Computer Sciences Commons^™

Spatialimaginer: Towards Adaptive Visual Imagination For Spatial Reasoning, Yian Li, Yang Jiao, Bin Zhu, Tianwen Qian, Shaoxiang Chen, Jingjing Chen, Yu-Gang Jiang

Research Collection School Of Computing and Information Systems

Spatial intelligence, which refers to the ability to reason about geometric and physical structure from visual observations, remains a core challenge for multimodal large language models. Despite promising performance, recent multimodal large language models (MLLMs) often exhibit fragile reasoning traces in spatial intelligence tasks that involve consistent spatial state recognition. We argue that these failures stem from a mismatch between the spatial recognition mechanism and the text-only reasoning behavior of these MLLMs. Effective spatial reasoning requires low-level geometric structure to be faithfully preserved and updated throughout the reasoning process, whereas textual representations tend to abstract away precisely these critical details. …

Assessor Experiences In Cmmc Level 2 Certification Assessments: An Interpretative Phenomenological Analysis Of Role Expectations, Samuel Heuchert, John Hastings

Research & Publications

The Cybersecurity Maturity Model Certification program requires that third-party assessments be conducted under a non-consultative model. The model is intended to ensure impartiality for organizations seeking certification. While this structure defines expectations for assessor behavior, assessor experiences and interpretations of these constraints remain underexamined. The study examines the lived experiences of CMMC-Certified Assessors and how they navigate role expectations within the non-consultative model. Using Role Conflict Theory as a guiding framework, the study applied Interpretative Phenomenological Analysis (IPA) to semi-structured interviews to explore how assessors make sense of their roles. The analysis identified experiential themes that describe how assessors construct …

Chemnetworks: New Capabilities For High-Throughput, Real-Time Chemical Graph Construction And Analysis, Daniel J. Pope, Jackson Elowitt, Bo Zhang, Manish Parashar, Aurora E. Clark

Michigan Tech Publications

A major revision of the ChemNetworks software (originally published in the Journal of Computational Chemistry, 2014, 35, 495–505) is presented. While the original ChemNetworks provided foundational graph construction capabilities for chemical systems, it was limited to simple distance and 3-body angular edge criteria, was not designed for high-performance computing environments or real-time operation alongside running simulations. This release addresses these limitations through three core contributions. First, a recursive Z-matrix-based search algorithm is introduced that enables chemically intuitive, arbitrarily descriptive three-dimensional structure searches, supporting geometric, energetic, and logical criteria. Second, the DataSpaces data staging framework is incorporated as an optional I/O …

Online Multidimensional Multiple Choice Knapsack Path Planner For Uavs, Manisha Wadhwa, Neelima Gupta, Sanjay Madria

Computer Science Faculty Research & Creative Works

The rapid proliferation of Unmanned Aerial Vehicles (UAVs) in safety–critical and time-sensitive applications such as disaster management, battlefield reconnaissance, urban surveillance and infrastructure inspection demands online path planning strategies. A fundamental requirement in such missions is generating flight paths that pass through some pre-specified waypoints while servicing dynamically requested spatio-temporal task points arriving randomly such as capturing aerial imagery. Existing UAV path planning approaches, including A* and its variants, meta-heuristic algorithms, and dynamic programming, find the shortest distance path from source to destination, but they are predominantly offline and assume complete prior environmental knowledge, thus, incur significantly higher computational cost …

Mutation-Based Multi-Agent Test Case Update, Dawei Tian, Jiakun Liu, Yun Peng, Yichen Zhang, Jianlei Chi, Jun Sun, Xiaohong Su

Research Collection School Of Computing and Information Systems

Modern software systems evolve rapidly under CI/CD practices, where tests are critical for quality. However, substantial code changes often render existing test cases obsolete, causing pipeline disruptions, reduced productivity, and compromised quality. Recent automatic test update approaches leverage LLMs to refine test cases via execution feedback and exact-matching context retrieval, prioritizing executability and line coverage but suffering three limitations: (1) neglecting test assertion adequacy, weakening fault detection; (2) relying on coarse line coverage instead of specific uncovered lines/branches; (3) using exact-matching retrieval, which fails for LLM hallucinated queries. To address these, we propose MuMuTestUp, a mutation-guided multi-agent framework with three …

