Christopher Henshaw
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Christopher Henshaw has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Alan J. Michaels
Virginia Tech National Security Institute, Blacksburg, VA 24060, USA
6 shared publications - Elliott Rheault
Virginia Tech National Security Institute, Blacksburg, VA 24060, USA
2 shared publications - Jack Kolenbrander
Bradley Department of Electrical and Computer Engineering, Virginia Tech, Blacksburg, VA 24061, USA
2 shared publications - Jared Byers
Virginia Tech National Security Institute, 1311 Research Center Drive, Blacksburg, VA 24061, USA
2 shared publications - Ethan Husmann
Virginia Tech National Security Institute, Blacksburg, VA 24060, USA
1 shared publication - Jacob Dennis
Virginia Tech National Security Institute, Blacksburg, VA 24060, USA
1 shared publication
Latest Publications
- Open Access
- 20 September 2026
- Published by MDPI AG in Electronics
- Vol. 15 (18) , 4310
Augmentation of Sentiment Analysis Models
Sentiment analysis is widely used to quantify the emotional tone of text. However, many pretrained sentiment models are only designed for short inputs and can struggle when applied to longer compilations of text or even full documents. These limitations are especially important when a continuous sentiment score range is needed rather than a simple positive, neutral, or negative label. In this research, sixteen pretrained sentiment models were evaluated on sentence-level, paragraph-level, and document-level benchmark datasets to determine their compatibility for continuous sentiment scoring on a normalized output scale. To mitigate the effect of input truncation and preserve sentiment granularity on longer texts, an augmentation layer that combines sentence-level, paragraph-level, and document-level sentiment estimates through weighted aggregation and sliding window sentence aggregation was added to each compatible model. Model performance was calculated using RMSE against ground-truth benchmark datasets, and model behavior was further examined through score distributions and residual correlations. Results show that the augmentation layer generally improved performance for paragraph-level and document-level text, with more noticeable improvements at the document level. Among the evaluated pretrained models, DeBERTa provided the best overall balance of benchmark performance and continuous score resolution, making it the most defensible model for downstream sentiment analysis, although substantial rank agreement was observed among the strongest candidate models.
- 20 September 2026
- Published by MDPI AG in Electronics
- Vol. 15 (18) , 4310
Open AccessAugmentation of Sentiment Analysis Models
Sentiment analysis is widely used to quantify the emotional tone of text. However, many pretrained sentiment models are only designed for short inputs and can struggle when applied to longer compilations of text or even full documents. These limitations are especially important when a continuous sentiment score range is needed rather than a simple positive, neutral, or negative label. In this research, sixteen pretrained sentiment models were evaluated on sentence-level, paragraph-level, and document-level benchmark datasets to determine their compatibility for continuous sentiment scoring on a normalized output scale. To mitigate the effect of input truncation and preserve sentiment granularity on longer texts, an augmentation layer that combines sentence-level, paragraph-level, and document-level sentiment estimates through weighted aggregation and sliding window sentence aggregation was added to each compatible model. Model performance was calculated using RMSE against ground-truth benchmark datasets, and model behavior was further examined through score distributions and residual correlations. Results show that the augmentation layer generally improved performance for paragraph-level and document-level text, with more noticeable improvements at the document level. Among the evaluated pretrained models, DeBERTa provided the best overall balance of benchmark performance and continuous score resolution, making it the most defensible model for downstream sentiment analysis, although substantial rank agreement was observed among the strongest candidate models.
