Repository navigation
Any way for this library to work without the active learning part i.e. in an unsupervised manner?
#1155
|
I wish to automate the process of generating cluster Ids for my dataset without having to provide labeled pairs or any input other than the data and schema. Is this possible with this library? |
Answered by
NickCrews
Apr 10, 2023
Replies: 4 comments
|
Dedupe does not support unsupervised. If you need unsupervised training then you will want to look at a different software package. This list is a bit outdated, but it's a reasonable starting point. https://github.com/J535D165/data-matching-software |
0 replies
Answer selected by
ppb-rapid
|
@NickCrews is right. i think there's some very interesting possibilities, but nothing in dedupe yet. |
0 replies
|
Thank you for the quick response and the list @NickCrews @fgregg ! Dedupe is wonderfully written, I learnt a lot. |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Dedupe does not support unsupervised. If you need unsupervised training then you will want to look at a different software package. This list is a bit outdated, but it's a reasonable starting point. https://github.com/J535D165/data-matching-software
I personally recommend splink.