K-Means Clustering in Python - Clients Segmentation with Unsupervised Learning

[@RDataScience]

Data Science Guide, Cheat Sheets and Free eBooks here: github.com/rocketingdatascience/RDS_Guide

[@zhalehmohammadalipour3542]

Is there a way to show centroids in 3D clusters? If so, can you explain how?

[@RDataScience]

Hey Zhaleh, I'm sorry for this late response. Yes it is! You would need to add the centroids coordinates like this: centroids_data = go.Scatter3d( x=centroids[:, 0], y=centroids[:, 1], z=centroids[:, 2], mode='markers', marker=dict(color='black', size=10, symbol='diamond') ) and then add it to the data variable: data = [trace, centroids_data ]

[@nuzailahnoor2069]

Can you please explain what is happening with the lambda function in the visualizing cluster section?

[@RDataScience]

Hello Nuzailah, I'm sorry for the late reply. I'm assigning one color o each cluster in the dataframe. So I can use these colors when building the vizualisation.

[@pcpc8817]

Great video, but I follow the codes, my scatter graph shows 5 centroids horizontally at 50 level in y axis, I have no idea why, any advice? Many thanks!

[@RDataScience]

Thank you ! Well, it might be hard to find something without watching the code. You can go to my GitHub repo and check the code. Or if you want you can share yours from GitHub here

[@MarianneHMiettinen]

great, except that it could be zoomed to be easier to see

[@RDataScience]

Thank you for the feedback, I’m using zoom more frequently in recent videos 👍

[@os4764]

drop the background music for god's sake, it's distracting

[@RDataScience]

Thank you for the feedback. I'm not using anymore.

[@santiagoruge]

in the real world you need to work with the gender too... why did you drop it?

[@RDataScience]

That's right! You can also get insights about the gender. The goal here was to work only with the numerical data and get insights about the remaining variables. You made a very good question, that critical thinking is essential in this field! If you've followed the tutorial, feel free to proceed with the gender and analyse how different insights are