Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hello,
I want to create a count of sentiment themes in my load script but can't work out an efficient way to do this.
The data set is based on customer surveys which has a free type section. This free type section has already been broken down into a new data set with key themes.
For example
Customer ID KeyTheme
1 Call Control;Wait Time;Helpfulness
2 Wait Time;Professionalism
3 Behaviour;Friendness;Trust
.......
etc
The solution I'm currently using is:
If(WildMatch([Key Theme],'Wait Time*'),'Wait Time'
but the load time is long.
I've read through the help guide on Qlik and a lot of forum posts. The preferred solution seems to be an inline load. So far I've only used inlines loads in connection with a Mapping Load.
Potentially, due to my limited understanding, I don't think an inline load would work here. The KeyTheme field has 200,000+ possible values over 12 million rows. If I've understood how to use the inline load I need to include all these value in the script.
The wildmatch does work, it lets me state that X customers mentioned KeyTheme Y in month Z.
But the load time is extremely long.
Is there an efficient way of doing this?
Due to our company policy I cannot load data file as an example - not if I want to stay employed at least.
Have you considered using the Subfield() function to create an additional table?
KeyWords:
Load [Customer ID],
subfield([KeyTheme],';') as KeyWord
Resident [Customer KeyThemes Table];
Have you considered using the Subfield() function to create an additional table?
KeyWords:
Load [Customer ID],
subfield([KeyTheme],';') as KeyWord
Resident [Customer KeyThemes Table];
Thank you,
I've spent roughly 20 hours trying to make a solution work, and this is perfect.
Simply, elegant and most importantly works. Reduced the load time by 3 hours.