# kakao/khaiii

Kakao Hangul Analyzer III

Repository: https://github.com/kakao/khaiii
Canonical: https://ross.abutalabs.com/products/khaiii
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-07-17T17:09:48+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 93, release rhythm 8, longevity 100
- inputs: {"age_days": 2837, "days_push": 47, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1450, forks 300 (observed 2026-08-28T04:04:45.913092+00:00)

## What it is
Khaiii (Kakao Hangul Analyzer III) is a data-driven Korean morphological analyzer developed by Kakao, using a CNN-based neural network model trained on the Sejong corpus. It segments Korean text into morphemes with part-of-speech tags and supports user dictionaries for pre-analysis and error patching.

## Use cases
- tokenize korean text into morphemes
- korean pos tagging for nlp pipelines
- preprocess korean sentences for search indexing
- analyze korean text with poor spacing
- add custom dictionary entries for korean named entities
- fast korean morpheme analysis in production services

## When to choose
- you need accurate, fast Korean morphological analysis
- your input text has spacing errors and needs robust segmentation
- you want a C++ core with Python bindings for low-latency NLP preprocessing

## When to avoid
- you need morphological analysis for languages other than Korean
- you need a pure-Python dependency-free solution
- you require frequent model retraining or GPU inference

## Facets
- artifact type: library
- maturity: maintenance
- function: nlp, machine-learning
- domain: machine-learning, developer-tools
- platform: python, cpp
- tags: korean, morphological-analysis, tokenization, cnn, hangul, pos-tagging, natural-language-processing, linux, macos

## Member repositories
- kakao/khaiii (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:45.913092+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:35:56.703274+00:00, confidence not recorded.
  - readme: https://github.com/kakao/khaiii (fetched 2026-08-28T04:04:45.913092+00:00, sha b0f33cd1602d)
- Data as of 2026-08-30T08:39:29.467469+00:00.