Orange3 Text extends Orange3, a data mining software package, with common functionality for text mining. It provides access to publicly available data, like NY Times, Twitter, Wikipedia and PubMed. Furthermore, it provides tools for preprocessing, constructing vector spaces (like bag-of-words, topic modeling, and similarity hashing) and visualizations like word cloud end geo map. All features can be combined with powerful data mining techniques from the Orange data mining framework.

Please note that Text add-on won't work on 32-bit Windows systems. The add-on depends on conda-forge and they have removed support for Windows 32 in April 2018.

The easiest way to install Orange3-Text is with Anaconda distribution. Download Anaconda for your OS (Python version 3.5). In your Anaconda Prompt first add conda-forge to your channels:

conda config --add channels conda-forge

Then install Orange3-Text

conda install orange3-text

Run

python -m Orange.canvas

to open Orange and check if everything is installed properly.

To install the add-on from source

# Clone the repository and move into it
git clone https://github.com/biolab/orange3-text.git
cd orange3-text

# Install the dependencies:
pip install -r requirements.txt

# Finally install Orange3-Text in editable/development mode.
pip install -e .

To register this add-on with Orange, but keep the code in the development directory (do not copy it to Python's site-packages directory), run

python setup.py develop

If you're not using Anaconda distribution, you can manually install biopython library before installing the add-on. First, download the compiler Visual Studio and run the setup with:

python setup.py build_ext --inplace --compiler=msvc install

After the installation, the widgets from this add-on are registered with Orange. To run Orange from the terminal, use

python3 -m Orange.canvas

or

orange-canvas

The new widgets are in the toolbox bar under Text Mining section.