| Name | Role | Phone | Bio |
|---|---|---|---|
| Khaled Alkurd | Team Lead | (703) 231-4491 | GMU Spring '26 Accounting and Business Analytics, Alteryx, Tableau, R |
| Pranavi Doodala | Project Manager | ------ | GMU Spring '26 Business Analytics; Data Mining, Predictive Modeling, Project Management |
| Mariam Debas | Visualization Lead | ------ | GMU Spring '26 Accounting, Data Analytics, Tableau, R |
| Nikita Chandrasing | Product Manager | ------ | GMU Fall '25 Business Analytics and MIS; Data Mining, Data Visualization, Product Design |
| Andy Yaro | Product Developer | ------ | GMU Fall '28 Cybersecurity Engineering; python, git, aws |
AuditsMadeSimple • AGA Datathon 2026
A public-facing educational platform helping citizens understand federal financial assistance and audit data.
🌐 Live Website: https://gmufiscalpatriots.bytechisel.com
📊 Presentation: View Presentation on Canva
| Category | Tools |
|---|---|
| Data Processing | Alteryx, Python |
| Visualization | Tableau Public |
| Machine Learning | scikit-learn |
| Website | HTML, CSS, AWS |
| AI Assistance | Claude, Google Gemini |
| Library | Purpose |
|---|---|
scikit-learn |
HistGradientBoostingClassifier, model evaluation metrics (ROC-AUC, PR-AUC), permutation importance |
pandas |
Data cleaning, feature engineering, entity-year aggregation |
matplotlib |
Feature importance visualization |
- $8.58 trillion in federal financial assistance distributed (FY2019–2024)
- Top 10 states received 50.9% of all federal grants
- California alone received $1.05 trillion (12% of total)
- HHS and USDA account for 67% of all federal assistance
- 57,448 entities required to submit Single Audits (FY2016–2024)
- 16,300+ entities had at least one audit finding
Federal funding data is large, fragmented, and hard to interpret quickly. Audit data is even harder because it uses specialized terminology and the impact is not always obvious. Our goal is to make these systems approachable for everyone.
This project helps users:
- Explore funding patterns by geography and time
- Learn audit terminology with plain-language definitions
- Follow guided examples (case studies and next steps) to investigate audit data more deeply
Dashboards that highlight:
- Funding distribution across states
- Funding trends over time
- Major funding sources by federal agency
- Higher-risk entities with funding context
A glossary translating common audit and spending terms into short, usable definitions.
Plain descriptions of USAspending, FAC, and SAM: what each tracks, what each misses, and how to use them responsibly.
Short walkthroughs and a checklist-style “What’s next?” page that keeps visitors engaged after their first chart.
We connect official public sources using UEI (Unique Entity Identifier) whenever possible.
| Source | What It Tracks |
|---|---|
| USAspending.gov | Federal spending and award transactions: who received funds, how much, and from which agency |
| FAC.gov | Single Audit submissions for non-federal entities expending $750K+ in federal awards |
| SAM.gov | Governmentwide exclusions: debarments, suspensions, and other exclusion actions |
deliverables/ Submission artifacts (dashboards, slides, report, video)
webapp/ Website source (HTML pages + assets)
data/ Clean outputs (CSV + Hyper) organized by domain
pipeline/ Alteryx workflows + pipeline notes
docs/ Data dictionaries + appendix hubs + team/competition docs
assets/ Repo visuals used in README/docs
These live in:
docs/appendix_hubs/methodology/screenshots/alteryx/
USAspending transformations
USAspending transaction-level data was transformed into recipient-level summaries for each fiscal year (2019–2024), then combined for cumulative analysis. Outputs were exported to CSV and Tableau Hyper formats for dashboard integration.
FAC Master Clean
Four FAC tables (General, Findings, Corrective Action Plans, Federal Awards) were joined and aggregated to entity level by auditee_uei, producing 57.4K clean audited entity records. Outputs preserve audit flags, finding counts, and federal expenditure amounts for risk scoring and USAspending integration.
Audit Health Score construction
The Audit Health Score was calculated for each entity using weighted risk factors including going concern (25 pts), material weakness (20 pts), repeat findings (15 pts), and significant deficiencies (10 pts), with mitigating factors applied. Entities were tiered into Red, Yellow, and Green categories to support dashboard visualizations and funding analysis.
FAC + USAspending Merge
FAC audit data was joined to USAspending financial assistance records on UEI to link audit findings with federal funding received. This merge enables analysis of how much taxpayer money flowed to entities with material weaknesses, repeat findings, or going concern flags.
SAM + FAC Merge
SAM exclusion data was joined to FAC audit records on UEI to identify entities with both exclusion history and audit findings. This cross-reference supports transparency gap analysis of debarred or suspended entities.
SAM Exclusion Cleaning
SAM exclusion records were split by UEI availability (38K with UEI, 120K legacy records without), with date parsing to calculate exclusion duration and active status. Outputs support transparency gap analysis and cross-referencing of excluded entities against federal award recipients.
These live in:
docs/appendix_hubs/methodology/screenshots/tableau/
Funding Distribution & Trends
These visualizations show where federal financial assistance flows geographically and how funding levels changed over the six-year period. The Top 10 states account for over half of all federal grants, with California alone receiving 12% of the national total.
Audit Oversight Signals
These maps highlight where audit findings are concentrated relative to funding received. Material weakness rates vary significantly by state, revealing geographic patterns in internal control quality.
Risk Tiering
This visualization shows the relationship between Audit Health Score and federal funding received. Red-tier entities received disproportionately more funding than Green-tier entities with cleaner audit records.
- Visit the live site: https://gmufiscalpatriots.bytechisel.com
- Open the Tableau dashboards (linked from the site)
- Read the glossary to understand audit terminology
- Use “What’s next?” to follow guided questions and case studies
| Category | Tools |
|---|---|
| Data Processing | Alteryx, Python |
| Visualization | Tableau Public |
| Machine Learning | scikit-learn |
| Website | HTML, CSS, AWS |
This project is for educational and public understanding purposes. Audit and exclusion signals require context. The presence of a finding or a higher-risk tier is not proof of wrongdoing. Always validate conclusions using primary documentation, program context, and appropriate investigative standards.
Fiscal Patriots • AuditsMadeSimple • AGA Datathon 2026
George Mason University • Costello College of Business