Open Access. Powered by Scholars. Published by Universities.®

Digital Commons Network™

Open Access. Powered by Scholars. Published by Universities.®

University of South Florida

Discipline
Keyword
Publication Year
Publication
Publication Type
File Type

Articles 1 - 30 of 17188

Full-Text Articles in Entire DC Network

Visitor Perceptions Of Lampenflora And Implications For Adaptive Show Cave Management: A Case Study From Ryusen-Do Cave, Japan, Kanato Ando, Kazuhide Nakajima, Shuji Watanabe, Masuo Komukai, Hidesato Kuwahara Oct 2026

Visitor Perceptions Of Lampenflora And Implications For Adaptive Show Cave Management: A Case Study From Ryusen-Do Cave, Japan, Kanato Ando, Kazuhide Nakajima, Shuji Watanabe, Masuo Komukai, Hidesato Kuwahara

International Journal of Speleology

Artificial lighting in show caves promotes lampenflora growth, altering cave ecosystems and posing persistent management challenges. Although various control methods have been proposed, their effectiveness and social acceptance remain context dependent. This study investigated visitors’ perceptions of lampenflora through questionnaire surveys and semi-structured interviews to explore adaptive management strategies. Results indicated that visitors generally held positive impressions of lampenflora, but these impressions became more critical after receiving information about its artificial origin and ecological impacts. At the same time, support for conservation-oriented management, including willingness to contribute financially, tended to increase following information provision. Tourism-related evaluations were more strongly associated …

Go to article

Attention-Enhanced Multimodal Fusion Of Elevation And Aerial Imagery For Sinkhole Segmentation, Junfeng Zhu, Desireé Cunningham, Nathan Jacobs Oct 2026

Attention-Enhanced Multimodal Fusion Of Elevation And Aerial Imagery For Sinkhole Segmentation, Junfeng Zhu, Desireé Cunningham, Nathan Jacobs

Sinkhole Conference 2026

Image segmentation, the task of assigning a class label to each pixel in an image, has been widely applied in remote sensing to identify surficial geological and hydrological features from data sources such as multispectral satellite imagery, aerial photography, and digital elevation models. UNet is among the most widely adopted convolutional neural network architectures for segmentation, owing to its encoder–decoder structure and skip connections that enable precise spatial localization. UNet has been applied to detect sinkholes using lidar-derived high-resolution digital elevation data; however, existing models still fall short of the accuracy of manual sinkhole mapping. In this study, we aim …

Go to article

Comparing Manual And Automated Methods In Sinkhole Delineation: A Cantabrian Mountains (Spain) And Tasman Mountains (New Zealand) Case Study, Mario Puente-Sierra, Javier Santos-González, Rosa B. González-Gutiérrez Oct 2026

Comparing Manual And Automated Methods In Sinkhole Delineation: A Cantabrian Mountains (Spain) And Tasman Mountains (New Zealand) Case Study, Mario Puente-Sierra, Javier Santos-González, Rosa B. González-Gutiérrez

Sinkhole Conference 2026

Studies of karst landforms, such as sinkholes, have advanced in recent years due to technological advances and the rise of automated detection methods. Such methods offer notable time-saving strengths, but their limitations in complex terrains are frequently overlooked in modern literature. To compare these automated methods to traditional manual methods, we selected four karst massifs in the Cantabrian Mountains (Spain) and two in the Tasman Mountains (New Zealand) as study areas. High-resolution LiDAR-derived Digital Elevation Models were used to identify and map sinkholes. Automated detection was performed using a Geographic Information System and included: i) a depression-filling method; ii) a …

Go to article

Statewide Sinkhole Susceptibility Mapping In Florida Using Explainable Geoai, Osama M. Altarawneh, Weiwei Zhan, Clint Kromhout Oct 2026

Statewide Sinkhole Susceptibility Mapping In Florida Using Explainable Geoai, Osama M. Altarawneh, Weiwei Zhan, Clint Kromhout

