David Eargle
Information Systems Management Researcher
Updated 2021/09/25
Currently
2017-present
I am an Assistant Professor in the Organizational Leadership and Information Analytics group
at the
Leeds School of Business, University of Colorado Boulder
2017-present
I also have a courtesy appointment as Assistant Professor in the
Department of Information Science
at the University of Colorado Boulder
Research Interests
- Human side of Information Systems Security (Behavioral InfoSec)
- Dark Side of Information Technology, including online group polarization
- Neuroscience applications to Human-Computer Interaction and Information Security
- Machine learning and visualization approaches to performing literature reviews
Teaching Interests
- Cybersecurity management
- Descriptive, Predictive, and Prescriptive Business Analytics
- Programming for business majors (Python, R); Databases; Cloud Computing, Web and app development
Education
2017
Ph.D., Information Systems and Technology Management
Katz Graduate School of Business, University of Pittsburgh, Pittsburgh, Pennsylvania, USA.
Dissertation: “Security Messages: Or, How I Learned to Stop Disregarding and Heed the Warning” (pdf)
- Two papers exploring whether integrating human facial expressions of fear and threat into security messages can help invoke attention and more secure behavior. A third paper exploring the degree to which various magnitudes of monetary cost impact work/study performance-security tradeoff behaviors.
- Used online lab study methods (amazon mturk) and fMRI; used deception protocols.
2013
Master of Information Systems Management
Bachelor of Science, Information Systems
Marriott School of Management, Brigham Young University, Provo, Utah, USA.
Magna Cum Laude with University Honors
Other Work Experience
2011–13
Academic Research Assistant
Department of Information Systems at Brigham Young University
- Duties included literature reviews, online lab experiment app development and server administration
2012
Digital Forensics Analyst Intern
Paraben Corporation, Ashburn, VA
- Consulted with police chiefs to create tools to log officer access to mobile forensic terminals
- Created a powershell tool to improve quality control for production of Paraben Data Recovery USB sticks
- Completed level 2 mobile forensics training
2011-12
Network and Systems Administrator
Better Logic LLC, Orem, UT
- Set up a a VMware vSphere ESXi server and migrated several bare-metal systems to be virtualized on it.
- Configured Bacula on the hypervisor
2010-12
Web Developer
Center for Teaching and Learning, BYU
- Core developer for syllabus section of Learning Suite, BYU’s Learning Management System
Scholarship metrics
- My Google scholar profile
- ORCiD
- 728 citations per Google Scholar (641 citations since 2016)
- 10 H-index per Google Scholar
Journal Publications
Forthcoming
Vance, A., Eargle, D., Eggett, D., Straub, D., Ouimet, K. “Do Security Fear Appeals
Work When They Interrupt Tasks? A Multi-Method Examination of Password Strength,”
MIS Quarterly, forthcoming.
2020
Kirwan, C., Bjornn, D., Anderson, B., Vance, A., Eargle, D., Jenkins, J.2020. “Repetition
of Computer Security Warnings Results in Differential Repetition Suppression Effects as
Revealed With Functional MRI,” Frontiers in Psychology, 11, pp. 1–10.
2019
Veen, Hendrik van, Nathaniel Saul, David Eargle, and Sam Mangham. “Kepler Mapper:
A Flexible Python Implementation of the Mapper Algorithm.”
Journal of Open Source Software 4, no. 42 (2019): 1315.
2016
Anderson, B.B., Vance, A., Kirwan, C.B., Jenkins, J. and Eargle, D. “From warnings to
wallpaper: Why the brain habituates to security warnings and what can be done about
it.” Journal of Management Information Systems, 33, 3 (2016), 713-743.
doi: 10.1080/07421222.2016.1243947
2016
Anderson, B.B., Jenkins, J., Vance, A., Kirwan, C.B. and Eargle, D. “Your memory is
working against you: How eye tracking and memory explain habituation to security
warnings.” Decision Support Systems, 92 (2016), 3-13. doi: 10.1016/j.dss.2016.09.010
2016
Jenkins, J., Anderson, B., Vance, A., Kirwan, B. and Eargle, D. “More harm than good?
How security messages that interrupt make us vulnerable.”