Ddor: Delta Debugging For Explainable Overrefusal Testing And Repair, Qinyan Zhou, Peixin Zhang, Jun Sun, Haonan Zhang, Dongxia Wang

Research Collection School Of Computing and Information Systems

While safety alignment and guardrails help large language models (LLMs) avoid harmful outputs, they can also induce overrefusal, i.e., unwarranted rejection of benign queries that merely appear risky. We present DDOR (Delta Debugging for OverRefusal), a fully automated and explainable framework for overrefusal testing and repair in a black-box setting, where only model inputs and outputs are accessible and internal safety mechanisms remain opaque. DDOR applies delta debugging to localize minimal refusal-triggering fragments (mRTFs) that provide phrase-level, explainable evidence for why a refusal occurs. Conditioned on these mRTFs, DDOR generates diverse, context-rich prompts and performs multi-oracle validation to filter intrinsically …

Hands-On Ransomware: An Experiential Wannacry Case Study For Undergraduate Cybersecurity Education, Eli Creek Richmond, Thomas R. Devine

Military Cyber Affairs

Ransomware represents one of the most disruptive threats in the cyber landscape, yet hands-on malware analysis remains rare in undergraduate cybersecurity curricula. This paper presents the design, implementation, and evaluation of an experiential learning module centered on the WannaCry ransomware case study, deployed in a senior-level course at West Virginia University. Students performed static and dynamic analysis using industry-standard tools. Pre- and post-module assessments demonstrated measurable gains in self-reported competency across seven technical dimensions. The module's competencies align directly with DoD Cyber Workforce Framework Work Role 212, Cyber Defense Forensics Analyst, supporting education-to-workforce pipeline development.

From Framework To Toolchain: Implementing Zero Trust Architecture In Cloud-Native Environments For Dow Compliance, Shelby C. Snyder

Military Cyber Affairs

Federal agencies face a fiscal year 2027 target for enterprise-wide Zero Trust deployment, but NIST SP 800-207A defines logical components without identifying the Kubernetes technologies that implement them. This paper proposes a three-tier mapping of the Policy Engine, Policy Administrator, and Policy Enforcement Point to service mesh, microsegmentation, and perimeter tooling, stating the criteria by which each component is classified. It then applies a defined rubric to six Zero Trust vendors across component alignment, Kubernetes capability, federal authorization posture, and evidence quality, finding that no single vendor covers all three tiers. The mapping is a testable architectural proposition; a Stage …

Characterizing Advanced Persistent Threats With Cyber Attack Flow Metrics, Tyler Miller, Caleb Chang, Shouhuai Xu

Military Cyber Affairs

Cyber attack campaigns vary not only in scale but in structure, yet conventional characterizations often reduce them to a single dimension such as technique count or impact severity. In this paper we extend the concept of cyber attack flows by defining three new metrics, novelty, technique complexity and flow complexity. Then we characterize the attack flows of three advanced persistent threat campaigns using these metrics and draw insights regarding their capabilities. Our findings include that low novelty does not equate to low attack capabilities and that exploitation of an internet-facing appliance is a common initial attack vector.

Semantic Shields: Automating Critical Infrastructure Defense Via Nlp-Driven Ransomware Profiling, Henry Trowbridge, Ian Zalcberg, Ryan Schley, Carter Yagemann, Natasha Phan, Srikar Maduposu, Vimal Buck

Military Cyber Affairs

Ransomware poses a growing threat to critical infrastructure, where successful attacks can disrupt operational technology (OT) and industrial control systems (ICS) with significant public safety consequences. However, attributing ransomware incidents to specific threat actors remains challenging due to ransomware-as-a-service ecosystems, actor rebranding, and the obfuscation of traditional indicators of compromise. This paper presents Semantic Shields, an NLP-driven attribution framework that leverages BERT-generated semantic embeddings and DBSCAN clustering to profile ransomware actors through the linguistic characteristics of ransom notes. Using a dataset of 295 ransom notes from 189 distinct threat groups, the framework achieved an 87.2% true positive clustering rate and …

Closing The Interpretability Gap: Explainable Ml-Based Malware Detection For Defensive Cyberspace Operations, Tashi Stirewalt, Sean Hodgson, Puumaaya Tahiru, Assefaw Gebremedhin

