The dynamics and social organization of innovation in the field of oncology/Tutorial esmo.py

This page documents some use cases of esmo.py instrument from the esmominer repository.

Tuning the parameters

1. Open a terminal/command(cmd) window and change directory to ./esmo
2. Open EsmoDescription.py located in the ./esmo/src folder
   • This file contains the text mining attributes for every ESMO year. A few notable attributes are size and font
3. From the ./esmo directory, you can print the list of conferences or the commands available using

   src/esmo.py --list
   src/esmo.py --help
   

   Here we'll assume you already gone through the commands for downloading and producing a first version of mined text for one year, ending with one of the -to-json commands.
4. You can view the abstracts, in this example for 2004, by using the command below. Adding --inspect will also pair the text from every document with it's json counterpart which contains the detailed attributes

   src/esmo.py 2004 --changes --view
   src/esmo.py 2004 --changes --view --inspect
   

5. To change what esmo.py considers in it's text mining process, change attributes for specific years within EsmoDescription.py and save the document. Then run the following command

   src/esmo.py 2004 --changes --update
   

   This will secure the changes.
6. Using the following command will allow you to see the effect that your change to EsmoDescription.py produced

   src/esmo.py 2004 --changes --diff --inspect
   

7. If the changes satisfy you, use the following command to implement your changes for good

   src/esmo.py 2004 --changes --accept
   

8. Return to step 4 and view the updated way in which esmo.py views the documents, or straight to step 5 to keep improving your text mining.