Load only Meaningful Data in Qlik Sense

[pradeep92]

Hello,

I would like to load only meaningful data into the Qliksense Applications. I am not aware of what all the data that is available in columns as there are millions of Records.

Example:

Field

[

fdfkjdkf,

Arjun,

dfsds,

mm,

Quality,

dfdf,

Product,

sfs

];

The desired output has to be Arjun,Quality,Product. 

This is just a small piece of data, where I am not aware of what all the meaningful data that will be available.

How to load only meaningful data into app.

Thanks!

[anthonyj]

Hi Pradeep,

If you're using the Data Manager to import your data you can untick the fields that you don't want. In which case you can leave the Arjun, Quality and Product fields ticked.

If you're using the Data Load Editor then you can call them out specifically in the load statement.

eg.

Data:

load

Arjun,

Quality,

Product

from [yourData.csv];

If I've misunderstood your question then please clarify and I'll do what I can to assist.

Thanks

Anthony

[pradeep92]

Hello Anthony,

Thank you for the response .
sorry , you have misunderstood the question .

All the values are present as single column ( field)

eg:

customer is a field and below are the values,

customer

[
Arun,

Ram,

dkdkd,
rorkdk,

llaksk,

Vishnu];

I need to have the output of Arun , Ram and Vishnu into the app and not load other meaning less data.

[Vegar]

Take a look at the LOAD FROM_FIELD functionality in Qlik Sense. It allows you to handle a single field as a data source.

https://help.qlik.com/en-US/sense/August2021/Subsystems/Hub/Content/Sense_Hub/Scripting/ScriptRegula...

[Vegar]

I found this blog post that describes the use of from_field The Qlik Sense FROM FIELD Function by Susie Bann.

[pradeep92]

Hello Vegar,

Thank you so much for the response . 
but however I have all the values in BACKEND ORACLE TABLE with varchar type.

Those are not URLS.

so the table is like ,

Fieldname

[
Arjun

Ram

dkxkc

dkdkkd

];

I need to load only Arjun and Ram ; I.e only meaningful words . 

There are million of rows . 
I am not actually not able to do where exists or match function as I am not aware of all the meaningful words

[Lemac]

Dear Pradeep,

What you ask for is quite hard and maybe even impossible. Words like Arjun, Amit, Vishnu, Asif, Pradeep may be meaningful to an Indian, but to a Dutch like myself it is still not meaningful.  If I would confront you with dutch words like 'gezellig, stroopwafel, kaairen, klompen, virmanie, windmolen, tulp.' It still would be gibberish to you. You probably wouldn't notice that there are two words in there which are absolute nonsense.

That is because meaning is derived from the author and given by the reader. 

So the best you could probably do is make a definitive list of all the meaningful wordst (start with a dictionairy perhaps?) Or, you could try to lookup each and every word in Google and return the number of hits? That only words with >1,000 hits in the English language are used?

[stevejoyce]

The only thing i can think of defining "meaningful" is there is other associated data with those attributes (fact data).  In that case you can use Keep/Join to limit dimension values only where there are records in another table.

fact:

load * inline [

customerid, value

1, 100

2, 200

] 

;

left keep (fact)

customer:

load * inline [

customerid, customer

1, Arun

2, Ram

99, asdf

98, hwer

]

;

In this case, asdf and hwer will be dropped in the customer table.

[pradeep92]

Hello ,

Thank you so much . We felt it is impossible and now we are creating nee field in the backend itself.

Thank you so much !!