Military Cyber Affairs

This paper presents an end-to-end, explainable malware triage pipeline designed for defense-oriented cyber operations. It combines high-performance static detection methods with analyst-centered interpretability. Utilizing the EMBER 2024 Windows PE subset, we train and evaluate four classifiers and select LightGBM as the production model based on its predictive performance, inference efficiency, and compatibility with exact tree-based attribution. The deployed system consists of four sequential components: PE feature extraction, malware probability scoring, dual explainability (using SHAP and LIME), and large language model (LLM) report generation, all integrated within a Flask web interface. On a temporal test set of 1,080,000 samples, LightGBM achieves …

A Calibrated And Conformal Deep Learning Framework For Trustworthy Antinuclear Antibody Pattern Recognition With Selective Referral To Experts, Hussein Ali Hussein Al Naffakh, Ahmed Dheyaa Radhi, Raghdah Maytham Hameed, Muntaha Abdullah Reishaan, Fouad A. Majeed, Rozaida Ghazali

Karbala International Journal of Modern Science

Reading antinuclear antibody patterns on human epithelial cells by indirect immunofluorescence is the reference screen for systemic autoimmune rheumatic diseases, but it is slow, subjective, and variable between observers. Deep learning reaches high accuracy on this task, yet most systems return a single prediction without stating how reliable it is, which is unsafe in a diagnostic workflow. This paper presents an intelligent decision support framework built around a single calibrated uncertainty signal. That signal is the control variable for four reliability modules: confidence calibration, conformal prediction, error detection, and selective referral. A feature space out of distribution detector serves as …

Resilient Ethical Governance By Design: Building Resilience For Future-Ready Smart Cities Against Emerging Technology Disruptions In Public Sector – A Case Study Of Dallas Smart City, Emmanuel Asamoah Asare

Doctoral Dissertations and Projects

This dissertation examines how Dallas’s smart city strategic policies foster resilience against disruptions associated with emerging technologies, particularly artificial intelligence and Internet of Things systems. Using a convergent mixed-methods comparative case study, Dallas serves as the primary case, Austin serves as a U.S. municipal comparator, and Barcelona serves as a qualitative benchmark for ethical and trust-centered smart city governance. The study integrates document analysis, broadband and digital equity indicators, stakeholder engagement evidence, and exploratory quantitative comparisons to assess how governance adaptability, preparedness, functionality, equity, and stakeholder trust shape resilience outcomes.

The findings provide partial support for the study’s hypotheses. Dallas …

Cuip-X25: A Real-World Network Intrusion Dataset For Next-Generation Ai-Driven Security, Arshad Iqbal, Sohail Asghar

Turkish Journal of Electrical Engineering and Computer Sciences

The efficacy of artificial intelligence (AI) in intrusion detection systems (IDS) is critically dependent on high-fidelity training data. However, as detailed in the manuscript's literature review, existing benchmark datasets are predominantly synthetic, outdated, or imbalanced and fail to capture the complexity of the contemporary threat landscape. To bridge this gap, this study introduces CUIP-X25, a novel real-world cyber-attack dataset captured over a four-month period using a dionaea honeypot deployed on a public network. Unlike synthetic alternatives, this dataset provides an authentic representation of modern adversarial tactics, techniques, and procedures, encompassing 3.16 million real events across ten distinct attack categories, including …

Boundary Layer Sliding Mode Control Strategy For Variable-Speed Compressor In Deep Freezers: Experimental Validation Of Energy Efficiency And Freezing Capacity, Sertan Aksoy, Kami̇l Çeti̇n, Sezai̇ Taşkin

Turkish Journal of Electrical Engineering and Computer Sciences

This study presents the design and experimental validation of a nonlinear sliding mode controller developed for a deep freezer equipped with a variable-speed compressor. The proposed control strategy aims to minimize energy consumption while maintaining rapid and stable cooling performance under varying ambient conditions. A detailed thermal model of the deep freezer was established using an equivalent resistance–capacitance network representation, enabling precise analysis of temperature dynamics. The sliding mode-based control algorithm dynamically adjusts the compressor’s operating frequency according to temperature deviation, ambient conditions, and time-dependent factors, providing robust performance without requiring parameter retuning for different models. Beyond theoretical-based analysis, a …

