Zahra Abbas
Data Analyst | Python, SQL, Dashboards & Visualization
New to Contra
Zahra is building their profile!
Health Insurance Claims EDA & Predictive Analysis
Built an end-to-end healthcare data analytics project focused on insurance claim patterns, patient risk factors, and cost prediction using Python and machine learning techniques.
This project involved data cleaning, exploratory data analysis (EDA), visualization, statistical insights, and predictive modeling to understand how factors like age, BMI, smoking habits, and medical conditions impact insurance charges.
What I worked on:
• Data preprocessing and cleaning
• Exploratory Data Analysis (EDA)
• Correlation and trend analysis
• Feature engineering
• Predictive ML models for insurance cost estimation
• Visual dashboards and statistical insights using matplotlib/seaborn
Tools & Technologies:
Python • Pandas • NumPy • Matplotlib • Seaborn • Scikit-learn • Jupyter Notebook
1
14
Developed an interactive sales analysis dashboard to identify customer purchasing behavior and business performance trends using Excel-based analytics and visualization techniques.
The project involved cleaning and organizing raw transactional sales data, performing exploratory analysis, and building an interactive dashboard to highlight insights related to customer demographics, regional sales patterns, and purchasing trends.
Key tasks included:
• Data cleaning and preprocessing
• Pivot table analysis
• Dashboard creation and visualization
• Trend and sales pattern analysis
• Interactive filtering and reporting
Tools & Technologies:
Microsoft Excel, Pivot Tables, Charts, Dashboards
The final dashboard was designed to present business insights in a visually clear and user-friendly format, enabling quick analysis of sales performance and customer behavior.
1
15
Developed a machine learning-based Heart Disease Prediction System with an interactive Streamlit interface for real-time prediction and analysis. The project involved data cleaning, exploratory data analysis, feature engineering, model training, and evaluation using Python and machine learning libraries.
The goal was to create a simple and user-friendly system that could help analyze patient health indicators and predict possible heart disease risk. Alongside the predictive model, the project also focused heavily on data visualization and interpretability.
Key tasks included:
• Data preprocessing and cleaning
• Exploratory Data Analysis (EDA)
• Correlation analysis and visualization
• Machine learning model building and evaluation
• Interactive Streamlit web app development
• Report writing and presentation preparation
Tools & Technologies:
Python, pandas, matplotlib, scikit-learn, Streamlit
This project was collaboratively developed with my teammate Gaurina as part of a university machine learning project.
1
34
Developed a large-scale statistical and data analysis project focused on air quality trends across major cities in Pakistan using Python and advanced analytical techniques.
The project involved analyzing 21,000+ environmental records to identify pollution patterns, seasonal variations, and relationships between different pollutants. Multiple statistical methods and forecasting models were applied to better understand air quality behavior and generate meaningful insights from real-world environmental data.
Key tasks included:
• Data cleaning and preprocessing
• Exploratory Data Analysis (EDA)
• Correlation analysis and visualization
• Hypothesis testing and ANOVA
• Regression modeling
• ARIMA forecasting and trend analysis
• Statistical reporting and visualization
Tools & Technologies:
Python, pandas, matplotlib, seaborn, statsmodels, scikit-learn
The project also included publication-style visualizations, analytical reporting, and interpretation of statistical findings to make the results more understandable and actionable.
1
22