2025 Graduate Student Research Symposium

Follow

• 

  Block copolymer-mediated synthesis of TiO2/ RuO2 nanocomposite for Efficient Oxygen Evolution Reaction

  K. C. Raj Binod, Bishnu Bastakoti Ph.D., and Dhananjay Kumar Ph.D.

  An amphiphilic block copolymer, poly (styrene-2-polyvinyl pyridine-ethylene oxide), was used as a structure-directing and stabilizing agent to synthesize TiO2/RuO2 nanocomposite. The strong interaction of polymers with metal precursors led to formation of a porous heterointerface of TiO2/RuO2. It acted as a bridge for electron transport, which can accelerate the water splitting reaction. Scanning electron microscopy, energy-dispersive X- ray spectroscopy, transmission electron microscopy, and X-ray diffraction analysis of TiO2/RuO2 samples revealed successful fabrication of TiO2/RuO2 nanocomposites. The TiO2/RuO2 nanocomposites were used to measure electrochemical water splitting in three- electrode systems in 0.1-M KOH. Electrochemical activities unveil that TiO2/RuO2-150 nanocomposites displayed superior oxygen evolution reaction activity, having a low overpotential of 260 mV with a Tafel slope of 80 mVdec−1

• 

  Transforming Food Bank Operations with Data Visualization: A Study on Neighbor Food Preferences

  Enoch Bonsu and Steven Jiang Ph.D.

  Food insecurity remains a critical global challenge, and food banks play an essential role in addressing this issue by ensuring food reaches those in need. However, operational inefficiencies, inequitable distribution, and a lack of data driven decision-making often hinder their effectiveness. This research develops an advanced visualization framework that transforms raw data into actionable insights, enabling food banks to optimize resource allocation, enhance equity, and improve operational efficiency. By integrating demographic, socio-economic, and geographic data, this study introduces interactive dashboards that provide real-time intelligence on food preferences, service area coverage, and allocation disparities. The framework employs intuitive navigation, standardized visual hierarchies, and structured categorization to enhance usability for decision-makers. Furthermore, census tract- level mapping and heatmaps reveal spatial trends, helping identify underserved communities and gaps in food distribution strategies. Beyond operational improvements, this work establishes a scalable, replicable model for humanitarian logistics, demonstrating how data visualization can bridge the gap between analytics and real-world impact. Future work will focus on developing a standardized visualization framework that serves as a go-to reference for food banks, policymakers, and humanitarian organizations, enabling them to adapt and customize data-driven insights for their specific needs

• 

  Soft Error Reliability Aware EDA

  Morgan Brown

  This review explores advancements in GPU-accelerated EDA tools, such as DREAM Place and GPU-based Rectilinear Steiner Tree generation, which leverage AI and parallelism to optimize placement, routing, and logic synthesis tasks. These tools significantly reduce runtime and enhance placement accuracy, facilitating faster design iterations and improving fault tolerance by minimizing soft error hotspots. Additionally, AI-driven frameworks, including deep learning-based layout optimization, show promise in addressing pin density challenges, congestion, and manufacturability constraints. While these innovations enhance performance, gaps remain in ensuring radiation-hardened reliability for aerospace and military applications. Existing research primarily focuses on optimizing computational efficiency, but further investigation is needed into integrating fault-tolerant microcontroller design within AI-accelerated EDA workflows. Specifically, challenges in mitigating Single Event Upsets (SEUs) and enhancing real-time processing capabilities for mission-critical hardware require deeper exploration. Future work should focus on bridging the gap between AIdriven optimization and the development of robust, radiation-resistant microcontrollers to ensure reliable operation in extreme environments

• 

  A Microphysiological Model for Investigating Aβ-Induced Microvascular Dysfunction and Immunotherapeutic Impact