Sinkhole Conference 2026

Sinkholes cause persistent damage to infrastructure, property, and groundwater resources across Florida. Statewide sinkhole hazard mitigation has been constrained by incomplete inventories and the absence of a high-resolution susceptibility map. This study presents a statewide sinkhole susceptibility assessment using an explainable GeoAI framework that integrates the citizen-reported Subsidence Incident Reports (SIR) database and a geologist-confirmed collapse-type sinkhole inventory from the Florida Geological Survey (FGS). Random Forest (RF) is used as the primary learner with ten geologic, hydrologic, topographic, and anthropogenic predictors. The influence of negative-sample buffer geometry on model behavior is evaluated across five annular buffer conditions (100-1000 m through …

Go to article

Mapping Sinkholes And Closed Depressions In The Edwards Aquifer Recharge Zone: Comparison Of Lidar-Based Semi-Automated Detection Methods And Field Validation, Enzo Franceschi Genesi, Horia G. Olariu, Bradford Wilcox Oct 2026

Mapping Sinkholes And Closed Depressions In The Edwards Aquifer Recharge Zone: Comparison Of Lidar-Based Semi-Automated Detection Methods And Field Validation, Enzo Franceschi Genesi, Horia G. Olariu, Bradford Wilcox

Sinkhole Conference 2026

Sinkholes and closed depressions play a critical role in karst hydrology by facilitating rapid recharge to aquifer systems (Ford & Williams, 2007; White, 1988), yet their spatial distribution is often poorly quantified in large recharge zones, and the automated methods used to detect them vary widely in approach and reliability. This study compares three LiDAR-based automated detection methods to determine their relative performance for mapping sinkholes and closed depressions within the Edwards Aquifer Recharge Zone, using a 0.5-meter resolution bare-earth Digital Elevation Model derived from high-density airborne LiDAR point cloud data (12.66 points/m²) obtained from OpenTopography. The three methods evaluated …

Go to article

Springs Of The Catskill Mountains: A Conduit Flow System In Predominantly Siliciclastic Strata, Daniel H. Doctor, Joshua R. Benton, Martin A. Briggs, William E. Odom, Neil C. Terry Oct 2026

Springs Of The Catskill Mountains: A Conduit Flow System In Predominantly Siliciclastic Strata, Daniel H. Doctor, Joshua R. Benton, Martin A. Briggs, William E. Odom, Neil C. Terry

Sinkhole Conference 2026

The Catskill Mountains of New York state are composed of thick sequences of interbedded mudstone, sandstone, and siltstone that are intercalated with thin beds of paludal limestone and calcareous mudstone. Sandstones often contain rip-up clasts of calcareous material at their bases and contain minor amounts of carbonate cement throughout their mass. This minor amount of calcareous material weathers out to produce thin zones of solutional porosity and permeability integrated within perched aquifers that feed perennial hillslope springs. Solutional features are evident within thin carbonate lenses and along fractures in sandstones, occasionally resulting in explorable cave-sized conduits. This study focuses on …

Go to article

Karst At A Potential Construction Site Adjacent To The Delaware River Near The Glacial Terminus, New Jersey, Usa, Ira D. Sasowsky Oct 2026

Karst At A Potential Construction Site Adjacent To The Delaware River Near The Glacial Terminus, New Jersey, Usa, Ira D. Sasowsky

Sinkhole Conference 2026

The state of New Jersey is not known for extensive karst landscapes, but there are over 100 caves listed in a survey of those features. Numerous sinkholes have also been identified, though no comprehensive database yet exists. Occasional news stories have reported damaging collapses, many of which seem infrastructure related, in several counties. This paper reports on the conditions at a proposed major construction site in a glaciated limestone terrane, Warren County. The area is underlain by the Allentown Dolomite and the Beekmantown Group carbonates (Cambrian/Ordovician age), and covered with a variable layer of glaciofluvial sediments. The area is of …

