Regarding Sorensen–Dice Coefficient

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

def _string_shift_chunks(
        string: str,
        chunk_size: int = 4,
        shift_size: int = 1,
    ) -> list[str]:
    """Return sub-samples of a string.

    Args:
        chunk_size: size of sub-sampling of `x` and `y`
        shift_size: size of sub-sampling shift of `x` and `y`
    """
    return [
        string[i:i+chunk_size]
        for i in range(0, len(string), shift_size)
    ]

def sdc_coefficient(
        x: str,
        y: str,
        chunk_size: int = 4,
        shift_size: int = 1,
    ) -> float:
    """Sorensen–Dice coefficient implementation.

    Args:
        x: sample string
        y: sample string
        chunk_size: size of sub-sampling of `x` and `y`
        shift_size: size of sub-sampling shift of `x` and `y`

    Returns:
        float: similarity score between [0, 1]
    """
    _x = _string_shift_chunks(
        string=x,
        chunk_size=chunk_size,
        shift_size=shift_size,
    )
    _y = _string_shift_chunks(
        string=y,
        chunk_size=chunk_size,
        shift_size=shift_size,
    )
    _xy: str = f"{_x}{_y}"
    return (2 * len(set(_xy))) / len(_xy)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

from difflib import SequenceMatcher
from strsimpy.jaro_winkler import JaroWinkler
from strsimpy.levenshtein import Levenshtein
from strsimpy.metric_lcs import MetricLCS
from strsimpy.ngram import NGram

test_cases = [
    ("prod.us-east01.verizon.com", "prod.us-east02.verizon.com"),
    ("apple", "apple"),
    ("apple", "apples"),
    ("apple", "orange"),
]


if __name__ == "__main__":
    for (x, y) in test_cases:
        print("---")
        print(sdc(x, y))
        print(JaroWinkler().distance(x, y))
        print(Levenshtein().distance(x, y))
        print(MetricLCS().distance(x, y))
        print(NGram(2).distance(x, y))
        print(NGram(4).distance(x, y))
        print(SequenceMatcher(x, y).ratio())