  Charity Campbell, Yeoheung Yun, and Samuel Uzoechi

  Anti-amyloid immunotherapies represent a promising avenue for Alzheimer’s disease (AD) treatment by targeting amyloid-beta (Aβ); however, their administration is associated with adverse cerebrovascular effects, including ARIA-E/H (edema/hemorrhage) and Cerebral Amyloid Angiopathy (CAA). Despite growing evidence implicating Aβ in microvascular pathology, the mechanisms governing its impact on the neurovascular unit remain unknown. This study employs a microfluidic-based Familial Alzheimer’s Disease-Brain Microphysiological System (FAD- BMPS) to elucidate Aβ-mediated blood-brain barrier (BBB) disruption, vascular amyloid deposition, and inflammatory cascades. Our findings indicate that Aβ compromises endothelial integrity by destabilizing tight junctions, increasing BBB permeability. While promoting the secretion of tight junction and adhesion proteins. Elevated Aβ concentrations correlate with an upregulation of pro-inflammatory cytokines and chemokines, including, TNF-α, and MMP-9. To further investigate the therapeutic implications, this model will assess the impact of inflammatory function on BBB stability and ARIA-related pathologies. We propose that such interventions modulate endothelial function, exacerbating vascular permeability and inflammatory responses, ultimately contributing to ARIA pathogenesis. The FAD-BMPS offers a platform for dissecting Aβ-induced vascular pathology and evaluating therapeutic interventions, granting mechanistic insights into CAA and ARIA.

• 

  Exploring the Use of Generative Artificial Intelligence for Bias Mitigation

  Alexis Cathcart

  Artificial intelligence (AI) and related topics are growing in popularity across various industries as there is a desire to improve the accuracy and efficiency of decision-making processes. The algorithms at the foundation of such technologies are initially developed and trained on datasets, then implemented in real-world applications that can directly affect humans, such as in healthcare, criminal justice, and finance. However, with this power comes the potential for a lack of fairness, an issue that is becoming a prominent concern in the realm of AI research. While the development of AI technology is on the rise, there is a need to ensure that the algorithms are constructed in a way that eradicates the influence of unfairness and bias, especially when considering the potential of outcomes that negatively impact marginalized communities. One approach that has been considered is the usage of generative AI, such as generative adversarial networks (GANs). The tools can not only be employed to generate data for computer vision problems, especially when data is limited and/or difficult to obtain, but also address concerns surrounding fairness and bias in AI. Additionally, the tools can potentially be implemented in COVID-19 research to better understand the virus’s impact in marginalized communities

• 

  Predicting the Sheet Resistance of Titanium Oxynitride Thin Films using AI Tools-Machine and Deep Learning

  Sheilah Cherono and Ikenna Chris-Okoro

  The objective of the project is to employ AI tools such as machine learning (ML) and deep learning (DL) to predict the sheet resistance of titanium oxynitride (TiON) thin films based on input features such as deposition parameters, film composition and thickness. Sheet resistance is a critical property for thin films in electronic applications. The complex relationships between material properties and deposition parameters calls for a shift from the traditional paradigm of experimental science to the modern paradigm of data exploration. The research focuses on the growth of high-quality titanium oxynitride thin films, using pulsed laser deposition (PLD) method, which allows control of growth parameters, leading to formation of high quality epitaxial thin films with precise control of its electrocatalytic properties. The films are used as catalysts to examine the electrochemical reactions during water splitting, with an aim of producing hydrogen, which is a source of renewable energy. MATLAB regression learner App together with experimental data has been used to train the models. The utilization of AI tools, along with the algorithmic processing of experimental data, has the potential to support data analysis and it can facilitate the systematic correlation of material structure and properties.

• 

  TinyML and Reliability: Does Quantization affect the Reliability of Machine Learning Models?

  Deriech Cumming II

  Machine Learning (ML) ability to handle and automate complicated tasks allows for its wide application (facial recognition, predictive text, ChatGPT, etc.). TinyML is a field that aids in ML’s shortcomings (ML is complex and memory expensive) finding ways to compress these large models, which widens the scope of application that ML can aid. One such scope involves mission-critical tasks (self-navigation, healthcare, manufacturing, etc.) which have catastrophic consequences for failure in these tasks. Ensuring we can safely implement these machine learning applications such that they run consistently is important.