- Open Access
- 4 September 2026
- Published by MDPI AG in Electronics
- Vol. 15 (17) , 3996
Comparative Sentiment Analysis of Campaign and Third-Party Communications in the 2024 U.S. Elections
In an era defined by constant digital interaction, personal information is frequently collected, shared, and repurposed in ways that raise serious concerns about privacy and transparency. As part of a broader project, which investigates how individuals’ personal information is used, an experiment was designed to test how political campaigns wittingly or unwittingly use personal information. To study these behaviors in a realistic setting, believable synthetic identities were generated and a controlled sign-up framework was built to register those identities across almost every pre-primary campaign in the 2024 U.S. election cycle, enabling the longitudinal collection of emails, voicemails, and SMS messages. Over three years, these communications were accumulated and aggregated, and a sentiment-analysis pipeline was then developed to characterize the tone of incoming messages. This analysis consisted of classifying each of the 77k emails in a continuous range from extremely negative (0) to extremely positive (1) and then analyzing their behavior. The descriptive results showed that while the Democratic and Republican parties were close to one another with moderate, marginally positive average sentiments, Republican emails contained a higher observed proportion of emails in both tails of the sentiment distributions. These party-level differences were not tested for statistical significance and should therefore be interpreted as descriptive observations of the collected dataset. More extreme sentiments are routinely displayed, however, toward the candidates by news sources and special interest groups, with a visible translation of expressed sentiment to apparent political bias.
- 4 September 2026
- Published by MDPI AG in Electronics
- Vol. 15 (17) , 3996
Open AccessComparative Sentiment Analysis of Campaign and Third-Party Communications in the 2024 U.S. Elections
In an era defined by constant digital interaction, personal information is frequently collected, shared, and repurposed in ways that raise serious concerns about privacy and transparency. As part of a broader project, which investigates how individuals’ personal information is used, an experiment was designed to test how political campaigns wittingly or unwittingly use personal information. To study these behaviors in a realistic setting, believable synthetic identities were generated and a controlled sign-up framework was built to register those identities across almost every pre-primary campaign in the 2024 U.S. election cycle, enabling the longitudinal collection of emails, voicemails, and SMS messages. Over three years, these communications were accumulated and aggregated, and a sentiment-analysis pipeline was then developed to characterize the tone of incoming messages. This analysis consisted of classifying each of the 77k emails in a continuous range from extremely negative (0) to extremely positive (1) and then analyzing their behavior. The descriptive results showed that while the Democratic and Republican parties were close to one another with moderate, marginally positive average sentiments, Republican emails contained a higher observed proportion of emails in both tails of the sentiment distributions. These party-level differences were not tested for statistical significance and should therefore be interpreted as descriptive observations of the collected dataset. More extreme sentiments are routinely displayed, however, toward the candidates by news sources and special interest groups, with a visible translation of expressed sentiment to apparent political bias.
- Open Access
Number Recognition Through Color Distortion Using Convolutional Neural Networks
Machine learning applied to image-based number recognition has made significant strides in recent years. Recent use of Large Language Models (LLMs) in natural language search and generation of text have improved performance for general images, yet performance limitations still exist for data subsets related to color blindness. In this paper, we replicated the training of six distinct neural networks (MNIST, LeNet5, VGG16, AlexNet, and two AlexNet modifications) using deep learning techniques with the MNIST dataset and the Ishihara-Like MNIST dataset. While many prior works have dealt with MNIST, the Ishihara adaption addresses red-green combinations of color blindness, allowing for further research in color distortion. Through this research, we applied pre-processing to accentuate the effects of red-green and monochrome colorblindness and hyper-parameterized the existing architectures, ultimately achieving better overall performance than currently published in known works.
Open AccessNumber Recognition Through Color Distortion Using Convolutional Neural Networks
Machine learning applied to image-based number recognition has made significant strides in recent years. Recent use of Large Language Models (LLMs) in natural language search and generation of text have improved performance for general images, yet performance limitations still exist for data subsets related to color blindness. In this paper, we replicated the training of six distinct neural networks (MNIST, LeNet5, VGG16, AlexNet, and two AlexNet modifications) using deep learning techniques with the MNIST dataset and the Ishihara-Like MNIST dataset. While many prior works have dealt with MNIST, the Ishihara adaption addresses red-green combinations of color blindness, allowing for further research in color distortion. Through this research, we applied pre-processing to accentuate the effects of red-green and monochrome colorblindness and hyper-parameterized the existing architectures, ultimately achieving better overall performance than currently published in known works.