Go to article

Karst Sinkholes And Civil Engineering: Overview Of Characterization And Treatment Options From A Geophysical And Geotechnical Viewpoint, Mia Painter, Stephen Welsh Oct 2026

Karst Sinkholes And Civil Engineering: Overview Of Characterization And Treatment Options From A Geophysical And Geotechnical Viewpoint, Mia Painter, Stephen Welsh

Sinkhole Conference 2026

Sinkholes pose a significant geohazard to civil engineering infrastructure in regions underlain by carbonate karst terrain. Selecting an appropriate stabilization method requires balancing site conditions, infrastructure risk, environmental considerations, and cost, all of which depend on a thorough understanding of the planned or existing infrastructure and subsurface conditions. This paper provides an overview of common sinkhole characterization and stabilization methods used in the Mid-Atlantic and Appalachian regions of the United States from both geotechnical and geophysical perspectives. The application of geophysical methods in conjunction with intrusive investigations is reviewed as a means of reducing subsurface uncertainty and refining treatment designs. …

Go to article

Sinkhole Hazards, Risks, And Remediation: A Fifty-Year Case Study In New Jersey, Usa, Oladoyin Kolawole, Rita L. Sousa, Christina Comuso, Mohab Hussein Oct 2026

Sinkhole Hazards, Risks, And Remediation: A Fifty-Year Case Study In New Jersey, Usa, Oladoyin Kolawole, Rita L. Sousa, Christina Comuso, Mohab Hussein

Sinkhole Conference 2026

New Jersey presents a uniquely complex sinkhole and ground subsidence hazard shaped by intersecting geological, industrial, and anthropogenic forces. Sinkhole herein refers primarily to collapse sinkholes, that is, the abrupt or progressive downward failure of the ground surface into subsurface voids, as distinct from solutional closed depressions of purely karst origin. This paper synthesizes five decades (1976-2026) of documented sinkhole and ground subsidence records across the state, examining root causes, societal and environmental impacts, and remediation strategies employed in response to such hazards. The review shows that New Jersey’s hazard is governed by three dominant mechanisms: karst dissolution in carbonate …

Go to article

Production Variability In The Edwards Aquifer, South-Central Texas, Usa, Jennifer Raye (Adkins) Schudrowitz, Linda K. Bevis Oct 2026

Production Variability In The Edwards Aquifer, South-Central Texas, Usa, Jennifer Raye (Adkins) Schudrowitz, Linda K. Bevis

Sinkhole Conference 2026

The San Antonio Water System (SAWS) serves more than two million people and is among the largest water utilities in the United States. Much of SAWS’s supply is sourced from the Edwards (Balcones Fault Zone) Aquifer (Edwards BFZA), a major sole‑source karst aquifer in south‑central Texas. The aquifer exhibits pronounced spatial heterogeneity in well performance due to its karstic nature, structurally complex setting, and variable depositional facies. Recent severe drought and pumping restrictions have emphasized the need to understand where within the aquifer to drill and how inter‑station performance differs under stress. This study evaluates production variability across SAWS pump‑station …

Go to article

Ongoing Dye Tracing Prompts Major Reevaluation Of Groundwater Divides For The Lost River Karst System, Bowling Green, Ky, Usa, Lee Anne Bledsoe, Chris G. Groves, Nick Lawhon Oct 2026

Ongoing Dye Tracing Prompts Major Reevaluation Of Groundwater Divides For The Lost River Karst System, Bowling Green, Ky, Usa, Lee Anne Bledsoe, Chris G. Groves, Nick Lawhon

Sinkhole Conference 2026

The Lost River Groundwater Basin located in south-central Kentucky may be one of the most thoroughly investigated karst basins in the US. Over the past 50 years, numerous studies have been conducted primarily due to the environmental challenges created by urban development on a well-developed sinkhole plain. The City of Bowling Green (pop. ~80,000) covers much of the basin but lies primarily in the northern, downstream reach nearest the main outlet of the system, the Lost River Rise. However, land use in the southern, up-gradient, portion of the basin is quickly changing from undeveloped and/or agriculture to industrial and residential. …

