kandu / wseg

A word identification system

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

wseg

A word identification system

Usage

In the test directory, there are two plain text files that serve as dictionary files, one for characters and one for words. And there is a test.ml program that illustrates how to build dictionary and index with wseg and how to apply several rules to identify words from a sentence. Just invoke make runtest to play with the demo.

char.dic contains 12640 Chinese characters and word.dic contains 157202 words. So you can expand the dict or demo for common usage.

About

A word identification system


Languages

Language:OCaml 96.5%Language:Makefile 3.5%