Robert J. Joyce
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Robert J. Joyce has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Charles Nicholas
Computer Science and Electrical Engineering, UMBC, Baltimore, MD, USA
5 shared publications - Edward Raff
Booz Allen Hamilton, McLean, VA, USA
5 shared publications - James Holt
Laboratory for Physical Sciences
3 shared publications - Derek Everett
Booz Allen Hamilton, McLean, VA, USA
2 shared publications - Steve Simske
Department of Systems Engineering, Colorado State University, Fort Collins, CO, USA
2 shared publications - Boian Stoianov Alexandrov
Los Alamos National Laboratory
1 shared publication
Latest Publications
- Open Access
- 18 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Document Engineering Issues in Malware Analysis
- 18 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Open AccessDocument Engineering Issues in Malware Analysis
- Open Access
- 18 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-10
MASCOT-Android: A Curated Dataset and Automated Collection Pipeline for Android Malware Source Code Specimens
- 18 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Open AccessMASCOT-Android: A Curated Dataset and Automated Collection Pipeline for Android Malware Source Code Specimens
- Open Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5988-5996
Zipf-Gramming: Scaling Byte N-Grams Up to Production Sized Malware Corpora
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 5988-5996
Open AccessZipf-Gramming: Scaling Byte N-Grams Up to Production Sized Malware Corpora
- Top Cited PapersOpen Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5516-5526
EMBER2024 - A Benchmark Dataset for Holistic Evaluation of Malware Classifiers
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5516-5526
Top Cited PapersOpen AccessEMBER2024 - A Benchmark Dataset for Holistic Evaluation of Malware Classifiers
- Open Access
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 277-286
ClarAVy: A Tool for Scalable and Accurate Malware Family Labeling
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 277-286
Open AccessClarAVy: A Tool for Scalable and Accurate Malware Family Labeling
- Open Access
- 26 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 185-196
AVScan2Vec: Feature Learning on Antivirus Scan Data for Production-Scale Malware Corpora
When investigating a malicious file, searching for related files is a common task that malware analysts must perform. Given that production malware corpora may contain over a billion files and consume petabytes of storage, many feature extraction and similarity search approaches are computationally infeasible. Our work explores the potential of antivirus (AV) scan data as a scalable source of features for malware. This is possible because AV scan reports are widely available through services such as VirusTotal and are ~100x smaller than the average malware sample. The information within an AV scan report is abundant with information and can indicate a malicious file's family, behavior, target operating system, and many other characteristics. We introduce AVScan2Vec, a neural model trained to comprehend the semantics of AV scan data. AVScan2Vec ingests AV scan data for a malicious file and outputs a meaningful vector representation. AVScan2Vec vectors are ~3 to 85x smaller than popular alternatives in use today, enabling faster vector comparisons and lower memory usage. By incorporating Dynamic Continuous Indexing, we show that nearest-neighbor queries on AVScan2Vec vectors can scale to even the largest malware production datasets. We also demonstrate that AVScan2Vec vectors are superior to other leading malware feature vector representations across nearly all classification, clustering, and nearest-neighbor lookup algorithms that we evaluated.
- 26 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 185-196
Open AccessAVScan2Vec: Feature Learning on Antivirus Scan Data for Production-Scale Malware Corpora
When investigating a malicious file, searching for related files is a common task that malware analysts must perform. Given that production malware corpora may contain over a billion files and consume petabytes of storage, many feature extraction and similarity search approaches are computationally infeasible. Our work explores the potential of antivirus (AV) scan data as a scalable source of features for malware. This is possible because AV scan reports are widely available through services such as VirusTotal and are ~100x smaller than the average malware sample. The information within an AV scan report is abundant with information and can indicate a malicious file's family, behavior, target operating system, and many other characteristics. We introduce AVScan2Vec, a neural model trained to comprehend the semantics of AV scan data. AVScan2Vec ingests AV scan data for a malicious file and outputs a meaningful vector representation. AVScan2Vec vectors are ~3 to 85x smaller than popular alternatives in use today, enabling faster vector comparisons and lower memory usage. By incorporating Dynamic Continuous Indexing, we show that nearest-neighbor queries on AVScan2Vec vectors can scale to even the largest malware production datasets. We also demonstrate that AVScan2Vec vectors are superior to other leading malware feature vector representations across nearly all classification, clustering, and nearest-neighbor lookup algorithms that we evaluated.