Go to article

A Field-Inspired Investigation Of Conduit Network Impacts On Spring Flow In Karst Aquifers, Jennifer Huang, Jannes Kordilla, Marco Dentz, John Barry, Christian Langevin, Peter Kang Oct 2026

A Field-Inspired Investigation Of Conduit Network Impacts On Spring Flow In Karst Aquifers, Jennifer Huang, Jannes Kordilla, Marco Dentz, John Barry, Christian Langevin, Peter Kang

Sinkhole Conference 2026

Effective management of karst aquifers necessitates accurate characterization of flow and transport. However, modeling these highly heterogeneous systems is challenging due to the complex partitioning of recharge between bedrock and the embedded conduit network. The properties of conduit networks impact flow and transport, but there has not been a systematic, quantitative study of how conduit network structure and geometry impact spring hydrographs in naturalistic field conditions. In this field-inspired modeling study, we attempt to identify key geometric and structural properties of conduit networks that control flow in karst aquifers and use these insights to constrain our understanding of a nitrate …

Go to article

Characterization Of The Galena Karst Of East-Central Olmsted County, Minnesota, John D. Barry, Ronald C. Spong, Peter J. Hutchinson, Martin Larsen, Anthony C. Runkel Oct 2026

Characterization Of The Galena Karst Of East-Central Olmsted County, Minnesota, John D. Barry, Ronald C. Spong, Peter J. Hutchinson, Martin Larsen, Anthony C. Runkel

Sinkhole Conference 2026

The Galena Group is mostly composed of carbonate rock with subordinate beds of fine siliciclastics, mainly shale. In southeastern Minnesota, it is a source of water for domestic and agricultural use. Hydrologic and geochemical characterization of a Galena Group spring, Bear Spring, began in 1979. Recent springshed delineation through dye tracing and continuous level, temperature, and nitrate monitoring has shed light on the aquifer’s hydrologic response to precipitation, but the degree of karstification in the Galena was unclear. Current and planned future modeling require additional insight into the pervasiveness of fractures and conduits, and their properties. A suite of characterization …

Go to article

Investigating Sediment Accumulation Dynamics Via Integration Of Gis And Remote Sensing Tools Along With Physical/Chemical Characteristics Of Sediments In Owl Cave, Highland County, Virginia, Dhanuska B. Wijesinghe, Spencer Schmidt Oct 2026

Investigating Sediment Accumulation Dynamics Via Integration Of Gis And Remote Sensing Tools Along With Physical/Chemical Characteristics Of Sediments In Owl Cave, Highland County, Virginia, Dhanuska B. Wijesinghe, Spencer Schmidt

Sinkhole Conference 2026

The Valley region of Virginia, USA, features cave and karst systems. Variation in surface topography and vegetation within these systems significantly affects sediment accumulation in cave ecosystems. Remote sensing and GIS tools effectively analyze surface conditions, including vegetation cover and its influence on sediment buildup in caves. This study involved collecting high-resolution drone imagery, such as LiDAR, multispectral, and thermal data near Owl Cave in Highland County, VA, to understand the sediment accumulation process inside the cave system. High-resolution digital elevation models (DEMs) were created from LiDAR point clouds, and vegetation indices like Normalized Difference Vegetation Index (NDVI) were calculated …

Go to article

A Comprehensive Look At Ohio’S Karst Inventory Mapping—Reconnaissance, Ground-Truth, And Beyond, Brittany Parrick, Douglas Aden Oct 2026

A Comprehensive Look At Ohio’S Karst Inventory Mapping—Reconnaissance, Ground-Truth, And Beyond, Brittany Parrick, Douglas Aden

