Skip to content

thatsabhishek/Whirling_Triangle_based_classification

Repository files navigation

Whirling Triangle Feature Extraction Method

Overview

The Whirling Triangle Feature Extraction Method is a geometric–computational framework that performs rotational triangular partitioning and extracts structured features from each generated region.

This method is designed to model spatial complexity using whirling (rotating) triangle formations and convert geometric structures into machine-learning-ready datasets.

The project integrates computational geometry, spatial analytics, and feature engineering into a reproducible research pipeline.


Key Concepts

  • Whirling Triangle Decomposition
  • Rotational Spatial Partitioning
  • Recursive Triangle Generation
  • Geometric Feature Engineering
  • Synthetic Dataset Creation

Repository Structure

.
├── Whirling_Triangle_Feature_Extraction_Method.ipynb
├── Whirling_Triangle_Features.csv
├── LICENSE.txt
└── README.md

Notebook Description

🔹 Whirling_Triangle_Feature_Extraction_Method.ipynb

This notebook implements the complete feature extraction workflow:

  • Generates whirling triangular partitions
  • Applies rotational geometric transformations
  • Computes triangle region boundaries
  • Labels each triangle uniquely
  • Extracts geometric and statistical features
  • Exports structured data into CSV format

The implementation ensures reproducibility and stable feature generation.


Dataset Description

File: Whirling_Triangle_Features.csv

Dataset Characteristics

  • Region-wise feature representation
  • Stable feature values
  • Machine Learning ready format
  • Suitable for classification & clustering tasks

Each row corresponds to one triangular region generated during the whirling decomposition process.


Extracted Feature Categories

🔸 Geometric Features

  • Triangle Area
  • Perimeter
  • Edge Lengths
  • Aspect Ratio
  • Compactness

🔸 Positional Features

  • Centroid Coordinates
  • Vertex Positions
  • Orientation Angles
  • Rotation Index

🔸 Statistical Features

  • Mean
  • Variance
  • Standard Deviation
  • Skewness
  • Kurtosis

🔸 Structural Features

  • Triangle Level / Depth
  • Region Index
  • Spatial Hierarchy Position

These features encode both shape geometry and rotational spatial behavior.


Requirements

Install dependencies:

pip install numpy pandas matplotlib scipy

How to Run

  1. Clone the repository:
git clone https://github.com/your-username/whirling-triangle-feature-extraction.git
cd whirling-triangle-feature-extraction
  1. Launch Jupyter Notebook:
jupyter notebook
  1. Open:
Whirling_Triangle_Feature_Extraction_Method.ipynb
  1. Run all cells to:
  • Generate whirling triangles
  • Extract features
  • Export CSV dataset

Applications

  • Computational Geometry Research
  • Pattern Recognition
  • Spatial Feature Engineering
  • Synthetic Dataset Generation
  • Machine Learning Modeling
  • Academic Research & Publications

License

This project is licensed under the GNU General Public License v3.0.

Releases

No releases published

Packages

 
 
 

Contributors