• 

  Ag-decorated copper microsphere for electrochemical reduction of CO2 into methane

  Rabin Dahal

  We synthesized the copper microsphere decorated with silver nanoparticles using a hydrothermal process followed by photoreduction of Ag ions into Ag nanoparticles. Sub-100 nm Ag nanoparticles anchored on the surface of Cu microspheres improve the electrochemical performance and selectivity of the CO2 reduction into CH4. By enhancing the conductivity and active site of the catalyst and lowering the charge transfer resistance, Ag nanoparticles on Cu accelerate the rate of CO2 reduction. The faradaic efficiency of methane in a copper microsphere coated with silver nanoparticles was 70.94%, about twice as high as that of a copper microsphere (44%). Higher catalytic performance, stability, and faradaic efficiency of silver decorated copper microspheres were noted in the electrochemical reduction of CO2.

• 

  AI-Powered Intrusion Detection for UAV Security

  Subhram Dasgupta and Kushal Badal

  This poster presents an Artificial Intelligence (AI) powered Intrusion Detection System (IDS) for Unmanned Aerial Vehicles (UAV) security. The system leverages ML and DL algorithms to detect cyber and physical attacks on UAVs, with plans to implement a Federated Learning (FL) framework and LLM's for performance enhancements

• 

  Robust Adaptive Control Technology to Mitigate Radiation Signal Noise for Robots in Radioactive Environments

  Francis Djirackor

  The deployment of robotic systems in nuclear stations plays a crucial role in monitoring and maintaining safety standards. However, one of the most significant challenges in such environments is the degradation of control signals due to radiation-induced interference. This paper explores novel robust control methodologies that can enhance the stability and accuracy of robotic operations in radioactive environments. By integrating advanced control strategies such as adaptive control, sliding mode control, and interference cancellation techniques, this study aims to propose an efficient framework for reducing signal corruption caused by ionizing radiation. The paper also highlights critical research gaps and presents practical implementations based on state-of-the-art technologies.

• 

  Proliferation of Industrial Internet of Things (IIoT) deployments

  Josephine Djirackor and Francis Djirackor

  The proliferation of Industrial Internet of Things (IIoT) deployments utilizing LoRaWAN technology has powered significant advancements in remote monitoring and data acquisition. However, widespread adoption remains hindered by persistent security vulnerabilities. Existing LoRaWAN security frameworks, primarily reliant on key-based encryption, are susceptible to evolving cyber threats, necessitating innovative countermeasures. This research introduces a novel security paradigm leveraging the inherent natural frequencies of individual network agents (sensors, actuators) to augment traditional key-based security. Incorporating unique frequency signatures as dynamic authentication parameters, we propose a multi- layered security architecture that enhances network resilience against unauthorized access and data manipulation. This approach exploits the intrinsic physical characteristics of each device, rendering it exceptionally challenging for malicious actors to replicate or spoof legitimate nodes. An analytical model that demonstrates the feasibility and effectiveness of this frequency-based security mechanism, considering factors such as environmental noise, device variability, and network scalability shall be developed. Simulations and preliminary experimental data validate the proposed methodology, showcasing substantial increase in network security without compromising LoRaWAN’s low-power, long-range capabilities.

• 

  Gold Nanoparticles Conjugated with Iodinated Copolymers as a Potential Dual X-ray Imaging Contrast Agent