Information Systems Research, 27, 4 (2016), 880-896. Awarded ISR’s “Best Published Paper” for 2016.
doi: 10.1287/isre.2016.0644
2016
Anderson, B., Vance, A., Kirwan, B., Eargle, D. and Jenkins, J. “How users perceive and
respond to security messages: A NeuroIS research agenda and empirical study.”
European Journal of Information Systems, 25, 4 (2016), 364-390.
doi: 10.1057/ejis.2015.21
2015
Anderson, B., Kirwan, B., Eargle, D., Jensen, S. and Vance, A. “Neural correlates of
gender differences and color in distinguishing security warnings and legitimate websites:
A neurosecurity study.” Journal of Cybersecurity, 1, 1 (2015), 109-120.
doi: 10.1093/cybsec/tyv005
2014
Vance, A., Anderson, B.B., Kirwan, C.B. and Eargle, D. “Using measures of risk
perception to predict information security behavior: Insights from
electroencephalography (EEG).” Journal of the Association for Information Systems,
15, 10 (2014), 679-722.
Conference Publications
2020
Larsen KR, Gefen D, Petter S, Eargle D. (2020) “Creating Construct Distance Maps
with Machine Learning: Stargazing Trust.” In
Conference of the Association for Information Systems (AMCIS 2020). Online.
Awarded AMCIS’ “Best Completed Paper” for 2020. 60% acceptance rate.
2019
A Vance, D Eargle, JL Jenkins, CB Kirwan, BB Anderson. (2019) “The Fog of
Warnings: How Non-Essential Notifications Blur with Security Warnings.”
In Fifteenth Symposium on Usable Privacy and Security (SOUPS 2019). Santa Clara, CA: USENIX
Association, 2019. https://www.usenix.org/conference/soups2019/presentation/vance
2019
Kirwan, C.B., Anderson, B., Eargle, D., Jenkins, J., and Vance, A. (2019, October).
Attentional habituation to non-essential computer notifications generalizes to security
warnings: an fMRI study. Program No. 665.##. Neuroscience 2019 Abstracts.
Washington, DC: Society for Neuroscience, 2019. Online.
2019
Anderson, B., Kirwan, B., Eargle, D., Jenkins, J., Vance, A., “Neural Evidence of
Generalization of Software Notifications to Security Warnings,”
Security and Human Behavior Workshop, Harvard University, June 2019
2019
Kirwan, C.B., Anderson, B., Eargle, D., Jenkins, J., and Vance, A. (2019, June). Using
fMRI to Measure Stimulus Generalization of Software Notification to Security Warnings.
Retreat on NeuroIS, Vienna, Austria. Information Systems and Neuroscience, 93-99.
2017
Eargle, D., Galletta, D., Shadi, J., Dimitar, K., and Shivendu, S. “The Chaos of Order:
Sequence and Mindlessness Effects in Obtaining Successive App Permissions.” In
Workshop on Information Security & Privacy. Seoul, South Korea: AIS SIGSEC and IFIP
TC11.1. (2017).
2016
Eargle, D., Galletta, D. and Jenkins, J. “How much is your security worth? Applying a
risk tradeoff paradigm to explain the bimodal nature of user elaboration over interruptive
security messages.” In Workshop on Information Security & Privacy, Dublin, Ireland: AIS
SIGSEC and IFIP TC11.1 (2016).
2016
Eargle, D., Galletta, D. and Cranor, L. “On the use of motivational components as
attention hooks in security message interface design: Avoiding “tl;dr”.” In
Dewald Roode Workshop on Information Systems Security Research, Albuquerque, New Mexico: IFIP
WG8.11/WG11.13 (2016).
2016
Eargle, D., Galletta, D., Kirwan, C. B., Vance, A., and Jenkins, J. 2016. “Integrating
Facial Cues of Threat into Security Warnings – an fMRI and Field Study.” Paper
presented at the Americas Conference on Information Systems (AMCIS), San Diego,
California.
2015
Galletta, D., Eargle, D., Shadi, J., Kunev, D. and Singh, S. “Integrating social and
economic models of responding to privacy messages in mobile computing: A research
agenda.” In Workshop on Information Security & Privacy, Fort Worth, Texas: AIS
SIGSEC and IFIP TC11.1 (2015).
2015
Eargle, D., Godfrey, J., Miao, H., Stevenson, S., Shay, R., Ur, B. and Cranor, L. “Poster:
You can do better – motivational statements in password-meter feedback.” In
Symposium on Usable Privacy and Security (SOUPS ‘15), Ottawa, CA: (2015).
