Construct example on how to use data

JMSLab / LaroplanOCR

Swedish primary school curricula (Läroplaner för grundskolan) in digital format.

MIT License

2 stars 0 forks source link

Construct example on how to use data #4

Closed santiagohermo closed 2 years ago

santiagohermo commented 2 years ago

In this issue we will come up with an illustration on how to use the data. We will implement it in code and add it to the repo.

Notes:

An idea on how to do this is in https://github.com/JMSLab/LaroplanOCR/issues/1#issuecomment-1022342046

santiagohermo commented 2 years ago

From our last call I got the sense you wanted to help out @miikapaal, so I assigned you here. Let me know if you don't have time.

santiagohermo commented 2 years ago

Per call, prioritize this issue over #3.

santiagohermo commented 2 years ago

Here is a first pass on the example @miikapaal @jmshapir @dagese.

I created a folder /example/ which contains its own readme (access current state of folder here). I added a reference to the example in the main README (see here)
I wrote the code linearly, thinking that people will open RStudio and run it manually. See code here
I added two types of ways of searching for words: wildcard and exact. I search one word using both to illustrate the difference.
I tried words like 'technology' and 'computer' but they had very few hits, only in the 1980 laroplan. Thus I decided to change to primary school subjects. We can of course change them.

I'm open to changing everything (keywords, name of folders and files, etc). Curious to hear your thoughts!

jmshapir commented 2 years ago

Thanks @santiagohermo! I took an initial pass through. (I made a tiny edit on the main readme.)

Generally this looks great. From my side, I think you could open a pull request and we could continue commenting in detail there.

After we wrap up this issue, we could ask someone who is new to these data to go over the repository and try to "operate" the example, to see if that turns up any new issues we haven't anticipated.

santiagohermo commented 2 years ago

Thanks @jmshapir! I took another look and made one more small change.

I'll move to PR, and hopefully we'll get @miikapaal's thoughts then.

santiagohermo commented 2 years ago

Continues in #5

santiagohermo commented 2 years ago

Summary: In this issue we added an example in a new folder that shows how to use the data to search for counts of some words, and make a plot with the output. We added a link to the folder in the main README.

Changes merged to main here