Anonymize/Mask the Person's Name from the comment field

Hi ,

I would like to mask or anonymize the person's name in comment text field. for eg : Lee is good person should be mask to abc is good person etc.

is there any way to do this in Alteryx?

Thanks

Hsac

Fuzzy Match

Data Investigation

Accepted answers

ddiesel

Ah, I see. Since you do not have a list of expected names, you will need an NLP tool. I suggest that you look into Named Entry Recognition that is part of the Alteryx Intelligence Suite.

https://www.youtube.com/watch?v=se-wnjfP2UY

https://www.alteryx.com/products/intelligence-suite

https://help.alteryx.com/20221/designer/alteryx-intelligence-suite

https://help.alteryx.com/20221/designer/named-entity-recognition

Let us know if this is what you're looking for.

Thanks,
Deb

Robin_McIntosh

Here's the yxmd workflow.

Discussion_Anonymize_Try.yxmd

All comments

messi007

@harinder1301,

You can extract the list of the name from the data source

Use https://www.mockaroo.com/ to create a list of name

Then use join by position to get the new names :

Attached the workflow,

Hope this helps!

Regards,

Rename.yxmd

ddiesel

Hi @harinder1301!

To build on @messi007's suggestion, you might also look into the functionality of the Find and Replace tool.

Anonymize Mask the Person's Name from the comment field_dmd.yxmd

harinder1301

Thanks, Is there any way to find the names of the persons from nearly 5000 comments?

harinder1301

Thanks for the Quick reply.

I have attached some sample records for your reference.

Regards

sample.csv

messi007

@harinder1301,

Similar to @ddiesel workflow :

Hope this helps

Regards

FindReplace_20220622.yxmd

harinder1301

Thanks for the reply. Actually my question is " How can i identify the names from 5000 comment rows?"

Is there any tool in Alteryx to identify the names from the comment?

Appreciate your reply.

ddiesel

@harinder1301 Do you have the list of names? What do you want the final output to look like?

harinder1301

Hi,

No, We don't have person list. This can be first name/known as/or surname etc. This the feedback provided. Here is the desired final output:

ID	Comments	Final Output
1	Joe is bad person.	xyz is bad person.
2	yes, I like John.	yes, I like xyz.
3	Would not allow Kiara to come.	Would not allow xyz to come.
4	Wouldn't change a thing.	Wouldn't change a thing.
5	Yet, there is nothing much to change.	Yet, there is nothing much to change.
6	Trin is very slow.	xyz is very slow.
7	I like yun.	I like yun.
8	Yes, Peter is not a good.	Yes, xyz is not a good.
9
10
..5000

Thanks for your help!

Regards

ddiesel

Ah, I see. Since you do not have a list of expected names, you will need an NLP tool. I suggest that you look into Named Entry Recognition that is part of the Alteryx Intelligence Suite.

https://www.youtube.com/watch?v=se-wnjfP2UY

https://www.alteryx.com/products/intelligence-suite

https://help.alteryx.com/20221/designer/alteryx-intelligence-suite

https://help.alteryx.com/20221/designer/named-entity-recognition

Let us know if this is what you're looking for.

Thanks,
Deb

Robin_McIntosh

Doesn't look pretty but took a stab at it here. Assuming that all the names would start with a capital letter, I used RegEx to extract possible names. Only issue is if there is a word other than a name capitalized (but I did exclude singles like I and A). You would then just need to update this list to remove non-names. Only other thing is that all anonymizing is done with xyz so there would be no distinct replacement per name.

Discussion_Anonymize_Try.yxwz

harinder1301

Thanks Robin, Please can you share the yxmd file.

Regards

harinder1301

Many Thanks Robin and Deb,

Both solutions worked for us but we have chosen NLP as it worked better with our data.

Thanks

Hsac

Quick Links

This months top contributors

atcodedog05 19598

Qiu 15878

binu_acs 15708

MarqueeCrew 13708

apathetichell 13703