You can test the online version of MultiAzterTest at the address http://ixa2.si.ehu.eus/aztertest
Multiaztertest is an upgrade to the Aztertest application meant to evaluate texts in various languages by calculating multiple metrics and indicators of the texts' content and analyzing those results to determine the complexity level of those texts.
If you want more information about the metrics analyzed in MultiAzterTest you can read the preprint that we have uploaded to arxiv:
https://arxiv.org/abs/2109.04870
- Download
multiaztertest.pyand thedata,corpusandwordembeddingsfolders into the same directory - Use the following commnads to install the necessary operative system packages in Ubuntu 18.04:
sudo apt install build-essential
#dos2unix:to convert plain text files in DOS or Mac format to Unix format
sudo apt install dos2unix
#python-dev contains everything needed to compile python extension modules
sudo apt-get install python3-dev
#python3-pip: install pip3 for python3
sudo apt install python3-pip
#to install textract package in ubuntu 18.04 are necessary the following packages: libpulse-dev and swig
sudo apt-get install libpulse-dev
sudo apt-get install swig
- Use the following commands to install the necessary python packages:
pip3 install stanfordnlp
pip3 install nlpcube
pip3 install wordfreq
pip3 install pandas
pip3 install sklearn
pip3 install --upgrade scikit-learn==0.22.1
pip3 install --upgrade gensim
apt-get install python-dev libxml2-dev libxslt1-dev antiword unrtf poppler-utils pstotext tesseract-ocr \ flac ffmpeg lame libmad0 libsox-fmt-mp3 sox libjpeg-dev swig
pip install textract
Once MultiAztertest has been installed, run it using the following parameters:
python3 multiaztertest.py -c -r -f $dir/*.txt -l language -m model -d /home/workdirectory
Currently available languages: english, spanish, basque
Currently available models: stanford, cube