2015
Eargle, D., Galletta, D., Kirwan, C.B. and Vance, A. “Integrating facial threat signals into
security messages: An extension of media naturalness theory to an information security
context.” In Dewald Roode Workshop on Information Systems Security Research,
Newark, Delaware: IFIP WG8.11/WG11.13 (2015).
2015
Anderson, B., Kirwan, B., Jenkins, J., Eargle, D., Howard, S. and Vance, A. “How
polymorphic warnings reduce habituation in the brain: Insights from an fMRI study.” In
Proceedings of the ACM Conference on Human Factors in Computing Systems (CHI),
Seoul, South Korea: ACM (2015).
2014
Eargle, D., Taylor, R., Sawyer, L. and Gaskin, J. “Acquiring IS skill through habitual
use.” In 2014 47th Hawaii International Conference on System Sciences (HICSS):
(2014), pp. 3-12.
2014
Eargle, D., Galletta, D. and Siegle, G. “Using fearful facial facial expressions to increase
the effectiveness of protective security messages: Proposing an fMRI and field study.” In
The Dewald Roode Workshop on Information Systems Security Research, IFIP WG8.11/WG11.13, Newcastle, UK: (2014).
2014
Anderson, B., Vance, A., Kirwan, B., Eargle, D. and Howard, S. “Why users habituate to
security warnings: Insights from fMRI.” In
The Dewald Roode Workshop on Information Systems Security Research, IFIP WG8.11/WG11.13, Newcastle, UK: (2014).
2014
Anderson, B., Vance, A., Kirwan, B., Eargle, D. and Howard, S. “Users aren’t
(necessarily) lazy: Using NeuroIS to explain habituation to security warnings.” In
International Conference on Information Systems, Auckland, New Zealand: AIS (2014).
2013
Vance, A., Eargle, D., Ouimet, K. and Straub, D. “Enhancing password security through
interactive fear appeals: A web-based field experiment.” In
2013 46th Hawaii International Conference on System Sciences (HICSS): (2013), pp. 2988-2997.
2013
Vance, A., Anderson, B., Brock, K. and Eargle, D. “Using measures of risk perception to
predict information security behavior: Insights from electroencephalography (EEG).” In
JAIS workshop, Gmunden Retreat on NeuroIS, Gmunden, Austria: (2013).
2013
Eargle, D., Vance, A.O. and Lowry, P.B. “How moral intensity and impulsivity moderate
the influence of accountability on access policy violations in information systems.” In
Proceedings of the Eighth Pre-ICIS Workshop on Information Security and Privacy:
(2013).
2013
Anderson, B., Vance, A., Eargle, D. and Kirwan, C.B. “Your memory is working against
you: How eye tracking and memory explain susceptibility to phishing.” In
The Dewald Roode Workshop on Information Systems Security Research, IFIP WG8.11/WG11.13:
(2013).
2013
Anderson, B., Vance, A. and Eargle, D. “Is your susceptibility to phishing dependent on
your memory?” In Proceedings of the Eighth Pre-ICIS Workshop on Information Security and Privacy, Milan, Italy: (2013).
2012
Vance, A., Eargle, D., Ouimet, K. and Straub, D. “How interactivity can enhance the
effectiveness of fear appeals: A web-based field experiment of password security.” In
The Dewald Roode Workshop on Information Systems Security Research, IFIP WG8.11/WG11.13, Provo, UT: (2012).
2012
Eargle, D., Vance, A., Allen, G., Barrick, D., Bearnson, T. and Tialin, T. “Justifying
breaking the glass: How accountability can deter unauthorized access.” In WISP 2012
Proceedings, Orlando, Florida: AIS SIGSEC and IFIP TC11.1 (2012).
2012
Anderson, B., Vance, A., Hansen, J., Kirwan, B., Eargle, D., Hinkle, L. and Weagel, A.
“Neural correlates of gender differences in distinguishing malware warnings and
legitimate websites: A NeuroIS study.” In IFIP WG8.11/WG11.13, Provo, UT: (2012).
Works in progress
See https://daveeargle.com/projects for links to resources for these works in progress.
1.
