I’m new to Weaviate and Elysia and am attempting to use Elysia to perform some email data exploration. I have two CSV files describing the email dataset: one which contains the email content plus all metadata except the recipient information, and the other which captures the email recipients. Both CSV files contain an emailID column that provides the requisite key for joining the tables. So far I’ve imported the CSV files into collections in my Weaviate cluster. What I’m wondering about now is how I specify the relationship between the emailID columns in the two collections? Does this require writing some Python code to use the API after I import the individual CSV files using the import tool? I tried querying the data via Elysia without that association, and not surprisingly I see errors. I would appreciate any pointers! Thank you!
hi @diehl !!
Welcome to our community 
You are probably looking into using Cross References.
However, there are some caveats: First, I don’t believe Elysia supports cross references for now. Second: for the sake of performance, you should always avoid cross references or have a very compelling use case when using it.
So I believe the best scenario for the use case you have described, is to normalize all data into one single collection.
Let me know if this helps!
Happy coding!
Thanks @DudaNogueira - Appreciate the quick reply!
So it sounds like in this scenario, it’s impossible at present to do email analysis by email relationship, since I don’t see an obvious way to include the recipients of each message into a single data table. That would require the ability to have an array of strings for all the email recipients. Am I missing anything here?