- Open Access
- 13 November 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Privacy and Security
- Vol. 26 (4) , 1-27
Semi-Supervised Classification of Malware Families Under Extreme Class Imbalance via Hierarchical Non-Negative Matrix Factorization with Automatic Model Selection
Identification of the family to which a malware specimen belongs is essential in understanding the behavior of the malware and developing mitigation strategies. Solutions proposed by prior work, however, are often not practicable due to the lack of realistic evaluation factors. These factors include learning under class imbalance, the ability to identify new malware, and the cost of production-quality labeled data. In practice, deployed models face prominent, rare, and new malware families. At the same time, obtaining a large quantity of up-to-date labeled malware for training a model can be expensive. In this article, we address these problems and propose a novel hierarchical semi-supervised algorithm, which we call the HNMFk Classifier , that can be used in the early stages of the malware family labeling process. Our method is based on non-negative matrix factorization with automatic model selection, that is, with an estimation of the number of clusters. With HNMFk Classifier , we exploit the hierarchical structure of the malware data together with a semi-supervised setup, which enables us to classify malware families under conditions of extreme class imbalance. Our solution can perform abstaining predictions, or rejection option, which yields promising results in the identification of novel malware families and helps with maintaining the performance of the model when a low quantity of labeled data is used. We perform bulk classification of nearly 2,900 both rare and prominent malware families, through static analysis, using nearly 388,000 samples from the EMBER-2018 corpus. In our experiments, we surpass both supervised and semi-supervised baseline models with an F1 score of 0.80.
- 13 November 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Privacy and Security
- Vol. 26 (4) , 1-27
Open AccessSemi-Supervised Classification of Malware Families Under Extreme Class Imbalance via Hierarchical Non-Negative Matrix Factorization with Automatic Model Selection
Identification of the family to which a malware specimen belongs is essential in understanding the behavior of the malware and developing mitigation strategies. Solutions proposed by prior work, however, are often not practicable due to the lack of realistic evaluation factors. These factors include learning under class imbalance, the ability to identify new malware, and the cost of production-quality labeled data. In practice, deployed models face prominent, rare, and new malware families. At the same time, obtaining a large quantity of up-to-date labeled malware for training a model can be expensive. In this article, we address these problems and propose a novel hierarchical semi-supervised algorithm, which we call the HNMFk Classifier , that can be used in the early stages of the malware family labeling process. Our method is based on non-negative matrix factorization with automatic model selection, that is, with an estimation of the number of clusters. With HNMFk Classifier , we exploit the hierarchical structure of the malware data together with a semi-supervised setup, which enables us to classify malware families under conditions of extreme class imbalance. Our solution can perform abstaining predictions, or rejection option, which yields promising results in the identification of novel malware families and helps with maintaining the performance of the model when a low quantity of labeled data is used. We perform bulk classification of nearly 2,900 both rare and prominent malware families, through static analysis, using nearly 388,000 samples from the EMBER-2018 corpus. In our experiments, we surpass both supervised and semi-supervised baseline models with an F1 score of 0.80.
- Open Access
- 16 August 2021
- Published by Association for Computing Machinery (ACM)
- p. 1
Document engineering issues in malware analysis
We present an overview of the field of malware analysis with emphasis on issues related to document engineering. We will introduce the field with a discussion of the types of malware, including executable binaries, malicious PDFs, polymorphic malware, ransomware, and exploit kits. We will conclude with our view of important research questions in the field. This is an updated version of tutorials presented in previous years, with more information about newly-available tools.