  Aiyanna Drakes

  Gold nanoparticles and radiopaque iodinated polymers have great potential in the field of bioimaging due to their high-contrast X-ray imaging properties. Here, we report the synthesis of iodinated copolymers, poly(2-[2’,3’,5’-triiodobenzoyl]oxoethyl methacrylate-co2- hydroxyethyl methacrylate) using reversible addition-fragmentation chain transfer polymerization and the preparation of gold nanoparticles conjugated with the iodinated copolymers. GPC confirmed the copolymer with an Mn of 14,259 g/mol and a polydispersity index of 1.53 while the EDX analysis showed that the copolymer contains about 27% iodine. Gold nanoparticles conjugated with the copolymer were prepared via in situ approach, which produced nanoparticles with average sizes of 28 nm as measured by DLS. SEM visually confirmed the formation of spherical gold nanoparticles whose sizes agreed with that of the DLS measurement. The UV-Vis spectrum of the gold nanoparticles showed a strong absorption at λ 530 nm, which was attributed to the nanoparticle surface plasmon resonance. The work presented here illustrates a new platform for developing a potential dual X-ray imaging contrast agent that could be used in bioimaging and allied fields

• 

  Navigating the Truck Parking Challenge: A Comprehensive Review of Strategies, Emerging Technologies, and Future Directions

  Nana Duahn and Hieu Nguyen Ph.D.

  Freight transportation systems play a vital role in the economy of the United States, making adequate truck parking essential for safe and efficient operations. However, a significant gap between truck parking demand and supply has led to challenges including road safety risks, regulatory noncompliance, operational inefficiencies, and environmental impacts. Despite their importance, comprehensive reviews of potential solutions for truck parking are limited. This study addresses this knowledge gap by reviewing the current truck parking management approaches, focusing on intelligent truck parking information systems, data prediction models, and truck pattern analysis. The analysis encompasses various technological solutions, including the sensing infrastructure, GPS technology, and information dissemination methods. The review also examines emerging solutions in smart parking systems and the potential adoption of these solutions to address truck parking challenges. Furthermore, the current challenges in the truck parking problem, the possible solutions, and future scopes in its implementation are also presented.

• 

  Identification of Allergenic Peanut Protein and Peptides Resistant to Alcalase Hydrolysis

  Mahshid Eghbali and Seyi Adebayo

  Peanut allergy is one of most severe and persistent food allergies. While Alcalase hydrolysis significantly reduces major allergenic proteins, residual allergenicity remains a concern. This study aimed to identify resistant proteins/peptides contributing to the allergenicity of extensively hydrolyzed peanut protein concentrate (PPC). Peanut protein concentrate (PPC) at concentration of 10% was hydrolyzed with 4% Alcalase for 2-8 hours. The degree of hydrolysis (DH) increased from 34.91% to 44.68% accompanied with decreased IgE-binding. However, the SDS-PAGE and Western blot of both supernatants and precipitates with pooled sera from 7 peanut sensitive patients confirmed the presence of resistant allergenic peptides in the extensively hydrolyzed PPC, particularly, the two proteins/peptides with molecular weights 22.5kDa and 12.65kDa. To further characterize these resistant proteins/peptides, gel samples of these two peptides were sent a commercial service lab for sequencing using liquid chromatography-mass spectrometry (LC-MS/MS). The sequences obtained do not match any sequences of the native proteins in the PPC. This suggests that they are the peptides formed during Alcalase hydrolysis of larger proteins. Findings indicate that Alcalase hydrolysis alone is insufficient to eliminate peanut allergenicity. Further strategies are necessary to break down or mask all immunoreactive epitopes, enhancing the safety of peanut-based products.

• 

  A Spatial-Temporal Attention Deep Reinforcement, A Smart HVAC Control System for Energy Savings and Comfort

  Abdullahi Elnaiem

  Heating, Ventilation, and Air Conditioning (HVAC) systems play a vital role in maintaining indoor comfort but are among the largest energy consumers in buildings. Traditional HVAC controllers often operate inefficiently, failing to adapt to changing conditions such as weather variations, occupancy shifts, and electricity pricing. To address this, we introduce STA-DRL (Spatial-Temporal Attention Deep Reinforcement Learning), a framework designed to optimize HVAC performance in multi-zone buildings. STA-DRL combines convolutional neural networks (CNNs) to model heat transfer between rooms, long short-term memory (LSTM) networks to capture time-dependent factors, and an attention mechanism to prioritize energy distribution based on occupancy and cost fluctuations. This allows real-time adjustments to HVAC setpoints, improving efficiency while maintaining occupant comfort. By dynamically balancing energy use and comfort, STA-DRL offers a scalable solution for smart energy management, paving the way for more sustainable and cost-effective building automation. Future research will focus on real-world implementation and integration with smart infrastructure