- Open Access
- 13 August 2024
- Published by MDPI AG in Journal of Cybersecurity and Privacy
- Vol. 4 (3) , 572-593
Use and Abuse of Personal Information, Part I: Design of a Scalable OSINT Collection Engine
In most open-source intelligence (OSINT) research efforts, the collection of information is performed in an entirely passive manner as an observer to third-party communication streams. This paper describes ongoing work that seeks to insert itself into that communication loop, fusing openly available data with requested content that is representative of what is sent to second parties. The mechanism for performing this is based on the sharing of falsified personal information through one-time online transactions that facilitate signup for newsletters, establish online accounts, or otherwise interact with resources on the Internet. The work has resulted in the real-time Use and Abuse of Personal Information OSINT collection engine that can ingest email, SMS text, and voicemail content at an enterprise scale. Foundations of this OSINT collection infrastructure are also laid to incorporate an artificial intelligence (AI)-driven interaction engine that shifts collection from a passive process to one that can effectively engage with different classes of content for improved real-world privacy experimentation and quantitative social science research.
- 13 August 2024
- Published by MDPI AG in Journal of Cybersecurity and Privacy
- Vol. 4 (3) , 572-593
Open AccessUse and Abuse of Personal Information, Part I: Design of a Scalable OSINT Collection Engine
In most open-source intelligence (OSINT) research efforts, the collection of information is performed in an entirely passive manner as an observer to third-party communication streams. This paper describes ongoing work that seeks to insert itself into that communication loop, fusing openly available data with requested content that is representative of what is sent to second parties. The mechanism for performing this is based on the sharing of falsified personal information through one-time online transactions that facilitate signup for newsletters, establish online accounts, or otherwise interact with resources on the Internet. The work has resulted in the real-time Use and Abuse of Personal Information OSINT collection engine that can ingest email, SMS text, and voicemail content at an enterprise scale. Foundations of this OSINT collection infrastructure are also laid to incorporate an artificial intelligence (AI)-driven interaction engine that shifts collection from a passive process to one that can effectively engage with different classes of content for improved real-world privacy experimentation and quantitative social science research.
- Open Access
- 11 August 2024
- Published by MDPI AG in Journal of Cybersecurity and Privacy
- Vol. 4 (3) , 546-571
Use & Abuse of Personal Information, Part II: Robust Generation of Fake IDs for Privacy Experimentation
When personal information is shared across the Internet, we have limited confidence that the designated second party will safeguard it as we would prefer. Privacy policies offer insight into the best practices and intent of the organization, yet most are written so loosely that sharing with undefined third parties is to be anticipated. Tracking these sharing behaviors and identifying the source of unwanted content is exceedingly difficult when personal information is shared with multiple such second parties. This paper formulates a model for realistic fake identities, constructs a robust fake identity generator, and outlines management methods targeted towards online transactions (email, phone, text) that pass both cursory machine and human examination for use in personal privacy experimentation. This fake ID generator, combined with a custom account signup engine, are the core front-end components of our larger Use and Abuse of Personal Information system that performs one-time transactions that, similar to a cryptographic one-time pad, ensure that we can attribute the sharing back to the single one-time transaction and/or specific second party. The flexibility and richness of the fake IDs also serve as a foundational set of control variables for a wide range of social science research questions revolving around personal information. Collectively, these fake identity models address multiple inter-disciplinary areas of common interest and serve as a foundation for eliciting and quantifying personal information-sharing behaviors.
