search

setriplette / golden_age

4 stars 1 forks source link

Pilgrim Emotions #10

Open ahunker opened 7 years ago

ahunker commented 7 years ago

I've compiled a general list of emotions that I tagged in the text. @setriplette if there are any words on here you think I should remove or any you think I should look for, please let me know. I've only listed one form of the words, so I've written love and not lovingly, loved, loves, etc., otherwise the list would be extremely long.

ebeshero commented 7 years ago

Thanks @ahunker ! Is your XML for the Pilgrims text up to date here in the repo?

This is really timely as @setriplette and I are about to work on your project schema, and these look like they might be attribute values, if your team wants to use them. You may want to boil them down, though, for simplification. (For example, what's the difference between Melancholy and Unhappy?)

ahunker commented 7 years ago

I haven't updated my xml, so I can do that and push the changes. The attribute values on emotions, as according to @setriplette tag list, is differentiating the parts of speech, such as noun or verb, rather than the meaning of the emotions themselves.

ahunker commented 7 years ago

The Pilgrim xml is updated and pushed.

ebeshero commented 7 years ago

@ahunker Thank you for pushing the file and explaining! I see (and remember) now how this works: You're tagging words in the text that indicate emotion, and noting their grammatical part of speech, which is very reasonable and interesting!

I wonder how frequently certain words show up in the text? (You can use <xsl:for-each> to walk through a list of distinct values, and for each value, go in and walk through your XML tree to do a count() of how frequently that value is used in the text.) That would be a great thing to try in XSL Ex 7. I would set up the <xsl:for-each> in the template that matches on your document node, because you want it to walk through the whole tree and bring back a count number for each item on a list of distinct-values, a list that you need to make only once on the entire document.

etc...

If you make such a list, you can use <xsl:sort> to sort the output either ascending or descending, and show which words are being used most frequently and least frequently! You could also reach in and get their parts of speech, etc: whatever seems interesting.