POSTagger is a GUI app for semi-automatic POS tagging.
This is a beta version. I'm looking for your feedbacks.
To install the app copy the follwing command in your terminal:
>>> pip install POSTagger
You can also download the repository via Github website or the following command:
>>> git clone https://github.com/asdoost/POSTagger
After downloding the repository, go to the POSTagger folder and copy the following command:
>>> chmod +x POSTagger.py
To start POSTagger, simply type POSTagger in your command line:
>>> POSTagger
If you have downloded the repository using git clone, go to the POSTagger folder and type the following command in your command line:
>>> ./POSTagger.py
POSTagger starts with an openning wizard. Here, you can either import a project that you have already saved, or you can start a new project by importing a corpus.
In the fourth line of the wizard you can choose between tagsets. There are 5 tagsets: default, Brown tagset, PENN Treebank tagset, Universal POS Tagset, and Bijankhan tagset.
In the fifth line of the wizard, you can choose between three POS tagging style: XML, underscore, and tab.
After choosing the preferred setting, press OK to start tagging.
The main window consists of six parts:
- Text Box
- Frequency Label
- Word Entry
- Edit Buttons
- Tagset
- Tagging Buttons
Text Box displays the current sentence that your are tagging.
Word Entry displays the word that should be tagged.
Frequency Label displays the frequency of the word that you are tagging.
Edit Buttons consist of three buttons:
Merge button which combines the cuurent word with the following word.
Split button undoes what has been merged.
Re-tokenize button retokenize the sentence up to the last word.
Tagset provides POS categories.
Tagging buttons consist of two buttons:
Tag which assigns the chosen tag two the word, using the preferred tagging style.
Untag strips the last word out of the chosen tag (untags what has been tagged).
From the File menu choose Save the project to save what you have done.
When you finished your project, from the File menu choose Export to get your work in a single txt (for underscore and tab tagging style) or xml (for XML tagging style) file.
From the File menu choose Preferences to change the tagset, font, font size, or text direction.
