Data Preparation
SQL querying, data cleaning, validation checks, feature engineering, reproducible workflows
About Me
Greetings! I’m Amir Amrollahi, a Data Analyst with an MSc in Management Analytics from Queen’s University.
💻 Skilled in SQL, Python, Machine Learning, Power BI, and Excel for data analysis, workflow automation, predictive modeling, and dashboard development.
📊 2+ years of experience across healthcare, operations, and customer analytics, turning raw data into insights teams can actually use.
🤝 Collaborative team member who communicates clearly, supports stakeholders, and helps turn analysis into practical decisions.
Skills
BI & Reporting
Power BI dashboards, DAX, Power Query, KPI design, executive-ready reporting
Operations Analytics
KPI tracking, performance measurement, scheduling analytics, operational decision support
Predictive Modeling
Logistic regression, decision trees, clustering, performance comparison, insight generation
Featured Work
Healthcare Operations
Oncology Consultation Scheduling Optimization
Designed a decision-support workflow to improve referral intake, validation, and appointment assignment for oncology consultation scheduling.
72% → 89%
On-Time Scheduling Improvement
Tools
Python, Streamlit, SQL, Excel, Optimization Logic
Improved 14-day scheduling visibility
Reduced manual tracking effort by 44 man-hours per week
Created referral and appointment monitoring logic
▶ Demo Video
Watch Demo Video →Operations Analytics
Remote Patient Monitoring Dashboard & Readmission Risk
Built a dashboard project that combines remote patient monitoring indicators with readmission-risk analysis.
66% → 77%
Readmission Risk Prediction Accuracy
Tools
Power BI, DAX, Power Query, Python, Logistic Regression, Decision Tree
Integrated demographics, readings, alerts, and notes for 702 patients
Modeled readmission risk with ML methods
Translated results into managerial insights and recommendations
Project Output
View Live Dashboard →Recommendation Systems
Course Recommendation System for Online Learning
Designed a course recommendation system that grouped students by engagement, education level, and previous credits, then recommended suitable courses within each cluster.
78%
Accuracy of Recommendations
Tools
Recommendation System, K-Means Clustering, Collaborative Filtering, Python
Clustered approximately 22,000 students into 5 learner groups
Used collaborative filtering to recommend courses within each cluster
Evaluated recommendations with accuracy, precision, and recall
Project Output
View System →WORK EXPERIENCE
Research Assistant|Kingston Health Sciences Centre|Sep 2025 — Mar 2026
Research Assistant|Queen’s University|Jan 2025 — Aug 2025
Marketing & Customer Behavior Analyst|Amrollahi Mosaic Manufacturing Company|Oct 2023 — Oct 2024