Features

This package contains a variety of useful functions for text mining in Python 3.

It focuses on statistical text mining (i.e. the bag-of-words model) and makes it very easy to create a term-document matrix from a collection of documents. This matrix can then be read into a statistical package (R, MATLAB, etc.) for further analysis. The package also provides some useful utilities for finding collocations (i.e. significant two-word phrases), computing the edit distance between words, and chunking long documents up into smaller pieces.

The package has a large amount of curated data (stopwords, common names, an English dictionary with parts of speech and word frequencies) which allows the user to extract fairly sophisticated features from a document.

This package does NOT have any natural language processing capabilities such as part-of-speech tagging. Please see the Python NLTK for that sort of functionality (plus much, much more).

The original code and documentation is available in PyPI under the package name textmining. This package is a port to Python 3 and published in PyPI under the package name textmining3, and is based on the original.

Credits

The original textmining 1.0 package code was authored by Christian Peccei <cpeccei@hotmail.com>

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

1.1.0 (2018-13-19)

• Add new feature to export DTM to pandas.DataFrame

1.0.2 (2018-12-19)

• First port of textmining to Python 3

1.0.0 (2010-01-11)

• Original release of textmining on PyPI (see https://pypi.org/project/textmining/1.0/)