Hi Experts,
I am working on the employees attrition dashboard for my company. I came to know that using k means cluster i can draw the insight from the data.
I have all the demographic information of my staff like age, work experience, nationality etc.
Can someone please guide me which insights i can withdraw and based on which clusters.
Appreciate if some resources are shared.
Its a general questions and all feedback and guidance is welcomed. I am totally new in this area.
For myself (as I imagine many other forum members), I’ve not heard of “k means cluster” before, let alone its’ desired insights. To help the forum members further analyze your current state and visualize your issue, could you please provide as many as you can of:
• Your work-in-progress PBIX file, using sanitized data if necessary
This is not something you can run directly in Power BI, but you CAN call an R or Python script within Power Query to perform this analysis, and then output the results back to a table in Power Query.
I have a two-part video series on exactly how this process works. You can find the video here:
With regard to the K-means Cluster analysis itself, this is a fairly complex machine learning method that to implement properly, you will need to train your model, evaluate performance, and then likely adjust your parameters.
One of the best resources for this is Brent Lantz’s “Machine Learning with R” (newly published 4th edition).
Chapter 9 of this book is dedicated to “Finding Groups of Data - Clustering with k-means”.
I would add that it appears from your description that the reporting
requirements are not complete. If it is inspiration which you seek,
Enterprise DNA includes a number of examples related to HR reporting.
Additionally Microsoft offers the Human Resources sample for Power BI
which would be a reasonable starting point. It is well known and has been offered to the
community for several years. It is anonymously published to the service, available
for download as a .PBIX, and the underlying dataset is available.
Due to inactivity, we’d like to conclude that you no longer require assistance to solve your inquiry.
If you have a follow question or concern related to this topic, please remove the Solution tag first by clicking the three dots beside Reply and then untick the check box.