A Spoonful of Sugar: Blending Online News Source and Content to Counter Ideological-Alignment News Biases and Encourage Political Group Depolarization
With: Valerie Bartelt, Zlatana Nenova, Dennis Galletta
Anecdotes suggest that political group polarization may impact readers’ perceptions of news articles so strongly that readers may call articles “fake news” solely based on their ideological alignment with the publication source, regardless of the article’s content. While researchers have explored confirmation bias in social media, studies have not yet teased out the differential effects of reader ideological alignment with article content (“content-friendliness”) and source (“source-friendliness”) on attitudes, beliefs, and intended behaviors. Using a mixed design, 133 MTurk participants read and reacted to polarizing news articles, with article-content being presented as if from random sources.
2.
The Fog of Warnings: How Non-essential Notifications Blur with Security Warnings
With: Anthony Vance, Bonnie Anderson, Brock Kirwan, Jeff Jenkins
Through a series of lab and field experiments, the impact of exposure to system notifications of varying degree of visual similarity to security messages is assessed using objective methods such as reaction times and fMRI response data.
Targeting MISQ Submission October 2021
- Conference version
- A Vance, D Eargle, JL Jenkins, CB Kirwan, BB Anderson. (2019) “The Fog of Warnings: How Non-Essential Notifications Blur with Security Warnings.” In Fifteenth Symposium on Usable Privacy and Security (SOUPS 2019). Santa Clara, CA: USENIX Association, 2019. https://www.usenix.org/conference/soups2019/presentation/vance
3.
How much is your security worth? Applying a risk tradeoff paradigm to explain the bimodal nature of user elaboration over interruptive security messages
With: Dennis Galletta
Why do employees disregard computer security messages, opening the organization to potential information security breaches? One research perspective assumes that humans who fall prey to such attacks solely use automatic information processing, and therefore, user interfaces (such as Google Chrome browser security popups and overlays or Microsoft Word security dialogs) must be better designed to capture and hold attention, and to educate users, to the end that users more carefully consciously evaluate their information security decisions. However, this research project takes the view that employees also make monetary cost-benefit approaches to adhering to or disregarding security messages. It gathers data using a series of online deception-protocol website experiments, wherein users are exposed to security messages that interrupt an ostensible primary task. Psychometric measures of attention, including mouse-cursor tracking and reaction times, are captured and used to predict security behaviors. The monetary “cost” of disregarding a security message is experimentally varied, and its impact on prompting attention and security behaviors is examined. Survey data and focus group data is also captured to probe users’ thought processes.
4.
Creating Construct Distance Maps with Machine Learning: Stargazing Trust
With: Kai Larsen, David Gefen, Stacie Petter
A design-science approach to creating a tool to graph the nomological space of all survey items used in information systems literature. Applies methods from the domain of topological data analysis to visually graph the nomological space, based on predicted “distances” between item pairs generated by a machine learning predictive model trained on a sampling of survey item-pair relationships (distances) coded by domain experts. Besides leading to insights into already-used IS constructs, the resulting tool can be used to identify placement of new survey items in context in the nomological space.
Ongoing research.
- AMCIS Citation
- Larsen KR, Gefen D, Petter S, Eargle D. (2020) “Creating Construct Distance Maps with Machine Learning: Stargazing Trust.” In Conference of the Association for Information Systems (AMCIS 2020). Online. Awarded AMCIS’ “Best Completed Paper” for 2020. 60% acceptance rate.
5.
When Bots Attack: Threat Modeling and Mitigations of Attacks Against Online Behavioral Experiments
With: Todd M. Gureckis, Jordan W. Suchow
Psychology and behavioral data is increasingly shifting to being collected online, instead of in brick-and-mortar lab rooms. However, panic has arisen about the degree to which such data is impacted by “bots”, or by malicious actors gaming the system in order to maximize participation payouts. This paper applies models from cybersecurity – specifically, the NIST Cybersecurity Framework’s Five Functions – to systematically evaluate the threat of bots, and to show the process by which controls can be developed to mitigate identified threats. Several cross-industry controls are suggested, including the development of machine learning models to detect anomalous participant behavior, aggregated across participating researchers’ data. The behavioral research community can use these models to defend collected data, and to argue for cross-industry grants to develop novel approaches.