Challenges, Trends, And The Role Of Lstm In Ai-Based Predictive Maintenance Of Electrolyzers For Solar Hydrogen Systems: A Review, Yani Koerniawan Kuatno, Muhamad Zahim Sujod

Turkish Journal of Electrical Engineering and Computer Sciences

The transition toward low-carbon energy systems has increased interest in hydrogen as a clean energy carrier, with solar-driven water electrolysis emerging as a promising technology due to its high efficiency and compatibility with renewable energy sources. However, dynamic operating conditions and intermittent renewable input accelerate electrolyzer degradation, reducing reliability and system lifespan. Predictive maintenance (PdM), supported by artificial intelligence (AI), offers a data-driven approach to anticipate failures and improve operational durability. This review systematically investigates AI-based PdM approaches for electrolyzers, with an emphasis on long short-term memory (LSTM) networks and Internet of things (IoT) integration. Following PRISMA 2020 guidelines, 35 …

Empowering Edge Intelligence Through Reparameterized Lightweight Transformers And Distributed Inference, Hosein Esmaeili, Mohammad Ali Afshar Kazemi, Reza Radfar, Nazanin Pilevari

Turkish Journal of Electrical Engineering and Computer Sciences

Deploying advanced transformer-based models on resource-constrained edge devices remains a significant challenge due to their high memory footprint and substantial compute requirements. In this paper, we propose a reparameterized transformer framework that integrates High-Rank Factorization (HRF) during training, layer merging at inference, and dynamic, load-balanced distributed inference across multiple devices. To further reduce resource usage, our framework supports mixed-precision quantization down to 4-bit, enabling flexible accuracy–latency–energy trade-offs. Experimental evaluations on the ESC-50 environmental sound dataset demonstrate that our method matches or exceeds the performance of larger baseline models while using 20–30% fewer parameters, achieving up to 48% latency reduction in …

Dynamic Focal Loss Adjustment For Railway Defect Detection, Mehmet Koç, Ridvan Özdemi̇r, Ömer Gerek

Turkish Journal of Electrical Engineering and Computer Sciences

Railway infrastructure is critical to the safe and efficient operation of transportation systems, and the early detection of defects is essential for preventing catastrophic failures. Automated defect detection methods are therefore crucial for maintaining continuous safety while reducing maintenance costs. Although Focal Loss is widely used in object detection under class-imbalanced conditions, its fixed α parameter may limit its effectiveness in detecting rare defects. In this study, we propose an adaptive α-tuned Focal Loss approach that dynamically adjusts class weights based on average precision (AP) values. By iteratively optimizing α without relying on gradient-based optimization, the proposed method improves the …

Relative Rate Observer-Based Online Tuning Mechanism For Single-Input Interval Type-2 Fuzzy Pid Controllers, Oqba Aldreiei, Cenk Ulu, Mert Can Kurucu, Müjde Güzelkaya

Turkish Journal of Electrical Engineering and Computer Sciences

The characteristics of the footprint of uncertainty (FOU) in interval type-2 membership functions (IT2-MFs) are crucial to the performance and robustness of interval type-2 fuzzy controllers (IT2-FCs). However, existing IT2-FC design approaches mostly use fixed FOU structures. This study proposes an online membership function (MF) adjustment mechanism for a single-input interval type-2 fuzzy PID controller (SIT2-FPID)  that adjusts the FOU of the antecedent MFs and weights of the consequent MFs, respectively, to achieve high performance and robustness. The proposed online adjustment mechanism consists of a relative rate observer (RRO), a two-input rule-base adjustment system, and a first-order smoothing filter. The …

Robust Load Frequency Control For Multiarea Electrical Power Systems Via Analytical Proportional-Integral-Derivative Plus Second Order Derivative Controller Design, Yavuz Güler, Mustafa Nalbantoğlu, Ibrahim Kaya

Turkish Journal of Electrical Engineering and Computer Sciences

This research presents a proportional-integral-derivative plus second order derivative (PIDD2) controller design based on the Direct Synthesis Method (DSM) for load frequency control (LFC) of interconnected power systems. The parameters of the proposed PIDD2 controller are determined using the DSM, which offers an analytical approach for tuning. The design approaches have been developed specifically for single, two, and three-area power systems, encompassing nonreheated and reheated thermal turbines. In the proposed design method, the best values of PIDD2 controller parameters were found by using a multicriteria objective function that includes the integral of absolute error (IAE) and settling time. In response …

