Software for Glossing/Tagging and Mapping Patterns in Text

Question

Pleasantries

I'm an English major temporarily studying at Oxford and I feel my literary/rhetorical analyses and argumentation could use a touch of the 'objectivity,' real or supposed, that computational analysis does readily seem to lend. I know nothing about NLP, and I'm not a linguist. I do, though, have basic knowledge of Ruby, XML, and Python.

The Question

That said, I'm looking for software/modules/languages that allow me to

1. Input text (well, duh...)
2. Gloss/tag/markup (preferably through a context menu as opposed to actual XML-like tags) text so that I can categorize it by rhetorical/literary device, concept, conjugation...
3. Summarize, visualize, compare, and export results.

Pipe dream or extant product? Please do let me know.

Answer 1

Since you already know some python you could take a look at the Natural Language Toolkit - nltk - it has a large suite of tools and the ability to customise most of them.

Of your requirements:

1. Input text - of course
2. Gloss/tag/markup (preferably through a context menu as opposed to actual XML-like tags) text so that I can categorize it by rhetorical/literary device, concept, conjugation... Usually performed by tailoring the analysis mechanism rather than manually
3. Summarize, visualize, compare, and export results. Yes - it is python so you can do a lot

Some additional features:

• Free (Libre & FLOSS)
• Cross Platform
• Extensible
• Widely used
• Active support
• For some examples of the sorts of analysis that can be performed there is an online demo.