Ongoing research
Teaching Experience
Leeds School of Business, University of Colorado
2017-Present
In four years:
- Taught 4 unique courses
- Taught at 3000, 4000 (including honors students), and 5000 (graduate, including MBAs) levels
Courses:
Information Security Management (graduate)
- Taught within the Masters of Business Analytics—Security track.
- Compared to the undergraduate offering, this course’s lectures have a stronger focus on specific security behaviors that generate data amenable to machine learning – e.g., post mortem reports from Mandiant and the House Oversight Committee (Equifax, OPM)
Information Security Management (undergraduate)
- Exploration of human, organizational, and technical domains of information security management.
- Self-created hands-on Google Cloud virtual machine labs to teach students to “think like attackers”
Security Analytics
- A projects-based class focused on applying machine learning to security-related data. Topics include malware classification (binomial and multinomial), modeling using mobile sensor data, network traffic parsing (PCAPS => netflows) and malicious IP, domain classification
- A focus on using python-sklearn – on reading documentation and source code
- Also a focus on “open data science” – on hosted Jupyter notebooks, on using Git and Github to store and share code projects. Also on sharing and programmatically consuming shared data.
- Labs have students host models behind API endpoints (Flask app). Models are also deployed to AWS and GCP’s machine learning platforms.
Business Analytics
- Descriptive: querying, and ETL/wrangling data with Alteryx
- Predictive: supervised vs unsupervised machine learning algorithms
- Used Alteryx and DataRobot AutoML
- Covering topics such as association rules, k-means clusters, regressions, correlations, and text mining
Course Evaluation Metrics:
The table below shows course evaluation metrics for each semester-course I have taught at CU Boulder. The “Course Overall” and “Instructor Overall” columns also include college averages parenthetically, when available.
AY Year | Term | Course | Level | Enrolled | Course Overall (college avg.) | Instructor Overall (college avg.) |
---|---|---|---|---|---|---|
2017‑18 | Fall | Infosec Management | ugrad | 19 | 5.5/6.0 (4.6) | 5.8/6.0 (5.1) |
Spring | Business Analytics | ugrad | 41 | 3.8/6.0 (4.7) | 4.2/6.0 (4.2) | |
2018‑19 | Fall | Infosec Management | ugrad & grad (incl. MBAs) | 45 | 4.6/6.0 (4.5) | 4.6/6.0 (5.0) |
Business Analytics | ugrad | 45 | 3.6/6.0 (4.5) | 4.0/6.0 (5.0) | ||
2019‑20 | Fall | Infosec Management | ugrad | 39 | 5.2/6.0 (4.5) | 5.4/6.0 (5.0) |
Infosec Management | grad (incl. MBAs) | 10 | 5.8/6.0 (4.5) | 5.7/6.0 (5.0) | ||
Spring | Security Analytics with Python | grad (incl. MBAs) | 6 | 4.9/5.0 1,2 | 4.9/5.0 1,2 | |
2020‑21 | Fall | Infosec Management | ugrad (honors) | 33 | 4.4/5.0 (4.3) 1 | 4.4/5.0 (4.3) 1 |
Infosec Management | grad (incl. MBAs) | 13 | 4.7/5.0 (4.3) 1 | 4.7/5.0 (4.3) 1 | ||
Spring | Security Analytics with Python | grad | 14 | 4.7/5.0 (4.4) 1 | 4.7/5.0 (4.4) 1 |
1 Average of all available metrics. Typical course and instructor overall evaluation metrics not collected by CU during these semesters because of Covid.
2 Due to the mid-semester move to remote learning, college FCQ results for spring 2020 are not available.
College of Business Administration at the University of Pittsburgh
2017
Database Management
- Taught relational database structures and data querying in MySQL and R to Juniors and Seniors from various departments
2015
Introduction to Information Systems Management
- Full responsibility for a class of 60 undergraduate students from various departments of the University of Pittsburgh’s College of Business Administration.
- Complete direction over course curriculum, policies, and syllabus.
2013-2015
Microsoft Excel workshops for Katz Graduate students
- Taught four beginner-to-advanced-level Microsoft Excel workshops to part-time Katz MBA students
Department of Information Systems, Marriott School of Management, Brigham Young University
2013
Spreadsheets for Business Majors
Full responsibility for four college-level class sections on computer spreadsheet skills, with total enrollment of over 270 across four sections. Mix of online plus in-class teaching. Oversight of three teaching assistants.
