bytewise

func DamerauLevenshteinDistance ¶

func DamerauLevenshteinDistance(a, b string) (int, error)

DamerauLevenshteinDistance calculates the magnitude of difference between two strings using the Damerau- Levenshtein algorithm with adjacent-only transpositions, bytewise.

This edit distance is the minimum number of single-byte edits (insertions, deletions, substitutions, or transpositions) to transform one string into the other. DamerauLevenshtein differs from Levenshtein primarily in that DamerauLevenshtein considers adjacent-byte transpositions.

The larger the result, the more different the strings.

See: http://en.wikipedia.org/wiki/Damerau-Levenshtein_distance

func DiceCoefficient ¶

func DiceCoefficient(a, b string) (float64, error)

DiceCoefficent calculates the simiarlity of two strings per the Sorensen-Dice coefficient, bytewise.

The resulting value is scaled between 0 and 1.0, and a higher value means a higher similarity.

This algorithm is also known as the Sorensen Index, and is very close to the White Similarity metric, with the key distinctions that DiceCoefficient does not differentiate between whitespace and other characters and also does not account for bigram frequency count differences between the compared strings.

See: http://en.wikipedia.org/wiki/Sorensen-Dice_coefficient

Note that this algorithm implementation operates upon individual bytes and does not account for multibyte unicode runes.

Returns an error if both of the input strings contain less than two bytes.

func HammingDistance ¶

func HammingDistance(a, b string) (uint, error)

See: http://en.wikipedia.org/wiki/Hamming_distance

Returns an error if the string lengths are not equal.

Note that this algorithm implementation operates upon individual bytes, and does not account for multibyte unicode runes.

func LevenshteinDistance ¶

func LevenshteinDistance(a, b string) (int, error)

LevenshteinDistance calculates the magnitude of difference between two strings using the Levenshtein Distance metric, bytewise.

This edit distance is the minimum number of single-byte edits (insertions, deletions, or substitutions) needed to transform one string into another.

The larger the result, the more different the strings.

See: http://en.wikipedia.org/wiki/Levenshtein_distance

func WhiteSimilarity ¶

func WhiteSimilarity(a, b string) (float64, error)

WhiteSimilarity calculates the similarity of two strings through a variation on the Sorensen-Dice Coefficient algorithm, bytewise.

The resulting value is scaled between 0 and 1.0, and a higher value means a higher similarity.

WhiteSimilarity differs from DiceCoefficient in that it disregards bigrams that include (single-byte) whitespace, applies an upper-case filter, and accounts for bigram frequency.

See: http://www.catalysoft.com/articles/strikeamatch.html

Note that this algorithm implementation operates upon individual bytes and does not account for multibyte unicode runes.

Returns an error if neither of the input strings contains at least one byte bigram without whitespace.