- 16 August 2021
- Published by Association for Computing Machinery (ACM)
- p. 1
Open AccessDocument engineering issues in malware analysis
We present an overview of the field of malware analysis with emphasis on issues related to document engineering. We will introduce the field with a discussion of the types of malware, including executable binaries, malicious PDFs, polymorphic malware, ransomware, and exploit kits. We will conclude with our view of important research questions in the field. This is an updated version of tutorials presented in previous years, with more information about newly-available tools.
- 26 May 2016
- Published by Transducer Research Foundation Inc. (TRF)
- p. 274-277
STABILIZATION OF CORIOLIS VIBRATORY GYROSCOPES BY FREQUENCY LOCKING TO ULTRA STABLE CLOCKS
- 26 May 2016
- Published by Transducer Research Foundation Inc. (TRF)
- p. 274-277
STABILIZATION OF CORIOLIS VIBRATORY GYROSCOPES BY FREQUENCY LOCKING TO ULTRA STABLE CLOCKS
- 1 January 2008
- Published by Society of Exploration Geophysicists
- p. 241-254
Temporal geophysical investigations of the FT‐2‐ plume at the Wurtsmith Air Force Base, Oscoda, Michigan
The decommissioned Wurtsmith Air Force Base former Fire Training Cell (FT‐02) facility has been the focus of several geophysical investigations. After several decades of fire training exercises, significant amounts of hydrocarbons and some solvents seeped into the subsurface contaminating the vadose and saturated zones in the source area. Groundwater geochemistry studies defined a contaminant plume that was approximately 125 m wide and > 300 m long. The boundary of the plume was further defined by using GPR, SP, and resistivity techniques. The source of the geophysical anomalies was attributed to biogeochemical modifications of the contaminated zone resulting from intrinsic bioremediation. In 2007, another integrated geophysical study of the site was conducted. GPR, SP, and electrical resistivity surveys were conducted with expectations of achieving similar results as the past investigations. However, there was a marked decrease in geophysical response from all of our geophysical techniques. The GPR anomaly has migrated deeper into the subsurface, the positive SP response was significantly attenuated, and the conductive resistivity anomaly has been replaced by background resistivity values. Also, six Geoprobe cores at three different locations were collected in order to conduct laboratory microbial counts and IP measurements. We attribute the attenuation of the observed geophysical anomalies to ongoing soil vapor extraction initiated in 2001. Significant removal of the contaminant mass by the vapor extraction system altered the subsurface biogeochemical conditions and these changes were documented by the 2007 geophysical data. The results of this study show that the attenuation of the contaminant plume is detectable with geophysical methods.
- 1 January 2008
- Published by Society of Exploration Geophysicists
- p. 241-254
Temporal geophysical investigations of the FT‐2‐ plume at the Wurtsmith Air Force Base, Oscoda, Michigan
The decommissioned Wurtsmith Air Force Base former Fire Training Cell (FT‐02) facility has been the focus of several geophysical investigations. After several decades of fire training exercises, significant amounts of hydrocarbons and some solvents seeped into the subsurface contaminating the vadose and saturated zones in the source area. Groundwater geochemistry studies defined a contaminant plume that was approximately 125 m wide and > 300 m long. The boundary of the plume was further defined by using GPR, SP, and resistivity techniques. The source of the geophysical anomalies was attributed to biogeochemical modifications of the contaminated zone resulting from intrinsic bioremediation. In 2007, another integrated geophysical study of the site was conducted. GPR, SP, and electrical resistivity surveys were conducted with expectations of achieving similar results as the past investigations. However, there was a marked decrease in geophysical response from all of our geophysical techniques. The GPR anomaly has migrated deeper into the subsurface, the positive SP response was significantly attenuated, and the conductive resistivity anomaly has been replaced by background resistivity values. Also, six Geoprobe cores at three different locations were collected in order to conduct laboratory microbial counts and IP measurements. We attribute the attenuation of the observed geophysical anomalies to ongoing soil vapor extraction initiated in 2001. Significant removal of the contaminant mass by the vapor extraction system altered the subsurface biogeochemical conditions and these changes were documented by the 2007 geophysical data. The results of this study show that the attenuation of the contaminant plume is detectable with geophysical methods.