shuwei Guo

Dissertations and Theses

Metasurfaces are flat artificial optical elements composed of dielectric and/or metallic nanostructures. They can manipulate light in unprecedented ways by altering amplitude, phase, and polarization at subwavelength scales. The ability of these two-dimensional elements to perform multiple optical functions within compact optical systems—while offering high functionality, small form factors, and easy integration into optoelectronic devices—has sparked significant interest across various fields and industries. These applications span imaging, sensing, nonlinear and quantum optics, optical computing, automotive technology, augmented and virtual reality, and more.

In this dissertation, we explore and design new multifunctional metasurfaces with a focus on manipulating spectral responses. First, …

Cognitive Map Generation For Vision And Language Navigation, Alexander Sandoval Mesa

Dissertations and Theses

Visual-Language Navigation (VLN) presents significant challenges for autonomous agents, such as robots and virtual assistants, particularly in complex, dynamic environments where the seamless integration of visual perception and natural language understanding is critical. Traditional VLN systems often struggle with effectively aligning language instructions and visual scene understanding, limiting their adaptability and navigation efficiency.

This thesis proposes a novel Cognitive Map-based framework that addresses these challenges by transforming natural language navigation instructions into structured graph representations. The Cognitive Map consists of nodes representing waypoints, landmarks, decision points, and edges encoding spatial relationships and navigational actions. These maps are generated using Large …

Multi-Material Adhesive Bonded Joints Under Static And Dynamic Loading, Gizem Derya Demir

Dissertations and Theses

Adhesively bonded joints present significant potential for reducing structural weight and enhancing performance in modern and future engineering structures. By enabling the elimination or reduction of traditional mechanical fasteners and facilitating the integration of dissimilar materials, adhesives can contribute to more efficient, lightweight designs. However, to fully leverage these benefits—especially in high-performance applications—adhesive materials must exhibit robust behavior under both static and dynamic loading conditions, such as ballistic impact. Despite their promise, the failure mechanisms of adhesives under such high-strain-rate scenarios remain insufficiently understood.

This dissertation investigates the performance of adhesively bonded multi-material joints subjected to static loading and ballistic …

Data Driven Bridge Deck Deterioration Modeling And Maintenance Intervention Scheduling, Deepak Kumar

Dissertations and Theses

Majority of over 617,000 bridges across the United States are significantly impacted by environmental factors, aging materials, and heavy traffic loads. According to the 2021 Infrastructure Report, nearly 231,000 bridges require repair or preservation work, with 46,154 (7.5%) being classified as structurally deficient, thereby posing risks for mor than 178 million daily trips across these bridges. Despite recent improvements, 42% of U.S. bridges are at least 50 years old, underscoring an urgent need for increased investment to meet repair demands. The nation’s backlog of bridge repairs is estimated at $125 billion, with a 58% increase in annual spending required to …

Generation And Characterization Of Human Blood-Brain Barrier Models For Investigating Neuropsychiatric Disorders And Tumor Metastasis, Yunfei Li

Dissertations and Theses

The blood–brain barrier (BBB) is important in the normal function of the central nervous system (CNS). An altered BBB has been described in various neuropsychiatric disorders. The brain-specific microvascular endothelial cell (BMEC) is an essential structural component of the BBB. To test if integrity of the BBB formed by BMECs is compromised in 22q11.2 deletion syndrome (also called DiGeorge syndrome), which is one of the validated genetic risk factors for schizophrenia, a 2D iBBB (induced BBB) on a Transwell filter was generated from human microvascular endothelial cells (HBMECs) derived from the induced pluripotent stem cells (iPSCs) out of patients with …

Experimental And Analytical Investigation Of Particle Dynamics In Newtonian And Non-Newtonian Fluids In A Taylor-Couette Cell, Andres F. Velez Mendoza

Dissertations and Theses

This doctoral dissertation investigates heavy particle sedimentation in Newtonian and non-Newtonian fluids, aiming to understand their behavior in Drilling Fluid (DF) systems used in oil and gas industry, where such phenomenon can lead to operational issues such as lost circulation, well control problems, and even blowouts. To simulate DF dynamics under operational conditions, this study utilizes a Taylor-Couette (TC) cell, which consists of a rotating inner cylinder and a stationary outer cylinder separated by a 9.0 mm annular gap. The TC cell allows for the investigation of the sedimentation behavior of spherical particles within both Newtonian fluids (deionized water and …

Deep Learning Architectures For Improving Weather Research And Forecasting (Wrf) Precipitation Estimates And Landslide Susceptibility Mapping: A Case Study Of Puerto Rico, Said A. Mejia Manrique

Dissertations and Theses

Precipitation is a critical factor in the Earth’s hydrological cycle and has a profound impact on various environmental and socioeconomic systems, including management of water resources, flood forecasting, and the evaluation of natural hazards, particularly those associated with rainfall-induced landslides. For the latter, obtaining precise precipitation estimates is essential, as intense rainfall events can trigger landslides that pose significant risks to human life, infrastructure, and ecosystems, particularly in mountainous regions. Puerto Rico (PR), with its mountainous terrain, tropical climate, and vulnerability to hurricanes, provides a unique environment for studying precipitation patterns and the need for accurate rainfall estimates. The complex …

In Vitro Comparative Study Of Composite Coatings For Magnesium-Based Bone Implants, Abdelrahman Amin, Bryce Williams, Thomas Mcgehee, Alyssandra Navarro, Vipul Patil, Mostafa Elsaadany, Hanna Ibrahim

Biomedical Engineering Faculty Publications and Presentations

This study provides a comparative assessment of the corrosion, morphology, and biological properties of four of the most promising coating methods for magnesium in biomedical applications, namely micro-arc oxidation (MAO), graphene oxide (GO)-containing MAO coating, MAO/sol-gel composite coating, and MAO/polycaprolactone (PCL) polymer dip coating approaches. Despite the abundance of research focused on these promising coating approaches, there is a notable lack of comparative studies evaluating their properties. To this end, the investigated composite coatings are ZK60+MAO, ZK60+MAO/GO, ZK60+MAO/Sol-gel, and ZK60+MAO/PCL composite coating. Several tests were used to assess various properties of the prepared groups such as in vitro corrosion characteristics, …