Normalized compression distance in python. Inspired by https://aclanthology.org/2023.findings-acl.426.pdf

normalizedcompressiondistance.py

import gzip

x1 = "Japan’s Seiko Epson Corp. has developed a 12-gram flying microrobot."
x2 = "The latest tiny flying robot has been unveiled in Japan."
x3 = "Michael Phelps won the gold medal in the 400 individual medley."

def ncd(x: str, y: str) -> float:
    '''
    Returns the normalized compression distance between two strings.

            Parameters:
                    x (str): A text string
                    y (str): Another text string

            Returns:
                    ncp (float): normalized compression distance (number between 0 and 1)
    '''
    cx = len(gzip.compress(x.encode()))
    cy = len(gzip.compress(y.encode()))
    xy = " ".join([x, y])
    cxy = len(gzip.compress(xy.encode()))
    return (cxy - min(cx, cy)) / max(cx, cy)


# x3 is further from x1 than x2
ncd(x1, x2)
ncd(x1, x3)

# ncd is symmetric
ncd(x1, x2)
ncd(x2, x1)

# ncd does not go to exactly 0
ncd(x1, x1)
ncd(x3, x3)