- Introduction / About Me
- Featured Project — Healthcare Readmission Analytics
- Technical Toolkit
- Education and Credential
- Other Projects and Learning
- Connect With Me
I'm a Data Science undergraduate at the University of Advancing Technology, building practical projects in data analysis, business intelligence, and applied machine learning. My work emphasizes Python, SQL, PostgreSQL, Power BI, Streamlit, and careful model evaluation.
I bring experience in healthcare operations, warehouse and inventory processes, digital analytics, freelance consulting, and paid AI model-training assignments. These perspectives inform the real-world questions I explore through data.
I'm interested in Data Analyst, Business Intelligence Analyst, Healthcare Data Analyst, and junior Data Scientist opportunities. I value reproducible workflows, data quality, clear communication, and responsible interpretation of results.
Python · SQL · PostgreSQL · Power BI · Streamlit · pandas · scikit-learn
An end-to-end, retrospective healthcare analytics portfolio project investigating 30-day hospital readmissions using the UCI Diabetes 130-US Hospitals (1999–2008) dataset.
What I built
- Analyzed 101,766 historical hospital encounters and profiled an 86,547-encounter development cohort with an observed 11.18% 30-day readmission rate.
- Created and reconciled nine PostgreSQL reporting views for KPIs, demographics, admission patterns, utilization, and data completeness.
- Built a Power BI executive dashboard and a five-page Streamlit application for interactive research reporting.
- Evaluated Logistic Regression, Random Forest, and Histogram Gradient Boosting using patient-aware methods and documented model limitations.
Evaluation transparency: The Random Forest's reported ROC-AUC of 0.6494 comes from a patient-disjoint held-out test partition that had previously been used for baseline evaluation; it is not a strictly untouched, single-use test estimate or external validation. Histogram Gradient Boosting results are based on patient-grouped out-of-fold development validation. This is a research and portfolio demonstration, not a clinical decision-support system.
Explore: GitHub repository · Live interactive dashboard · Power BI preview
| Focus | Technologies and methods |
|---|---|
| Analysis and programming | Python, SQL, pandas, NumPy, data cleaning, exploratory analysis, statistical analysis |
| Databases and business intelligence | PostgreSQL, Microsoft Power BI, Microsoft Excel, data validation, KPI reporting |
| Machine learning | scikit-learn, classification, feature engineering, patient-aware validation, model evaluation |
| Development and communication | Streamlit, Jupyter Notebook, Git/GitHub, technical documentation, dashboards |
| Additional hands-on project exposure | TensorFlow, PyTorch, MATLAB, AWS, Terraform |
University of Advancing Technology — Bachelor of Science in Data Science
Expected graduation: December 20, 2026 · GPA: 3.89
Microsoft Applied Skills: Create an intelligent document processing solution with Azure AI Document Intelligence (earned October 2024).
- Data Science Projects — earlier learning and project archive
- Additional documented coursework and technical experiments are available through my GitHub repositories.
Earlier profile visuals and programming journey
Original data-science profile banner
Original data-science process animation
- Email: info@amoheric.com
- Website / Portfolio: amoheric.com
- LinkedIn: linkedin.com/in/amoheric
- Tableau Public: public.tableau.com/app/profile/amoheric
- GitHub: github.com/amoheric
Developer community profiles, badges, and GitHub activity
Building useful analytics solutions with transparency, reproducibility, and an emphasis on real-world decisions.