• 

  Investigating Evolutionary Pressures: Uncovering Adaptive Trade-offs of Iron Stress and Phage Resistance in E. coli B

  Franklin Chikezie Ezeanowai, Jude J. Ewunkem, Joseph Graves Jr., and Liesl Jeffers-Francis

  We subjected Escherichia coli B to 35 days of experimental evolution under elevated iron(III) sulfate and T4 phage stress, performing genomic, phenotypic, and cross-resistance analyses. Ten replicate populations were grown in LB broth with or without 1500 mg/L iron(III), with or without phage. All populations underwent daily transfers, ensuring consistent selective pressures. Whole genome sequencing revealed mutations that reached fixation or sweeping frequency. Iron(III)-selected strains showed enhanced tolerance at 1000–1750 mg/L iron but experienced reduced susceptibility thresholds to sulfanilamide, silver nitrate, and copper(II) sulfate, highlighting significant trade-offs. Conversely, iron(III) adaptation conferred cross- resistance to iron(II) and gallium(III) salts, likely reflecting overlapping uptake and detoxification pathways. Phage-mediated selection further shaped these trade-offs, occasionally increasing resistance to certain antibiotics while diminishing metal tolerance. Convergent mutations in hchA (C→A at position 144564) emerged across multiple regimes, implying a broad adaptive role in stress response and protein quality control. Large deletions in phage receptor-related genes appeared consistently under phage pressure, underscoring receptor modifications for phage resistance. These findings clarify how metals, phages, and antibiotics can jointly influence evolutionary trajectories, offering insights into effectively countering multidrug resistance.

• 

  GIS-based Analysis of Flooding Impacts on the Coastal Plains of Neuse River Basin, North Carolina

  Yared Gebregziabher

  Recurrent flooding of the Neuse River in North Carolina has been severely impacting the socio-economic conditions of the coastal plain in the River Basin. Effective flood forecasting and impact analysis are critical for planning emergency preparedness and mitigation measures. FEMA prepared 100-year return period flood risk maps for most of the study area. To enhance safety, this research focuses on modeling 500-year return period flood, assessing its impacts on human and economy, devising flood evacuation system, and proposing mitigation measures. HEC-RAS and GIS tools were used to conduct the study, with model parameters calibrated and validated using flood maps from FEMA. Flood impacts on the population, public infrastructures and land use were analyzed. Results indicate that over 100,000 acres of land across six counties would be inundated, affecting key infrastructures including interstate highways, major roads, schools and land parcels. A flood evacuation system was analyzed, and an optimized road network and public shelters like hospitals and schools are proposed for emergency response. To mitigate impacts, the study recommends community awareness programs, flood protection infrastructure, and nature-based solutions. The study demonstrates that integrating flood hydrodynamic modeling with spatial analysis provide effective solutions for mitigating flood risks and minimizing socio-economic impacts.

• 

  Direct Antenna Binary Phase-Shift Keying through Ferrimagnetic Loading

  Shantu Ghose

  This study proposes the use of Yttrium Iron Garnet (YIG), a ferrimagnetic material, in the form of a small ferrite rod, to load an X band open-ended waveguide antenna. It was observed that the phase of the radiated signal from the YIG-loaded antenna can be tuned by adjusting the magnetic bias field strength. By selectively switching between two magnetic bias fields that give 180-degree phase difference in the radiated signal, a direct antenna binary phase shift keying (DA-BPSK) is realized. Simulation results from HFSS and ADS are used to demonstrate the feasibility of the idea. Such direct antenna modulation (DAM) system can be instrumental for on-chip electrically small antennas.

