Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat: similarity and difference score #40

Merged
merged 2 commits into from
Jan 19, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
48 changes: 48 additions & 0 deletions crates/differ/src/hamming.rs
Original file line number Diff line number Diff line change
Expand Up @@ -79,4 +79,52 @@ mod tests {
assert_eq!(test_diff_4, hamming("0000", "1111"));
assert_eq!(test_diff_5, hamming("2173896", "2233796"));
}

#[test]
fn test_hamming_similarity() {
use crate::hamming::hamming;
use crate::DiffScoreConfig;

let sim = hamming("karolin", "kathrin");
let config = DiffScoreConfig::default();
let similarity = ((7.0) - 3.0) / (7.0);

let sim_v2 = hamming("karolin", "kerstin");
let mut config_v2 = DiffScoreConfig::default();
config_v2.sub_cost = 0.5;
let similarity_v2 = ((7.0) - 1.5) / (7.0);

let sim_v3 = hamming("kathrin", "kerstin");
let mut config_v3 = DiffScoreConfig::default();
config_v3.sub_cost = 1.5;
let similarity_v3 = ((7.0) - 6.0) / (7.0);

assert_eq!(similarity, sim.similarity(&config));
assert_eq!(similarity_v2, sim_v2.similarity(&config_v2));
assert_eq!(similarity_v3, sim_v3.similarity(&config_v3));
}

#[test]
fn test_hamming_difference() {
use crate::hamming::hamming;
use crate::DiffScoreConfig;

let diff = hamming("karolin", "kathrin");
let config = DiffScoreConfig::default();
let difference = ((7.0) - 3.0) / (7.0);

let diff_v2 = hamming("karolin", "kerstin");
let mut config_v2 = DiffScoreConfig::default();
config_v2.sub_cost = 0.5;
let difference_v2 = ((7.0) - 1.5) / (7.0);

let diff_v3 = hamming("kathrin", "kerstin");
let mut config_v3 = DiffScoreConfig::default();
config_v3.sub_cost = 1.5;
let difference_v3 = ((7.0) - 6.0) / (7.0);

assert_eq!(difference, diff.similarity(&config));
assert_eq!(difference_v2, diff_v2.similarity(&config_v2));
assert_eq!(difference_v3, diff_v3.similarity(&config_v3));
}
}
48 changes: 48 additions & 0 deletions crates/differ/src/levenshtein.rs
Original file line number Diff line number Diff line change
Expand Up @@ -166,4 +166,52 @@ mod tests {
assert_eq!(test_diff_3, levenshtein("RESET", "SETS"));
assert_eq!(test_diff_4, levenshtein("RESET", "RESETER"));
}

#[test]
fn test_levenshtein_similarity() {
use crate::levenshtein::levenshtein;
use crate::DiffScoreConfig;

let sim = levenshtein("Kittens", "kitten");
let mut config = DiffScoreConfig::default();
config.lowercase_sub_cost = 0.5;
let similarity = ((7.0) - 1.5) / (7.0);

let sim_v2 = levenshtein("cattle", "battle");
let mut config_v2 = DiffScoreConfig::default();
config_v2.sub_cost = 2.0;
let similarity_v2 = ((6.0) - 2.0) / (6.0);

let sim_v3 = levenshtein("Saturday", "Sunday");
let config_v3 = DiffScoreConfig::default();
let similarity_v3 = ((8.0) - 3.0) / (8.0);

assert_eq!(similarity, sim.similarity(&config));
assert_eq!(similarity_v2, sim_v2.similarity(&config_v2));
assert_eq!(similarity_v3, sim_v3.similarity(&config_v3));
}

#[test]
fn test_levenshtein_difference() {
use crate::levenshtein::levenshtein;
use crate::DiffScoreConfig;

let diff = levenshtein("Kittens", "kitten");
let mut config = DiffScoreConfig::default();
config.lowercase_sub_cost = 0.5;
let difference = 1.0 - ((7.0) - 1.5) / (7.0);

let diff_v2 = levenshtein("cattle", "battle");
let mut config_v2 = DiffScoreConfig::default();
config_v2.sub_cost = 2.0;
let difference_v2 = 1.0 - ((6.0) - 2.0) / (6.0);

let diff_v3 = levenshtein("Saturday", "Sunday");
let config_v3 = DiffScoreConfig::default();
let difference_v3 = 1.0 - ((8.0) - 3.0) / (8.0);

assert_eq!(difference, diff.difference(&config));
assert_eq!(difference_v2, diff_v2.difference(&config_v2));
assert_eq!(difference_v3, diff_v3.difference(&config_v3));
}
}
48 changes: 48 additions & 0 deletions crates/differ/src/lib.rs
Original file line number Diff line number Diff line change
Expand Up @@ -38,12 +38,32 @@ impl StringDiffOp {
}
}

#[derive(Debug, PartialEq)]
pub struct DiffScoreConfig {
pub sub_cost: f32,
pub lowercase_sub_cost: f32,
pub indel_cost: f32,
pub transpose_cost: f32,
// future properties here as needed
}

#[derive(Debug, PartialEq, Eq)]
pub struct Diff {
pub ops: Box<[StringDiffOp]>,
pub total_len: usize,
}

impl Default for DiffScoreConfig {
fn default() -> Self {
Self {
sub_cost: 1.0,
lowercase_sub_cost: 1.0,
indel_cost: 1.0,
transpose_cost: 1.0,
}
}
}

impl Diff {
pub fn new(diffs: Vec<StringDiffOp>, total_len: usize) -> Self {
Self {
Expand All @@ -55,6 +75,34 @@ impl Diff {
pub fn distance(&self) -> usize {
self.ops.len()
}

pub fn similarity(&self, score: &DiffScoreConfig) -> f32 {
let mut similarity_score: f32 = self.total_len as f32;
for i in self.ops.iter() {
match i.kind {
StringDiffOpKind::Delete => {
similarity_score -= score.indel_cost;
}
StringDiffOpKind::Insert(_x) => {
similarity_score -= score.indel_cost;
}
StringDiffOpKind::Substitute(_x, _y) => {
if _x.to_ascii_lowercase() == _y.to_ascii_lowercase() {
similarity_score -= score.lowercase_sub_cost;
} else {
similarity_score -= score.sub_cost;
}
}
StringDiffOpKind::Transpose => {
similarity_score -= score.transpose_cost;
}
}
}
similarity_score / (self.total_len as f32)
}
pub fn difference(&self, score: &DiffScoreConfig) -> f32 {
1.0 - self.similarity(&score)
}
}

pub(crate) fn get_operation_matrix(
Expand Down