morra 1.0.7

Morphological parser (POS, lemmata, NER etc.)

RuMor: Russian Morphology project

Morra: morphological parser (POS, lemmata, NER etc.)

A part of RuMor project. It provides tools to organize a pipeline for complete morphological sentence parsing and named-entity recognition.

Scores on SynTagRus: accuracy 98.45% for POS tagging; 98.74% for lemmata detection.

This project was making with a focus on Russian language, but it can also be used with some other languages (European, at least).

Installation

pip

Morra supports Python 3.5 or later. To install it via pip, run:

$ pip install morra

If you currently have a previous version of Morra installed, use:

$ pip install morra -U

From Source

Alternatively, you can also install Morra from source of this git repository:

$ git clone https://github.com/fostroll/morra.git
$ cd morra
$ pip install -e .

This gives you access to examples that are not included to the PyPI package.

Usage

Input and output format for Morra is CoNLL-U when input or output is a file, or Parsed CoNLL-U if it is an object. Also, it allows Corpuscula's corpora wrappers as input.

MorphParser Basics

Part of Speach Tagging

Lemmata Detection

Morphological Feats Tagging

Named-entity Recognition

Supplements

Examples

You can find them in the directory examples of our Morra github repository.

License

Morra is released under the BSD License. See the LICENSE file for more details.