• 

  Framework, Microbial community transcriptomics, Quorum Sensing

  Charity Greene, Godfrey Ejimakor, and Omowunmi Ode

  Both documented and undocumented immigrant farm labor play important roles in the US agricultural sector. Many livestock producers and meat processors rely on the immigrant labor pool for their operations. Anticipated changes in immigration policy at the national level could result in a reduction in the supply of such workers. A reduction of farm workers at the farm level is expected to result in an increase in the price of livestock such as hogs. The expected increase in the price of whole livestock is expected to result in a decrease in the supply of processed meat. The anticipated labor shortage is also expected to result in additional reductions in the supply of processed meat. This chain of events is expected to result in more volatility in the prices of livestock products at both the retail and farm levels. Producers who are able to access information on such price and quantity changes will be in a better position to plan production activities in order to take advantage of the changes. We use information on the farm-level demand for hogs to illustrate the expected impact of immigrant labor shortages on farm-level price, production and income. We assume labor shortages that could result in 5%, 10%, and 15% and 20% reductions in supply of hogs. We conclude that due to the inelastic nature of the farm-level demand for hogs, producers of hogs could benefit in the form of increased revenues if labor shortages result in price increases for hogs.

• 

  Generative-AI for Air Traffic Control Enhancement: Advancing Safety and Communication Efficiency for Autonomous & Manned Vehicles

  Stefan Green, Everett-Alan Hood, Na’im Baker, and Hamidreza Moradi

  With air travel expected to surpass 4 billion passengers post 2025, the demand for efficient air traffic management is growing. Air Traffic Control (ATC) coordinates flights, manages air and ground traffic, and ensures optimal scheduling, but controllers often handle up to thirty aircraft at once, creating significant workload challenges. Generative AI integration in ATC offers a solution by enhancing scheduling, optimizing routing, and improving crew management. By automating key processes, AI can reduce cognitive load, increase safety, and make airspace management more adaptive and scalable, ultimately ensuring safer and more efficient operations.

• 

  Engineering Synthetic Beneficial Biofilms: A Comprehensive Framework for Diagnostic and Therapeutic Applications

  Nona Hashem

  Biofilms, microbial communities adhering to surfaces, account for 70% of human microbial diseases. While their harmful effects are well-studied, growing interest in beneficial biofilms spans applications from biological control to industrial corrosion prevention. However, their complexity challenges the development of a systematic engineering framework to harness their potential. Synthetic Beneficial Biofilms (SBBs) offer a novel approach for various applications, particularly in healthcare. These engineered microbial communities can mimic harmful biofilms for treatment studies or be synthesized for cell therapeutics and drug delivery, aiding in diagnosing and treating gut microbiome diseases. We present an innovative interdisciplinary framework combining mechanical and bioengineering principles to control programmable biofilm production. This approach identifies crucial steps from selecting interaction surfaces to designing bacterial communities and ensuring their functionality. It outlines key phases of biofilm formation preformation, formation, and post- formation each with specific target genes and modulation criteria. Emerging technologies are integral to this framework, including nextgeneration transcriptomics, 3D bioprinting, light- controlled genetic devices, and non-invasive real-time monitoring. This comprehensive framework can unlock the full potential of SBBs, from diagnostic to therapeutic approaches.

• 

  Machine Learning-based Intrusion Detection and Explainable AI in IoT Networks

  Amber Hoenig, Laila Holland, and Alexander Nanor

  Machine learning-based intrusion detection systems (IDS) are one of the most vital and widely used defenses for Internet of Things networks. However, the ever-changing landscape of cyberattacks and the lack of explainability of most ML and deep learning methods leaves these systems vulnerable to intrusions, inaccuracies, compromised data, and new attacks. This research implements machine learning, deep learning, and explainable AI to assess the strengths and drawbacks of each method. Our results show that the inherently explainable model of decision trees (DT) provides excellent results similar to the highest performing models of LSTM and GRU. Achieving results within 2% of the best performing models and offering by far the lowest training and testing times of all models, Decision Trees demonstrates its viability as a computationally efficient, high-performing, inherently explainable model for IoT intrusion detection systems