Sinkhole Conference 2026

Keywords: Ohio, mapping, karst inventory, karst database, fieldwork, LiDAR, karst, GIS, geomorphology, digital elevation models The present iteration of the karst inventory mapping program at the Ohio Department of Natural Resources, Division of Geological Survey (herein the Survey) has been ongoing since 2009. With documentation of the Survey’s previous karst inventory mapping scattered across several decades-old publications—coupled with the karst community’s demonstrated need for a standardized database—this paper outlines the program’s evolution from its inception to the present. New Digital Elevation Models (DEM) derived from modern Light Detection and Ranging (LiDAR) datasets, along with modern mobile applications, have significantly improved …

Go to article

Cover Collapse Sinkholes In Stormwater Infrastructure In The Urban Karst Of Bowling Green, Kentucky: Occurrence, Mitigation, And Prevention, Nick Lawhon P.G., C.F.M. Oct 2026

Cover Collapse Sinkholes In Stormwater Infrastructure In The Urban Karst Of Bowling Green, Kentucky: Occurrence, Mitigation, And Prevention, Nick Lawhon P.G., C.F.M.

Sinkhole Conference 2026

Cover collapse sinkholes are an extremely common geohazard in karst terrains. These features, formed by concentrated flows of water entering the subsurface through discrete points in the epikarst, are particularly common in stormwater infrastructure in Bowling Green, KY. There are currently 787 known stormwater basins and hundreds of miles of stormwater conveyances in this fast-developing city, all of which concentrate stormwater flows over the naturally sinkhole-prone karst system. This paper reports on the occurrence of cover collapse sinkholes in stormwater infrastructure, particularly retention and detention basins, from inspections over the past 7 years and records from City-managed sinkhole repair projects, …

Go to article

Biogeomechanics Of Eicp Reinforcement Of Karstified Rock Masses, Mary C. Ngoma, Oladoyin Kolawole Oct 2026

Biogeomechanics Of Eicp Reinforcement Of Karstified Rock Masses, Mary C. Ngoma, Oladoyin Kolawole

Sinkhole Conference 2026

Sinkholes are among the most significant karst-related hazards, arising from the dissolution of soluble rock and subsequent collapse of overlying geomaterials, thereby compromising rock mass stability and underground infrastructure. Conventional sinkhole remediation approaches often rely on cement-based grouting to prevent failure; however, the associated environmental impacts underscore the need for eco-friendly and sustainable alternatives such as biocementation. While biocementation has been a widely applied technique in reinforcing soils and cemented structures, the potential of biocemented hybrid soil-embedded karstified rock systems remains unexplored. To address sinkhole-related post-hazard remediation, this work investigates the potential of biocementation as a sustainable reinforcement strategy for …

Go to article

A Comprehensive Analysis Of Stratigraphic Influence On Karst Feature Distribution And Density Across Southeast Minnesota, Usa, Julia R. Steenberg, John D. Barry, Andrew J. Retzler, Anthony C. Runkel, Robert G. Tipping, E. Calvin Alexander Jr. Oct 2026

A Comprehensive Analysis Of Stratigraphic Influence On Karst Feature Distribution And Density Across Southeast Minnesota, Usa, Julia R. Steenberg, John D. Barry, Andrew J. Retzler, Anthony C. Runkel, Robert G. Tipping, E. Calvin Alexander Jr.

Sinkhole Conference 2026

Karst features pose significant risks to infrastructure and land use, from reducing tillable acreage to causing catastrophic failures in wastewater lagoons. This study analyzes the stratigraphic and spatial distribution of karst features in southeastern Minnesota to better understand their geologic controls. By seamlessly merging raster surfaces of Paleozoic bedrock units and intersecting them with regional karst datasets in a GIS framework, we found that over 95 percent of mapped sinkholes occur where unconsolidated sediment overlying bedrock is 15 meters or less. Furthermore, 81 percent of sinkholes are concentrated in just six of the 21 mappable bedrock units, each of which …