- 11 August 2024
- Published by MDPI AG in Journal of Cybersecurity and Privacy
- Vol. 4 (3) , 546-571
Open AccessUse & Abuse of Personal Information, Part II: Robust Generation of Fake IDs for Privacy Experimentation
When personal information is shared across the Internet, we have limited confidence that the designated second party will safeguard it as we would prefer. Privacy policies offer insight into the best practices and intent of the organization, yet most are written so loosely that sharing with undefined third parties is to be anticipated. Tracking these sharing behaviors and identifying the source of unwanted content is exceedingly difficult when personal information is shared with multiple such second parties. This paper formulates a model for realistic fake identities, constructs a robust fake identity generator, and outlines management methods targeted towards online transactions (email, phone, text) that pass both cursory machine and human examination for use in personal privacy experimentation. This fake ID generator, combined with a custom account signup engine, are the core front-end components of our larger Use and Abuse of Personal Information system that performs one-time transactions that, similar to a cryptographic one-time pad, ensure that we can attribute the sharing back to the single one-time transaction and/or specific second party. The flexibility and richness of the fake IDs also serve as a foundational set of control variables for a wide range of social science research questions revolving around personal information. Collectively, these fake identity models address multiple inter-disciplinary areas of common interest and serve as a foundation for eliciting and quantifying personal information-sharing behaviors.
- 20 January 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 71 (4) , 3670-3686
CAN Bus Message Authentication via Co-Channel RF Watermark
Over the past decade, consumer vehicles have increasingly begun to resemble computers on wheels, with many newer vehicles containing over 100 independent electronic control units (ECU) that make independent decisions within the vehicle. Cost and power constraints for the vehicle significantly constrain processing capabilities and limitations on practical cybersecurity measures that may be employed in the vehicle today. Also, as automotive manufacturers are continuous integrators, many legacy communications links within vehicles were originally designed as standalone entities that assumed internal trust of communications between ECUs by virtue of closed bus structures and an emphasis on reliability, but not security. This paper presents a low-cost authentication mechanism to significantly improve message validation on wired message buses such as the Controller Area Network (CAN) bus by employing a co-channel underlay watermark that arrives time-aligned with the primary bus messages. Concept validation using a gate-level Simulink / HDL Coder simulation provided initial feasibility, while a complete CAN bus hardware testbed was constructed with a custom CAN transceiver built on a Red Pitaya software defined radio (SDR) connected to multiple commercial CAN transceivers to validate backwards compatibility. This testbed incorporates a custom transmission security (TRANSEC) engine that supports dynamic variation of the watermark, further improving security during deployment. Finally, this custom CAN testbed with the watermarking CAN transceiver was tested across a range of operational conditions, with initial performance results presented and discussed. Future work describes potential for near-term ruggedization and productization of the watermarking CAN transceiver.
- 20 January 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Vehicular Technology
- Vol. 71 (4) , 3670-3686
CAN Bus Message Authentication via Co-Channel RF Watermark
Over the past decade, consumer vehicles have increasingly begun to resemble computers on wheels, with many newer vehicles containing over 100 independent electronic control units (ECU) that make independent decisions within the vehicle. Cost and power constraints for the vehicle significantly constrain processing capabilities and limitations on practical cybersecurity measures that may be employed in the vehicle today. Also, as automotive manufacturers are continuous integrators, many legacy communications links within vehicles were originally designed as standalone entities that assumed internal trust of communications between ECUs by virtue of closed bus structures and an emphasis on reliability, but not security. This paper presents a low-cost authentication mechanism to significantly improve message validation on wired message buses such as the Controller Area Network (CAN) bus by employing a co-channel underlay watermark that arrives time-aligned with the primary bus messages. Concept validation using a gate-level Simulink / HDL Coder simulation provided initial feasibility, while a complete CAN bus hardware testbed was constructed with a custom CAN transceiver built on a Red Pitaya software defined radio (SDR) connected to multiple commercial CAN transceivers to validate backwards compatibility. This testbed incorporates a custom transmission security (TRANSEC) engine that supports dynamic variation of the watermark, further improving security during deployment. Finally, this custom CAN testbed with the watermarking CAN transceiver was tested across a range of operational conditions, with initial performance results presented and discussed. Future work describes potential for near-term ruggedization and productization of the watermarking CAN transceiver.