Skip to main content

a tool to detecting the language for a small piece of unicode text without any dependency to other libraries.

Project description

# lang-detect: a tool to detect language

Detecting the language for a small piece of unicode text without any dependency to other libraries.

Currently we support detecting de, en, es, fr, it, ja, nl, pl, ru, zh-hans, zh-hant, and zh-yue.

After some simple testing, we found that the result for long sentence is better.

## Method

We focus on the Basic Multilingual Plane in unicode encoding, and current language support set could be extended.

For each language, we use a uniformed ngram vector to represent the language itself. This vector can be seen at the data folder.

When we detect a text, we generate the uniformed ngram vector for this text, and just comparing the cosine value of the angle between the text vector and the language vector.

To get the language vector, we use feature articles on Wikipedia as corpus.

## Usage

cd to the project root

bin/langdetect YOUR_SENTENCE_HERE

Project details


Release history Release notifications

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
lang_detect-0.0.1-py2.6.egg (1.4 kB) Copy SHA256 hash SHA256 Egg 2.6 Aug 18, 2011

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page