External Service
2014 to Present
JMIS Website Editor
Web administrator for the Journal of Management Information
Systems’ web presence (https://jmis-web.org), working directly with Editor in Chief
Vladimir Zwass.
Fall 2011 to Present
AIS IS Theory Wiki Editor
Systems administrator and managing editor for the Information Systems Theory Community Wiki,
https://is.theorizeit.org, affiliated with the Association for Information Systems.
June 2016 to present
psiTurk Project Leader
Lead developer for an open platform for science on Amazon Mechanical Turk, hosted on github.
Used by researchers around the world.
Ad Hoc Reviewer
ISR, MISQ, EJIS, ECIS, ICIS, HICSS, CAIS, WISP, and The Dewald Roode Workshop
on Information Systems Security Research, IFIP WG8.11/WG11.13.
University Service
Fall 2017 to Present
Director of Security Analytics track within the Masters of Business Analytics degree at Leeds – University of Colorado Boulder
- Developed a proposal for a track with a pioneering collaboration between the school of business and school of engineering at CU Boulder
- Supervise graduate students who enrolled in the track across their three semesters in the program.
- Point-person from Leeds meeting with directors of the interdisciplinary telecom program from the school of engineering.
- Ambassador for the track; In charge of reach-out to and bond-forming with industry partners, including Webroot, IBM, and Crowdstrike.
Honors and Awards
Fellowships
- NSF Graduate Research Fellow (DGE-1247842) - $132,000
- GAR Foundation Fellow - $24,000
Grants
- David Berg Center for Ethics and Leadership at the University of Pittsburgh -$8,420
- Brigham Young University Marriott School of Business - $1,000
- Rollins Center for Entrepreneurship & Technology (2013) - $4,000
Scholarships
- Brigham Young University Undergraduate Scholarship - $8,278
- BYU Masters of Information Systems Management Scholarship - $4,380
- Robert K. Thomas BYU Honors Department Scholarship - $1,488
- Khona Family BYU Honors Department Scholarship - $1,488
- Ella M. Herman Scholarship - $1,054
Competitions
- AIS Global Competition 2012, Windows Phone Development Track, 2nd Place Worldwide
Areas of Expertise
- Statistics
- Machine learning
- Predictive Analytics, Business Intelligence
- Data visualization
- Databases
- Cloud Computing
- DevOps
- Server Administration
- Web development
Technical Skills
- R (tidyverse, dplyr, ggplot2), SAS, SPSS, python (pandas)
- MySQL / postregsql, MS Access, S3/Glue/Athena, MongoDB
- Python (scikit-learn, scipy, numpy), Tensorflow (a little), R (keras), RapidMiner, Alteryx
- vagrant, packer, terraform, git, chef, ansible, pytest
- AWS (EC2, Glue, Athena, S3, IAM, SageMaker)
- GCP (Compute, Cloud Storage, and whatever they call their AutoML offering now)
- html/css/less/sass/scss, bootstrap (tbs), bower, npm & grunt, pip, Jekyll,
- javascript/jquery
- Bash
- Vim
- Python
- Docker
- Powershell
- Java
- php (slim, CakePHP, propel)
- markdown, twig/Jinja2
- Nginx, apache, gunicorn
- Linux (Debian preferred), Windows server (active directory and group policies)
- VBA for Excel
Area-of-expertise Narratives
The below narratives illustrate how I have used the above technologies within the stated areas of expertise.
Check out my github activity for more! Not all of my activity is “open,” though – current research projects are typically marked as “private”
Habituation and Generalization Studies
This example describes what I call Full “Academic Stack” Development that I have done for one of the research projects in my cybersecurity stream. It includes the creation of web-based field experiments, data collection, analysis, and hypothesis testing.
- I used
npm
andyarn
for javascript package management, andgit
/github
for source code control. - Each “modal” (popup) in this study shares common code, so I used
webpack
to let me write javascript classes with inheritance.webpack
andgrunt
scripts also build and bundle the source javascript andscss
files that are used by the modals. - The front-end uses
jquery
/d3
/underscore
for interactivity. - The web task dynamically loads page elements in order to keep responsivity
high. I used python
Flask
for serving the task and for running a json api that the client-side javascript calls. - User data is stored in a
mysql
/postgres
database (depending on the server host). The database is built automatically and interfaced with by python code usingsqlalchemy
. - The task integrates a Qualtrics survey. Data is passed to the survey via web url parameters, so each participant’s identifier gets recorded into the Qualtrics data for matching later.
