How jotis86/people-analytics-project works
Project Type and Purpose
• The system combines interactive data analysis and visualization using Power BI with data cleaning, processing, and additional analysis performed in Python. There is also an interactive web app powered by Streamlit.
• This is a full-stack analytics project that integrates multiple tools – Power BI for interactive dashboards, Python (with Pandas, Matplotlib, and Seaborn) for data cleaning and visualizations, and Streamlit for dynamic web presentation.
Break Down the File Structure to Identify Key Components
• The “app/” directory contains the web app (Streamlit) that serves results to end users.
• The “assets/” folder holds related images and branding assets.
• The “data/” folder contains raw and processed data (e.g., raw CSVs or Excel files).
• The “notebook/” directory contains Jupyter notebooks used for data cleaning and visualization with Python.
• The “powerbi/” directory contains Power BI project files (dashboard.pbix, related media such as clip.mp4) to create interactive dashboards.
• The “summary_report/” folder holds the summary insights and supporting documentation (its own README and images).
• Other files like “requirements.txt”, “README.md”, and “LICENSE” provide project documentation and dependency management.
Main Components and Their Relationships
a. Data Sources:
– Raw data is stored in the “data” and “notebook” directories (data.xlsx, data_cleaned.xlsx, csv files).
b. Data Processing and Transformation:
– Jupyter notebooks in “notebook/” perform cleaning and visualization tasks using Python (Pandas for cleaning; Matplotlib/Seaborn for plotting).
c. Reporting and Visualization:
d. Web Application:
– Streamlit app in “app/” provides an interactive front-end that displays the processed and visualized data to end users.
e. Supporting & Presentation Layers:
– “assets/” hold related images.
– “summary_report/” provides conclusive insights and a report.
Architectural Patterns and Design Principles
• Separation of Concerns: Different directories clearly separate data ingestion and cleansing (notebooks, data), visualization (Power BI, notebooks), and presentation (Streamlit web app).
• ETL Flow: Data goes through extraction (data files), transformation (cleaning notebooks and Python scripts), and loading (streamlit app and Power BI dashboards).
• Integration Points: Although each analysis tool (Python, Power BI, and Streamlit) works with preprocessed data, they are integrated into the overall People Analytics platform.
• Modularity: Each component (data processing, visualization, web presentation) is decoupled for easier maintenance and future scalability.