I have a table with columns company, connection type and some other stuff. I want a Venn diagram showing companies using on-prem connections and roaming connections. Some companies will use both but not at the same time. I can't work out how to get such an item.
Ways in which this could work might be: - Select all rows with company x where company x has a row using "roaming". But I can't see how. - Do an aggregate in the data manager view which created a table with the right data. Unfortunately, I only seem to be able to get one dataset into Omniscope at a time. This seems like a crazy restriction so I guess I'm missing something.
You may need to use the tokenize operation to join them up. You may also need to create a category field as the way Venn diagrams work depends on grouped up category tags.
Venn View works with row subsets generated by filters/queries, so you should reset the filter states, to make sure you are looking at the whole dataset, then create data subsets or segments where: [Company]=A, [Connection type]=B. Use filters to achieve this, then click on the Query button and choose Create new query>from filtered data, and give that subset a name.
Reset the filters again and create a different segment, say all the records from [Company] and only [Connection type]= X.
Venn View works well with 2-4 queries, and you can add them to the visualisation via Sets menu in the Venn view.
Omniscope will allow you to store multiple values in the same cell, so for example you can store in the [Connection] field: "roaming, on-prem". You should specify in the Field Organiser in the DataManager that this field is tokenised (also possible in DataExplorer, right click, Field options>Tokenise). That will tell Omniscope that each value stored in the cell, separated with comma in this case, should be treated as a separate value, not as a string. Tokenising the field will enable you to have 1 record falling into 2 queries, one looking at records with "on-prem", another looking for "roaming". Records that have both values will appear in the intersection zone.
That would work, if I can create a filter that says "select * from data where company in (select company from data where connection = 'roaming')" but I can't figure out how to do that either.
When you set the field as tokenised, each value will appear on the sidebar filter as individual value, so even if you have only 1 record in the dataset, and content of the [Connection] field is "roaming, on-prem", you should see "on-prem" and "roaming" as individual filter choices on the sidebar. You can then choose one or more values in each field to create a query from filter.
Please feel free to post a sample dataset and we can configure this for you...
OK. Got it. Thanks for your help. What was eventually needed was: - Aggregate on data with a unique values list - A FieldOrganizer to rename the fields - A join with the original data - A tokenize operation on the list of connector