Python wordsegment Module

Python wordsegment Module

Python WordSegment Module is an Apache2 licensed module for English word segmentation, written in pure-Python, and based on a trillion-word corpus.

Based on code from the chapter “Natural Language Corpus Data” by Peter Norvig from the book “Beautiful Data” (Segaran and Hammerbacher, 2009).

Data files are derived from the Google Web Trillion Word Corpus, as described by Thorsten Brants and Alex Franz, and distributed by the Linguistic Data Consortium. This module contains only a subset of that data. The unigram data includes only the most common 333,000 words. Similarly, bigram data includes only the most common 250,000 phrases. Every word and phrase is lowercased with punctuation removed.

Features

  • Pure-Python
  • Fully documented
  • 100% test coverage
  • Includes unigram and bigram data
  • Command line interface for batch processing
  • Easy to hack (e.g. different scoring, new data, different language)
  • Developed on Python 2.7
  • Tested on CPython 2.6, 2.7, 3.2, 3.3, 3.4 and PyPy 2.2

User Guide

Installing WordSegment is simple with pip:

> pip install wordsegment

You can access documentation in the interpreter with Python’s built-in help function:

>>> import wordsegment
>>> help(wordsegment)

In your own Python programs, you’ll mostly want to use segment to divide a phrase into a list of its parts:

>>> from wordsegment import segment
>>> segment('thisisatest')
['this', 'is', 'a', 'test']

WordSegment also provides a command-line interface for batch processing. This interface accepts two arguments: in-file and out-file. Lines from in-file are segmented iteratively, joined by a space, and written to out-file. Input and output default to stdin and stdout respectively.

> echo thisisatest | python -m wordsegment
this is a test

API Documentation

segment(text)
"Return a list of words that is the best segmenation of `text`."
...
 
score(word, prev=None)
"Score a `word` in the context of the previous word, `prev`."
...
 
divide(text, limit=24)
"Yield (prefix, suffix) pairs from `text` with len(prefix) not exceeding `limit`."
...
 
# Mapping of (unigram, count) pairs. Loaded from the file 'unigrams.txt'.
unigram_counts
 
# Mapping of (bigram, count) pairs. Loaded from the file 'bigrams.txt'.
bigram_counts

Useful Links