GAUFS

Usage

To execute the code:

1. Place your datasets as CSV files in the datasets/ folder
2. Ensure your CSV files follow this format:
   • Column headers: var-0,var-1,var-2,var-3,...,var-n,ETIQ
   • Values should be normalized between [0,1]
   • ETIQ is the target column for classification
3. Run the main script:

   python main.py

After execution, results will be generated in the results/ folder. The script will automatically process all CSV files in the datasets/ folder and generate analysis results including clustering metrics and visualizations.

Project Structure

• datasets/ - Input CSV files for analysis
• results/ - Generated output files and visualizations
• src/ - Source code modules
  • build_test_genetic_2.py - Main test builder
  • genetic2_parallel.py - Parallel genetic algorithm implementation
  • alg_clustering.py - Clustering algorithms and metrics
  • analysis_weighted_variables_num_cluster.py - Variable analysis tools
  • data_generators - Methods related to synthetic data generation. The file can be executed to generate an example of a corners dataset on ./datasets/synthetic_data_corners.csv
• main.py - Main execution script for GAUFS algorithm/
• comparison/ — Files used for comparison with the AutoUFS tool.
  • dataset-papers/ — Input CSV files for the comparison.
  • results-papers/ — Output results, containing one folder for each compared dataset.
  • AutoUFSTool-main/ — Folder cloned from the AutoUFS-tool GitHub repository.
    • main-comparison.m — MATLAB script to run the comparison.
  • alg_clustering.py — Python module with utility functions for clustering.
  • automate_v2.py — Python script to run the automatic comparison process.
  • datasets_mat.ipynb — Jupyter notebook with tools for converting CSV files into MATLAB structures.