Go to article

Hasparos Embayment – An Example Of Intrastratal Hypogenic Speleogenesis, Peter Hutchinson Oct 2026

Hasparos Embayment – An Example Of Intrastratal Hypogenic Speleogenesis, Peter Hutchinson

Sinkhole Conference 2026

The Hasparos Embayment contains Neogene-aged deposits that overlie the Permian-aged San Andres Formation. The Hasparos Embayment consists of thousands of subsidence sinkholes, concentrated in two tracts that are overlain with deposits provisionally considered to be part of the Ogallala Formation. The two Ogallala-covered karst fields are divided north and south by the Hasparos Canyon. The northern karst field includes 12,400 hectares, whereas the southern tract includes 6,900 hectares. Intrastratal karst is a form of hypogene speleogenesis, in which karst develops in rocks buried by younger deposits after deposition of the younger rocks. Thus during the temperate, late Pleistocene epoch, Lake …

Go to article

Cognitive Resilience At The Edge: Hyperdimensional Computing Versus Deep Learning For Hardware-Degraded Rf Classification, Adrian B. Cisneros, Jeong Yang Sep 2026

Cognitive Resilience At The Edge: Hyperdimensional Computing Versus Deep Learning For Hardware-Degraded Rf Classification, Adrian B. Cisneros, Jeong Yang

Military Cyber Affairs

Autonomous Collaborative Combat Aircraft (CCA) operating in contested electromagnetic environments must classify Radio Frequency (RF) signals on edge silicon that degrades over the mission lifetime due to thermal stress, radiation, and manufacturing variation. Deep neural networks dominate RF classification on pristine hardware, but their weights are precise and interdependent, causing catastrophic accuracy collapse as the underlying chip ages. We investigate whether Hyperdimensional Computing (HDC), a brain-inspired paradigm that distributes information across thousands of dimensions, can provide a reliability floor where Deep Learning fails. Using the RadioML 2016.10A dataset filtered to five digital modulations relevant to drone command-and-control links, we trained …

Go to article

Hands-On Ransomware: An Experiential Wannacry Case Study For Undergraduate Cybersecurity Education, Eli Creek Richmond, Thomas R. Devine Sep 2026

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.

Go to article

Breaking The Build: Detecting Software Supply Chain Vulnerabilities In Ci/Cd Pipelines, Mercedes R. Wahl, Dr. Benjamin Yankson Sep 2026

Breaking The Build: Detecting Software Supply Chain Vulnerabilities In Ci/Cd Pipelines, Mercedes R. Wahl, Dr. Benjamin Yankson

Military Cyber Affairs

This study examines whether integrating structured DevSec- Ops security controls into CI/CD pipelines can reduce software supply chain risk by preventing vulnerable components from progressing through the software development lifecycle. Software supply chain attacks frequently originate from weaknesses or compromises within dependencies, build environments, and trusted development stages, making early detection essential. A controlled sandbox experiment compared two pipeline configurations: a baseline CI/CD pipeline with no automated security enforcement and a secure DevSecOps pipeline integrating automated vulnerability scanning, SBOM generation, and artifact integrity verification. A known vulnerable dependency, the Python requests package (version 2.19.0) associated with CVE-2018-18074, was intentionally introduced …

Go to article

Mitigating Common Vulnerabilities And Exposures In Cobol-Based Critical Systems Using The Strangler-Fig Pattern, Lauren E. Caruso, Vincent J. Compeau, Assefaw H. Gebremedhin Sep 2026

Mitigating Common Vulnerabilities And Exposures In Cobol-Based Critical Systems Using The Strangler-Fig Pattern, Lauren E. Caruso, Vincent J. Compeau, Assefaw H. Gebremedhin

Military Cyber Affairs

COBOL-based legacy systems continue to underpin critical infrastructure in banking and government sectors despite their age and associated cybersecurity risks. Originally developed through a Department of Defense–sponsored initiative to standardize business computing, COBOL remains widely used in mission-critical environments. However, these systems face increasing vulnerabilities due to outdated security architectures, workforce shortages, and rising maintenance costs. This paper examines cybersecurity and operational challenges associated with COBOL systems and evaluates the Strangler Fig pattern as a modernization strategy that enables incremental replacement while maintaining continuity. The findings highlight implications for financial institutions and public-sector organizations dependent on legacy infrastructure.