• 

  Harder, Better, Faster, Stronger: Predicting Magnesium Alloy Hardness with Machine Learning

  Kia Hubbard and Wisdom Akande

  Material hardness is crucial in engineering and manufacturing, affecting component performance and durability. This study examines magnesium (Mg) alloys, valued for their lightweight and stiffness properties, with applications in automotive, aerospace, electronics, and medical industries. We conducted Vickers Hardness (VH) testing to analyze the effects of alloy composition, microstructure (phase composition and grain size), and rolling on hardness. The study focused on Mg-6Al (6 wt% aluminum) in both cast and rolled conditions. Results showed that rolling increased VH, demonstrating how structural modifications impact hardness. Additionally, machine learning (ML) was employed to predict the hardness of Mg-6Al alloys. Hardness, defined as resistance to localized permanent deformation, depends on grain size, precipitates, dislocation density, and phase composition. The Mg alloy was cast and rolled at NC A&T, with 257 VH data points collected. Using MATLAB’s Regression Learner, Gaussian Process Regression (GPR) showed the best predictive performance. ML integration aims to improve hardness prediction accuracy, reduce testing errors, and enhance efficiency and cost-effectiveness in material evaluation.

• 

  A Tool for Reverse Analysis and Classification of Executables (T.R.A.C.E) Family of Algorithms

  Kitty Huynh

  National security faces increasing risks due to the evolving capabilities of enemy drones. In reverse engineering, analyzing executable files remains a challenge, as minor code modifications can bypass traditional signature based detection methods. Dynamic analysis, while effective, is often time consuming. Many static analysis approaches require expert interpretation, limiting their accessibility. Existing static methods, such as variable name prediction and sequence-based techniques, suffer from low accuracy due to compiler variations. Reverse engineering tools also fail to provide meaningful variable names for analysis, further complicating the process. Graph Neural Networks (GNNs) offer a promising solution for executable code analysis without requiring extensive expertise or external domain knowledge. While GNNs have shown success in static analysis using control flow graphs (CFGs) and function call graphs, limited research has explored their application in data flow graphs (DFGs) for algorithm identification. Current research primarily focuses on detecting malicious behavior, but there is a gap in classifying executable files based on their algorithmic families. This study aims to develop a system capable of analyzing Windows executable files and predicting their algorithm classification with confidence levels ranging from 0% to 100%. By leveraging data flow graphs and GNNs, this approach seeks to enhance executable file analysis, improving accuracy and efficiency.

• 

  Analyzing Driver Distraction During Animal Crossings on Rural Roads: Insights from Eye-Tracking and Vehicle in-the-Loop Simulation

  Amir Milad Moshref Javadi and Sita Ram Parasad

  Rural roads continue to be among the riskiest and least understood driving conditions. Using a vehicle-in-the-loop driving simulator and Tobii Glasses 3 eye-tracking technology, this study examined driving performance and behavior on rural roads in three different visibility scenarios: high (during the day), medium and low(during the night). A comparison of driver attention and decision-making in manual and Advanced Driver Assistance System (ADAS) modes was made possible by the inclusion of actual risks in the simulations, such as roadside distractions and animal crossings considering both animal walking and running cross the road. Important findings showed that drivers’ ability to notice hazards was seriously lacking in areas including failing to keep safe distances and ignoring animal crossings. The unpredictable and low visibility of rural areas, where cognitive load and attention demands were higher, made it difficult for ADAS systems to prevent crashes, even though they were successful in controlled circumstances. These findings highlight the necessity of more sophisticated sensors, cameras, and radars to improve ADAS functionality in rural areas. This work offers important insights into how visual attention, cognitive functions, and driving performance interact. These insights can be used to develop adaptive ADAS systems, improve road safety, and solve the particular difficulties associated with driving in rural areas