A Binary Multiobjective Hippopotamus Optimization Algorithm For Feature Selection In Phishing Website Detection, Fatima Belmessaoud, Sofiane Maza, Djaafar Zouache

Turkish Journal of Electrical Engineering and Computer Sciences

Phishing website detection remains a major challenge in cybersecurity as attackers continuously develop new techniques to deceive users. Identifying the most informative features from large datasets is essential to improve classification accuracy while reducing computational complexity. Feature selection is therefore widely addressed using metaheuristic optimization techniques due to their flexibility and global search capability. In this study, we propose a Binary Multiobjective Hippopotamus Optimization Algorithm (B-MOHOA) for feature selection in phishing website detection. The proposed method simultaneously optimizes two conflicting objectives: maximizing classification accuracy and minimizing the number of selected features. Unlike many existing studies that mainly focus on transfer …

Range–Angle-Dependent Oam Beamforming With A Concentric Helical Circular Fda, Uğur Yeşi̇lyurt

Turkish Journal of Electrical Engineering and Computer Sciences

Secure and spatially selective wireless transmission requires orbital angular momentum (OAM) beams that are confined to a specific range and angle, rather than propagating indefinitely along the beam axis. In this paper, a concentric helical circular frequency diverse array (CHCFDA) is proposed to generate range–angle-dependent OAM beams without requiring external phase shifters. The helical element positioning inherently provides the necessary interelement phase distribution through physical step height, while logarithmically increasing frequency offsets are applied across concentric rings—and optionally across individual elements—to eliminate range periodicity and achieve a single, well-focused OAM beam exclusively at the target location. Both linear and logarithmic …

Measurement-Aware Zero-Phase Iterative Learning Control For Robust Regulation Of Nonideal Boost Converters In Electric Vehicle Fast Charging, Aytaç Altan, Mohammed S. Alzaidi, Cağfer Yanarateş

Turkish Journal of Electrical Engineering and Computer Sciences

Integrating battery energy storage with DC-DC boost converters for electric vehicle fast charging exposes the regulator to ageing-induced parameter drift, periodic load pulses, and, critically, the nonidealities in the output-voltage sensing chain. This paper proposes a measurement-aware, zero-phase iterative learning control scheme for robust output-voltage regulation of a nonideal boost converter whose parameters are matched to those of a commercial Texas Instruments TPS6102x battery regulator. The controller combines an inner proportional-integral stabilizing loop with an outer zero-phase learning law that updates a feedforward correction based on the sensor-captured output trajectory; a forward-backward robustness filter suppresses the amplification of measurement noise …

Pahdf: A Privacy-Aware Hybrid Detection Framework With Class-Aware Weighted Stacking Ensemble (Cawse) For Fake Instagram Account Detection, Sura Jasim Mohammed, Safa Saad Abbas, Suhad Hatem Jihad

Journal of Intelligent Informatics, Networking, and Cybersecurity

The rapid growth of social media platforms has intensified concerns regarding online privacy, data security, and fraudulent activities that driven by fake accounts. This paper proposes a Privacy-Aware Hybrid Detection Framework (PAHDF) to detect Instagram fake account that integrates privacy preservation with high-performance machine learning. Unlike existing approaches that treat privacy and detection as separated objectives, therefore, the proposed framework jointly addresses both objectives by relying exclusively on publicly available, low-sensitivity profile metadata. PAHDF combines a deep learning model for latent feature representation with a Random Forest classifier for behavioural pattern learning through a Class-Aware Weighted Stacking Ensemble (CAWSE), where …

From Detection To Segmentation: Adapting Yolo26 For Surface Crack Delineation, Mohammed Al-Mustafa, Israa H. Ali

Journal of Intelligent Informatics, Networking, and Cybersecurity

Surface crack delineation plays an important role in infrastructure inspection because accurate pixel-level mapping of cracks can support condition assessment, maintenance planning, and automated monitoring of roads and other civil surfaces. However, recent crack segmentation methods often depend on heavy network designs or prompt-based foundation models. It also remains unclear how a lightweight detection-oriented model behaves across different data organizations and unseen images. This study therefore investigates whether YOLO26m-seg, the medium segmentation variant of the recent YOLO26 family, can be adapted into an effective and practical pipeline for surface crack delineation. The proposed workflow has four main steps. First, binary …

