1
54%
dev: 65%

Ran 12 Jan 2026 05:23AM UTC

Files 190

Run time 5s

Badge

Embed ▾

Committed 12 Jan 2026 05:20AM UTC coverage: 53.705%. First build

Job # 20908908159.1

Build Type

Pull #1175

github

Committed by

Copilot

Commit Message

Add complete_soundex_similarity function for phonetic similarity calculation

Implemented the similarity calculation formula from Tapsai et al. (2020) paper:
- Character-wise comparison: S(X,Y) = Sum(sim(c_xi, c_yi)) / max(len(X), len(Y))
- Binary matching: sim = 1 if characters match, else 0
- Normalized by maximum length of the two codes

Features:
- Follows Equation (1) from Section 3.3 (page 55) of the paper
- Compares codes position-by-position (not by syllable blocks)
- Returns float between 0.0 (no match) and 1.0 (perfect match)
- Handles edge cases: empty strings, different lengths

Testing:
- Added comprehensive unit tests for various scenarios
- Perfect match returns 1.0
- Tone-only differences show ~0.83 similarity (5/6 matches)
- Multi-syllable words correctly compared character-by-character
- All tests passing

Examples verified:
- Perfect match: ก้าน vs ก้าน = 1.0
- Tone difference: คข7M2- vs คข7M0- = 0.8333 (5/6)
- Multi-syllable: ปุญญา vs บุญญา = 0.75
- Different words: ทราย vs ก้าน = 0.4286

Exported complete_soundex_similarity in pythainlp.soundex module.

Co-authored-by: wannaphong <8536487+wannaphong@users.noreply.github.com>

Pull Request Pull Request #1175: Add Complete Soundex for Thai Words Similarity Analysis with Similarity Function

Run Details

4457 of 8299 relevant lines covered (53.71%)

0.54 hits per line

PyThaiNLP / pythainlp / 20908908159 / 1
54%
dev: 65%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 20908908159.1

PyThaiNLP / pythainlp / 20908908159 / 1 54% dev: 65%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 20908908159.1

PyThaiNLP / pythainlp / 20908908159 / 1
54%
dev: 65%

README BADGES
x