- The task is deployed to Heroku, Salesforce’s platform-as-a-service solution. This runs containerized applications on AWS hardware.
- I wrote “campaign” functionality to automate staggered posting of tasks to AWS
Mechanical Turk, using the python
boto3
library integrated with thepsiturk
library. - At analysis time, I queried data using
sqlalchemy
. After parsing json datastrings, I usepandas
DataFrames to merge together task behavior, mouse-tracking data, and survey data tables. I save these to csv files. - I did statistical analysis primarily in
R
, using thetidyverse
collection of packages (dplyr
,ggplot2
, etc.). I created models usingglm
andlm
, and I specified follow-up contrasts usingemmeans
. I usedknitr
to render a report from an r-markdown file.
Studies on confirmation bias for readers of online news
This example demonstrates consuming cloud-computing services – specifically, stringing together several AWS cloud computing services in order to capture mouse-tracking data for a web-based field experiment.
- I manually downloaded snapshots of dozens of news articles from various online
publication sources. Then I wrote python classes for each publication source
to scrape the articles’ content, headlines, and header image. This data was
stored in a
mongodb
. - I created an api that could display article content as if from any source,
using
Flask
templates. - I collected mouse-tracking data while participants read news articles, and
posted it to
AWS firehose
via anAWS api gateway
. Firehose stored data ins3
. This was crawled withAWS glue
, to create query-ableAWS Athena
tables. - I used Python scripts to query the Athena tables, and to group the
mouse-tracking data by article-interaction, based on timestamps recorded in
task-tracking data. I used
R
scripts to automate statistical analysis, usinglme4
for linear mixed models, since the data was of repeated-measures nature.
Topological Data Analysis for a Nomological Network of IS Survey Items
This example demonstrates some of my experience with using machine learning packages for research.
- I computed semantic space embeddings for IS survey items, using embedding
libraries such as
word2vec
andGloVe
. - I used
sklearn
to feed the data through various modeling algorithms. I also fit models using DataRobot’s AutoML platform. - I wrote a script to use python to interact with a Java trained ML model downloaded from DataRobot
- Then, I used
numpy
andscipy
to pull together model predictions from DataRobot - I applied various dimensionality reduction techniques (projections) to get
from 11-thousand down to 3- or 2-dimensions, using various
sklearn
manifolds, including IsolationForest,t-SNE
, l2-norm, and PCA. - I used
scipy.sparse.csgraph.shortest_path
to calculate shortest walking distances between all item-pairs from the graph, which I prepared usingscipy.spatial.distance.pdist
. - I used the Mapper algorithm (via python
KeplerMapper
library) to divide the projection into a grid of hyperspheres. Then I used agglomerative clustering (viasklearn.cluster.AgglomerativeClustering
) on each cluster to obtain nodes. I linked overlapping nodes, creating a “graph.” - I became a core developer of the scikit-tda python KeplerMapper library.
- I used KeplerMapper to visualize the graph using javascript
d3-force
graph.
InfoSec DevOps
This example demonstrates some of my self-taught efforts with cloud-based DevOps. It involved moving virtual machines from usb-based “golden image” virtualbox ISOs to full code-as-infrastructure DevOps.
- I used
packer
to automate the preparation of a raw image of Debian-based Kali OS to be able to run on Google Cloud (GCP – it’s cheaper for students than AWS). - I used
virt-manager
on a gcp-launched kali image (that’s nested virtualization!) so students can launch several “penetration testing lab” virtual machines for class assignments. - I used
vagrant
andchef
to provision the virtual machine images. The use of Vagrant means that I can update the vagrant build script and have students download and rerun the script to get fixed VM images mid-semester, if needed. I used chef because that is what metasploitable3 uses, and I wanted to be able to contribute back to metasploitable3. - I used
terraform
to launch a separate subnet, vpn server, and vulnerable midterm assessment server for each student team. The vulnerable servers are only accessible via a connection to the given vpn server, except that ssh is open to the world so that I can ssh in and fix images if needed, viaansible
.
Professional Affiliations
- IFIP Working Group 8.11/11.13, “Information Systems Security Research” Active Member
- Association for Information Systems Member
Foreign Language
- Spanish