A CLI (Command Line Interface)-based data analysis application built using Python to analyze stroke patient datasets. This project includes data cleaning, statistical analysis, algorithmic search, and interactive data visualization.
- Automatic Data Cleaning: Handles missing values (Mean/Mode imputation) and removes duplicate data.
- Interactive CLI: Interactive menu interface with easy navigation (Looping menu).
- Data Visualization:
- Pie Chart (Stroke Proportion)
- Heatmap (Feature Correlation)
- Histogram & KDE (Age Distribution)
- Scatter Plot (BMI vs. Glucose Relationship)
- Searching Algorithm: Searches for patients based on specific criteria (Age & Stroke Status).
- Sorting Algorithm: Sorts data by Age, Glucose, or BMI.
- Export Data: Saves the cleaned data to a new CSV file.
- Cross-Platform: Supports a clear screen on both Windows and Linux/Mac.
- Language: Python 3.x
- Libraries:
pandas(Data Manipulation)matplotlib&seaborn(Data Visualization)os&sys(File System Management)
- Clone this repository:
git clone [https://github.com/kelfinofarelino/stroke-data-analysis.git](https://github.com/kelfinofarelino/stroke-data-analysis.git)cd stroke-data-analysis - Install required libraries:
pip install -r requirements.txt - Run the application:
python main.py
-
v2.4 (Latest) - UX Improvements:
-
[UPDATE] Changed the Sorting menu input from string (type the column name) to numeric (select numbers 1-3) to minimize typos.
-
[FIX] Improved error handling if a dataset column name changes (e.g., Sex vs. Gender).
-
v2.3 - Clean Interface:
-
[NEW] Added an automatic clear_screen() feature. The terminal screen will clear every time you switch menus, creating a more professional appearance. typo.
-
[UPDATE] Added compatibility for Windows (cls) and Mac/Linux (clear).
-
v2.2 - Interaction Flow:
-
[NEW] Added a "Pause" feature (Press Enter to continue) after feature execution, giving the user time to read the analysis results before the main menu reappears.
-
[FIX] Improved automatic file path (File Path) using os.path.join to prevent FileNotFound errors.
-
v2.1 - Robustness:
-
[FIX] Improved Data Cleaning logic for separate handling of numeric and categorical data types.
-
[FIX] Backup original data (raw_df) before modifying it.
-
v2.0 - Major Refactor (OOP):
-
[NEW] Changed the procedural code structure to Object-Oriented Programming (OOP) with the StrokeAnalysisApp class.
-
[NEW] Added a main menu with a while loop to prevent the application from closing immediately.
-
[NEW] Added an Export to CSV feature.
-
v1.0 - Initial Release:
-
Initial release for Data Structures coursework.
-
Basic features: Load Data, Simple Stats, Simple Plotting.
stroke-data-analysis/
│
├── data/
│ ├── stroke_dataset.csv # Original Dataset
│ └── cleaned_stroke_data.csv # Cleaned Data Output (Generated)
│
├── main.py # Main Source Code
├── README.md # Project Documentation
└── requirements.txt # List of Dependencies
Author: Muhammad Farelino Kelfin Informatics Student, UPN "Veteran" Yogyakarta - 2024