Urban Spatial Development Control In Tanzania: Analysis Of Factors Influencing Gis Application Using Structural Equation Modelling (Sem)., Happiness Protas Mmanda, Nestory Yamungu

Tanzania Journal of Engineering and Technology (TJET)

Rapid urbanization in developing countries has intensified urban expansion, creating challenges for sustainable development. Geographic Information Systems (GIS) enhance spatial planning, but empirical evidence on factors influencing their effectiveness remains limited. This study examines determinants of GIS application in Urban Spatial Development Control (USDC). The objectives are to (1) identify and categorize factors affecting GIS use, (2) assess relative strength, and (3) develop a validated structural model explaining GIS adoption in USDC. Data were collected from 103 LGAs by a mixed sampling method. Exploratory and Confirmatory Factor Analysis classified influencing factors into technology-related (α = 0.869, CR = 0.881), process-related …

Decision Support Tool For The Maintenance Of Meter Gauge Railway Permanent-Way Infrastructures: A Concept Paper, Hamisi J. Maulid, Beatus A.T Kundi, Juma M. Matindana, Ismail W. R. Taifa Dr

Tanzania Journal of Engineering and Technology (TJET)

A Decision Support Tool (DST) represents a transformative way of modernising maintenance practice on the permanent-way infrastructure of Meter Gauge Railway (MGR). Maintenance practice remains largely reactive, as the MGR plays a strategic role in transporting both freight and passengers, resulting in inefficient resource allocation, high operational risk, and growing Lifecycle expenses. This concept paper explores the potential and the design challenges for a DST that combines fuzzy analytic hierarchy process (Fuzzy-AHP) modelling, multi-criteria decision analysis (MCDA), Geographic Information Systems (GIS) and predictive machine learning (ML) analytics for evidence-based, future-oriented maintenance management. The paper draws on peer-reviewed studies from …

Kanmultisign: Multi-Scale Sequence-Based Pose Animation From Sign Language Notation With Kolmogorov-Arnold Networks, Guanyi Du, Lintao Wang, Kun Hu, Ziyang Wang

Research outputs 2022 to 2026

Sign language production from symbolic notation offers a scalable route to accessible sign animation. We present KANMultiSign, a multi-scale sequence generator that translates HamNoSys notation into two-dimensional human pose sequences. Our framework makes two complementary contributions. First, we introduce a coarse-to-fine generation strategy with multi-scale supervision: the model is first guided by an intermediate body–hand–face scaffold to encourage global structural coherence, and then refines fine-grained hand articulation to improve finger-level detail. Second, we investigate integrating Kolmogorov–Arnold Network modules into a Transformer backbone, using learnable univariate function primitives to model the highly non-linear mapping from discrete phonological symbols to continuous body …

When Does Global Search Pay For Itself? A Measured Exploration Cost And Energy Break-Even Analysis Of A Metaheuristic Search Controller In Solar Energy Systems, Mohamed Ali Muammar Ezgour

Communications of the IIMA

Autonomous energy systems increasingly delegate the choice of operating point to embedded search algorithms, trading a fast local optimizer that can settle on a wrong point against a slower global search that guarantees the right one at a measurable cost. This paper reframes maximum power point tracking under partial shading as that decision and measures its economics on a fixed photovoltaic plant in MATLAB/Simulink. A Hippopotamus Optimization global search handed over to incremental conductance is compared with incremental conductance alone across seventeen initial duty cycles and thirty random seeds. The hybrid reached the global peak in all thirty seeds, whereas …

Cyberspace Collaborative Awareness: A Model For Unity Of Effort In Homeland Defense, Mike Knapp, Sean Atkins, Matthew Mclaughlin

Joint Force Quarterly

The increasing frequency and severity of cyberattacks against U.S. critical infrastructure continue to confound homeland defense efforts. Defending against state cyber campaigns that threaten the nation’s most critical systems requires a new awareness model that can enable unity of effort across public and private actors. Examining homeland defense awareness in other domains reveals principles and approaches that can inform the development of a collaborative awareness model in cyberspace. This new framework acknowledges the interconnectedness of government and commercial networks and the independent goals of each player in the domain. Doing so provides a viable path to achieving shared domain awareness …