Company Name Matching Tool

A Python-based CLI tool that matches company names from an input CSV file or folder against a master CSV file or folder.

Features

• Streamed CSV processing for low memory usage
• Fast fuzzy matching using rapidfuzz
• Support for processing a single input file or all CSV files in an input folder
• Returns up to 5 top matches with confidence scores for each input company, sorted by similarity (descending)
• Real-time progress tracking with estimated time remaining (ETA)
• Docker-ready execution to avoid polluting local Python environments
• Adds matched_company, confidence, and up to 4 additional match columns to output CSV files

Installation

pip install -r requirements.txt

Usage

Prepare an input file or folder and a master file or folder:

• input/ containing one or more .csv files
• master/ containing one or more .csv files

Run locally:

python match_companies.py --input input --master master --output output --threshold 80

If you prefer a single file:

python match_companies.py --input input/sample_input.csv --master master/master_list.csv --output output --threshold 80

Docker

Build the image:

docker build -t company-matcher .

Run the container:

docker run --rm -v "%cd%/input:/app/input" -v "%cd%/master:/app/master" -v "%cd%/output:/app/output" company-matcher --input input --master master --output output --threshold 80

Output

The tool generates output CSV files with the following additional columns:

• matched_company - The top matching company name from the master list
• confidence - Confidence score (0-100) for the top match
• matched_company_2 to matched_company_5 - 2nd through 5th best matches
• confidence_2 to confidence_5 - Confidence scores for matches 2-5

All matches are sorted by confidence score in descending order and only included if they meet the threshold score.

Notes

• threshold is the minimum accepted similarity score.
• If the input or master path is a folder, all .csv files in that folder are processed.
• The output folder will be created automatically if it does not exist.