Function compute_relevance

src/licensedcode/models.py:2574–2626

Return a computed ``relevance`` given a ``length`` and a threshold. The relevance is a integer between 0 and 100 where 100 means highly relevant and 0 means not relevant at all. The relevance is computed base on the rule or detection ``length`` using a relevance schedule based

(length)

def compute_relevance(length):
    """
    Return a computed ``relevance`` given a ``length`` and a threshold.
    The relevance is a integer between 0 and 100 where 100 means highly
    relevant and 0 means not relevant at all.

    The relevance is computed base on the rule or detection ``length`` using
    a relevance schedule based on the ``length``

    For instance a match to the "gpl" or the "cpol" words have a fairly low
    relevance as they are a weak indication of an actual license and could
    be a false positive and should therefore be assigned a low relevance. In
    contrast a match to most or all of the apache-2.0 license text is highly
    relevant. The relevance is used as the basis to compute a LicenseMatch
    and LicenseDetection score.

    For example::
    >>> compute_relevance(0)
    0
    >>> compute_relevance(1)
    5
    >>> compute_relevance(17)
    94
    >>> compute_relevance(18)
    100
    >>> compute_relevance(19)
    100
    >>> compute_relevance(100)
    100
    """
    if length > 18:
        return 100
    return {
        0: 0,
        1: 5,
        2: 11,
        3: 16,
        4: 22,
        5: 27,
        6: 33,
        7: 38,
        8: 44,
        9: 50,
        10: 55,
        11: 61,
        12: 66,
        13: 72,
        14: 77,
        15: 83,
        16: 88,
        17: 94,
        18: 100,
    }[length]


def compute_thresholds_occurences(