Student-Behavior-Analytics

A comprehensive data visualization project demonstrating various charting techniques using Matplotlib, Seaborn, and Plotly. This project analyzes educational platform user behavior data through multiple visualization approaches, from basic line charts to interactive animations.

What This Project Does

This project processes student behavior data from an educational platform database to create meaningful visualizations. Each exercise builds on the previous one, exploring different aspects of the data through various chart types and styling approaches.

File Descriptions

00_line_chart.ipynb

    Basic line chart showing daily page views using Pandas plotting functionality

01_line_chart_styles.ipynb

    Dual-line chart comparing page views and commits with custom styling

02_bar_chart.ipynb

    Bar chart analyzing commit patterns across different times of day

03_bar_charts.ipynb

    Comparative bar charts showing weekday vs weekend commit patterns

04_histogram.ipynb

    Overlapping histograms comparing commit distributions using Matplotlib

05_boxplot.ipynb

    Box plot analysis of control vs test group behavior with custom styling

06_scatter_matrix.ipynb

    Scatter matrix visualization showing correlations between multiple variables

07_heatmap.ipynb

    Heatmap visualizations for temporal commit patterns across users

08_seaborn.ipynb

    Advanced styling with Seaborn for project commit dynamics

09_plotly.ipynb

    Interactive animated line race chart using Plotly

Interesting Techniques Used

Data Manipulation Techniques

• SQL queries with SQLite3 for database operations and filtering non-admin users
• Pandas DataFrame operations including grouping, aggregation, and time-based filtering
• Time-based data categorization using datetime operations to classify commits by time periods
• Multi-index operations for complex data grouping and analysis

Visualization Techniques

• Custom color palettes and transparency effects using matplotlib alpha parameter
• Figure sizing and font customization with matplotlib's rcParams configuration
• Subplot positioning and layout management using matplotlib's figure and axes system
• KDE (Kernel Density Estimation) plots for distribution visualization
• Scatter matrix generation using pandas.plotting.scatter_matrix
• Heatmap creation with custom colormaps and axis labeling
• Animation frameworks using Plotly's graph_objects for interactive charts

Technologies and Libraries

Core Data Processing

• Pandas - Advanced data manipulation and analysis
• SQLite3 - Database connectivity and querying
• NumPy - Numerical computing for data processing

Visualization Libraries

• Matplotlib - Foundational plotting library with extensive customization options
• Seaborn - Statistical data visualization with built-in themes and color palettes
• Plotly - Interactive plotting library for web-based visualizations

Advanced Matplotlib Components

• mpl_toolkits.axes_grid1 - Advanced subplot positioning and colorbar management
• Custom transparency handling for overlapping visualizations

Project Structure

├── src/
│   ├── ex00/
│   ├── ex01/
│   ├── ex02/
│   ├── ex03/
│   ├── ex04/
│   ├── ex05/
│   ├── ex06/
│   ├── ex07/
│   ├── ex08/
│   └── ex09/
├── data/
└── misc/
    └── images/

src/: Contains all Jupyter notebooks organized by exercise number, each demonstrating different visualization techniques and data analysis approaches
data/: Stores the SQLite database and CSV files used for analysis. The main database contains pageviews and checker tables with student activity data
misc/images/: Reference images showing expected output for each visualization exercise

Key Features

The project demonstrates progression from basic static charts to advanced interactive visualizations. Each exercise explores different aspects of the same dataset, showing how visualization choice affects data interpretation. The final Plotly exercise creates an animated line race chart that dynamically shows commit progress over time.

The visualizations reveal patterns like peak usage hours, weekend vs weekday behavior differences, and user engagement metrics that would be difficult to spot in raw data tables.