Related Posts
Hi folks, I have 6.5yrs of experience in Java, Springboot, Microservices. Does anyone have any idea about Optum Global Solutions, Hyderabad. I was told that the Project which I will be assigned to is an IVR application responsible for middleware services and the shift will be from 2 pm-12 am. Any idea which project is it going to be and how is the work going to be? Do they expect us to extend beyond 12 am on a regular basis?
Optum Optum Global Solutions UnitedHealth Group UHG
More Posts
How do you find in-house positions?
Deloitte FDD vs PwC FDD? NYC
Additional Posts in Excel Genius
Anyone know what these badges are?

New to Fishbowl?
unlock all discussions on Fishbowl.







R or Python would be better suited for this task. In addition you’ll probably have better luck in the data & analytics bowl.
In my (very unprofessional) opinion it sounds like you’re trying to do cluster analysis. Hopefully this thread puts you in the right direction: https://datascience.stackexchange.com/questions/22/k-means-clustering-for-mixed-numeric-and-categorical-data
Thank you!
I found some answers on other forums. For those having the same problem, this is a list of leads:
- kmeans but only with numeric data
- kmodes with mixed data
- if all data is binary use and each property only occurs once, use Jaccard index
- if a property can occur several times and frequency is important use cosine index
- if you want to compare continuous data, use euclidian distance
- if you have mixed data, you can also use Gower’s general similarity measure
Just FYI, i’ve tried turning all data into categorical so i defined some ranges for the numeric and i did a Euclidian distance calculation. It’s okayish but it doesn’t really factor automatically that some categories influence more the pricing than others. I added manual weights on each category but not sure how to calibrate all of this.
Why TF would anyone give this away for free?
To prove you’re an Excel genius, obviously.
Why are you doing this? Maybe submit it to alyx, or the lumen.audit.pm email if it's audit related
QI Macros statistical testing across multiple variables 🤷♀️