add corpus linguistics and data analysis tools

[adbar]

You might want to revise the sections "Corpus Linguistics" and "Data Analysis", I tried to add the tools accordingly but the distinction is not always clear-cut.

Short pitch

I posted the link to your repo on Mastodon and this answer makes a good point: further links could be added, you could have a look at the Tapor list you reference below.

Checklist

• I have read and understood the contribution guidelines.
• Table of contents has been updated (if a section is added / removed).
• Contents have been sorted alphabetically.

[maehr]

@adbar Thanks for your pull request. I've looked at it and also at the discussion on Mastodon. We're not trying to provide a comprehensive list, but a useful one. Hence the question for you: how do you use the tools you suggested? Why do they belong on the list? (I don't use any of these tools, stylometry is not my strong suit.)

[adbar]

@maehr I get your point but the "Corpus Linguistics" part of your list could really be extended.

• AntConc is used to index text corpora and run queries against it. It is not open-source but it is widely used (look for mention of it in Scholar if you're interested).
• TXM is fully open, it offers more functions (multilingual POS-tagging, subcorpora, web interface, etc.) and more geared towards statistical analysis of texts (with graphical display for instance).
• Mallet is widely used for topic modeling and document clustering, i.e. finding which documents belong together along which topic for data-driven corpus exploration.
• Stylo is used to extract a series of features for text classification, e.g. in authorship attribution. It can be partly used like Mallet but provides a wider range of different methods.

[maehr]

Thank you very much for getting back to me so quickly. What is your take on the section "Corpus Linguistics"? Integrate into "Data Anaylsis"?

[adbar]

Maybe put "Data Collection" first and then "Corpus Linguistics" immediately before "Data Analysis" would already help comparing the tools.

If the list continues to grow I see two options for "Data Analysis":

1. Add subsections like "Corpus Exploration", "Stylometry" or "Topic Modeling"
2. Focus on the medium and split it into "Text Analysis", "Image Analysis" etc.

Up to you, I admit it's not an easy task to classify these packages.

[sheiden]

I would have suggested to try to use somewhat the Tadirah terminology and structure (https://tadirah.info/pages/Browser.html) produced by the DH community. But it doesn't contain a 'Text analysis' entry :-(

[diegosiqueir4]

Thank you @adbar, @maehr and @sheiden for your contributions.

[diegosiqueir4]

it looks good :)