The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here
Process using k means on customer segmentation
Contributor II
See my screen shot 'Process using k means.'
I ran it and to my surprise, it executed.
Result was the screen shot 'K means cluster model.'
What is this telling me?
I clicked on Plot in the side bar, got the screen shot
'K means cluster model plot.' What is this telling me?
This is the problem statement accompanying the data set.
Any help is appreciated.
I ran it and to my surprise, it executed.
Result was the screen shot 'K means cluster model.'
What is this telling me?
I clicked on Plot in the side bar, got the screen shot
'K means cluster model plot.' What is this telling me?
This is the problem statement accompanying the data set.
Any help is appreciated.
Tagged:
0
Best Answer
-
jacobcybulski
Member, University Professor Posts: 391 
Unicorn
Normally, to get good k-Means clustering you need to normalise the attributes, e.g. to the interval 0..1. If one attribute (such as D) is in the range of 0..475 and a dummy encoded attribute is in the range 0..1, the cluster chart is very difficult to interpret. Also you cluster model will be biased towards those large-valued attributes, as their values will dominate your distance measurements, which are at the core of k-Means (especially when you rely on Euclidean measures for numeric attributes, which I think is default). Also note that the presence of a large number of dummy variables will make difficult to interpret the cluster chart.
Jacob5
