Hello Julio, Thanks for your feedback! The Elbow method may produce plots with concavity changing from one point to a next. It means that some points must be removed from consideration, and my method takes care of this. In addition the math is adapted for clustering procedure and much simpler. This helps with efficiency and maintenance. Best regards, Mya
On Thu, Oct 22, 2020 at 9:04 AM <[email protected]> wrote: > Send scikit-learn mailing list submissions to > [email protected] > > To subscribe or unsubscribe via the World Wide Web, visit > https://mail.python.org/mailman/listinfo/scikit-learn > or, via email, send a message with subject or body 'help' to > [email protected] > > You can reach the person managing the list at > [email protected] > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of scikit-learn digest..." > > > Today's Topics: > > 1. Numeric version of Elbow method for finding an optimal > cluster number (Maiia Bakhova) > 2. Re: Numeric version of Elbow method for finding an optimal > cluster number (Julio Antonio Soto) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 21 Oct 2020 10:48:19 -0700 > From: Maiia Bakhova <[email protected]> > To: [email protected] > Subject: [scikit-learn] Numeric version of Elbow method for finding an > optimal cluster number > Message-ID: > < > cac7jaaoy0ub8zkua0a_w0aujcv5vg4m4fbttnh1n-exp-4s...@mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > Hello, > Sorry, this is not about scikit-learn. It is my method which I once offered > here, and was told that > it is too easy to code. Since then I have been talking to other people and > learn that linguists > do not understand enough math to work it out, and even people who are good > at math would prefer > a ready script. I wrote one and would like to share. > I did a numeric implementation of the Elbow method for calculating the > optimal cluster number. I will appecuat it greatly if you show my work here > and people would give me feedback on it. > https://github.com/Mathemilda/Numeric_ElbowMethod_For_K-means > Best, > Mya > -- > Maiia Bakhova > Mathematician in Data Science > http://myabakhova.blogspot.com > https://www.linkedin.com/in/myabakhova > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: < > https://mail.python.org/pipermail/scikit-learn/attachments/20201021/98b526fd/attachment-0001.html > > > > ------------------------------ > > Message: 2 > Date: Wed, 21 Oct 2020 19:53:08 +0200 > From: Julio Antonio Soto <[email protected]> > To: Scikit-learn mailing list <[email protected]> > Subject: Re: [scikit-learn] Numeric version of Elbow method for > finding an optimal cluster number > Message-ID: <[email protected]> > Content-Type: text/plain; charset="utf-8" > > Hi Mya, > > To me, it looks very similar to the kneed library: > https://github.com/arvkevi/kneed > > Best regards > > > > El 21 oct 2020, a las 19:48, Maiia Bakhova <[email protected]> > escribi?: > > > > ? > > Hello, > > Sorry, this is not about scikit-learn. It is my method which I once > offered here, and was told that > > it is too easy to code. Since then I have been talking to other people > and learn that linguists > > do not understand enough math to work it out, and even people who are > good at math would prefer > > a ready script. I wrote one and would like to share. > > I did a numeric implementation of the Elbow method for calculating the > optimal cluster number. I will appecuat it greatly if you show my work here > and people would give me feedback on it. > > https://github.com/Mathemilda/Numeric_ElbowMethod_For_K-means > > Best, > > Mya > > -- > > Maiia Bakhova > > Mathematician in Data Science > > http://myabakhova.blogspot.com > > https://www.linkedin.com/in/myabakhova > > > > _______________________________________________ > > scikit-learn mailing list > > [email protected] > > https://mail.python.org/mailman/listinfo/scikit-learn > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: < > https://mail.python.org/pipermail/scikit-learn/attachments/20201021/fa35d8ac/attachment-0001.html > > > > ------------------------------ > > Subject: Digest Footer > > _______________________________________________ > scikit-learn mailing list > [email protected] > https://mail.python.org/mailman/listinfo/scikit-learn > > > ------------------------------ > > End of scikit-learn Digest, Vol 55, Issue 5 > ******************************************* > -- Maiia Bakhova Mathematician in Data Science http://myabakhova.blogspot.com https://www.linkedin.com/in/myabakhova
_______________________________________________ scikit-learn mailing list [email protected] https://mail.python.org/mailman/listinfo/scikit-learn
