NLP Project Discussion Room

  NLP DataPoints:

Here are some examples of DataPoints (as per my understanding).

* There are lots of other options for metadata and how we can clean up and organize each unit of information in our project, but that should be a good starting point to start the discussion and fill up the gaps in future.

  Statistics of all the Variables on Seshat Websites

* Note that Seshat has two old websites. One is Seshat.info (aka Wiki) which serves as the place where all the facts go before being reviewed by Seshat Experts. The public Seshat Website is Seshatdatabank.info (aka Browser) is where only the expert-reviewed data shows up. From my experince, most of the data on Wiki is also of great quality and can be used (with minimun extra effort) in our research projects, among which the current NLP project. For example, all the examples in the above table are taken from Wiki.

* What you see below is the first draft of my analysis of all Seshat variables. As you can see there are so many gaps to be filled in order to take advantage of the full Seshat potential. Some gaps can (and will) be filled by me and Jakob, and for some other gaps, we need some expert and RA help. By filling the gaps, I mean, the process through which we reduce the difference between the current state of the data and the ideal super-clean database where every citation has a Zotero ID and page number and PDF files and correponding text.

• Wiki

  (seshat.info)

• Browser

  (seshatdatabank.info)