Go to article

From Framework To Toolchain: Implementing Zero Trust Architecture In Cloud-Native Environments For Dow Compliance, Shelby C. Snyder Sep 2026

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 …

Go to article

Hybrid Deep (Cnn-Bilstm) Intrusion Detection For Defense And Mission-Critical Networks, Corey A Cheng, Jermaine Anim-Addo, Asma Jakir Hussain, Zion O Smith-Fox, Sanjay Goel, Yuksel Celik Sep 2026

Hybrid Deep (Cnn-Bilstm) Intrusion Detection For Defense And Mission-Critical Networks, Corey A Cheng, Jermaine Anim-Addo, Asma Jakir Hussain, Zion O Smith-Fox, Sanjay Goel, Yuksel Celik

Military Cyber Affairs

This article examines how hybrid deep learning can strengthen intrusion detection for military and defense networks. Using the CSE-CIC-IDS2018 dataset, the study evaluates a CNN-BiLSTM model designed to detect benign traffic and multiple attack categories, including DDoS, DoS, botnet, brute-force, web attack, and infiltration activity. The model achieved strong multi-class detection performance, with 0.9893 accuracy and 0.9979 ROC-AUC. The findings suggest that AI-supported intrusion detection can improve cyber defense operations, analyst triage, and protection of mission-critical networks.

Go to article

Are Large Language Models Safe? A Vulnerability Analysis Of Generated Source Code, James Richards-Perhatch, Mitchell Milander, James M. Halvorsen, Assefaw Gebremedhin Sep 2026

Are Large Language Models Safe? A Vulnerability Analysis Of Generated Source Code, James Richards-Perhatch, Mitchell Milander, James M. Halvorsen, Assefaw Gebremedhin

Military Cyber Affairs

The increasing complexity of software and demands for rapid deployment have pushed the software industry to rely more on large language models (LLMs) in developing source code. However, as this technology is still relatively recent, questions can arise about the safety of the generated code. This paper presents an analysis of seven LLMs with respect to the presence of vulnerabilities within source code. Our findings show that LLMs are more likely to produce vulnerable web applications than vulnerable C programs, and that vulnerabilities are more likely to occur when program size and complexity increases.

Go to article

Characterizing Advanced Persistent Threats With Cyber Attack Flow Metrics, Tyler Miller, Caleb Chang, Shouhuai Xu Sep 2026

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.

Go to article

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 Sep 2026

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 …

Go to article

Closing The Interpretability Gap: Explainable Ml-Based Malware Detection For Defensive Cyberspace Operations, Tashi Stirewalt, Sean Hodgson, Puumaaya Tahiru, Assefaw Gebremedhin Sep 2026

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 …

Go to article

Llm-Generated Countermeasures For Iot Cyberattacks, James Alger, Michael Tu Sep 2026

Llm-Generated Countermeasures For Iot Cyberattacks, James Alger, Michael Tu

Military Cyber Affairs

The rapid expansion of the Internet of Things (IoT) has introduced significant cybersecurity challenges, particularly for resource-constrained devices that traditional intrusion detection systems often fail to protect effectively. This paper proposes a novel, two-phase autonomous security pipeline designed to bridge the gap between probabilistic threat detection and deterministic network enforcement. The framework first utilizes a custom Time Series Transformer (TST) to classify multivariate network traffic and identify specific attack vectors, such as ransomware, SQL injections, and malicious file uploads. In the second phase, an agentic AI layer, comprising a locally hosted Large Language Model (LLM) orchestrated via LangGraph, processes the …

Go to article

Next Last