Reputation: 1
I have the following problem that is causing me a bit of a headache. I have a set of data, on this data the K-means clustering algorithm was used in order to cluster certain outlets(shops) together based upon the price of wares ordered weekly from the preferred supplier vs the price of wares ordered weekly from the competitor. There are currently 4 clusters, each of which will be labelled; "Bronze", "Silver", "Gold" and "Platinum".
So an example of the data here would be the following;
Now I have operators that need to go to these outlets and take a survey. I need to divide the outlets up, evenly, under all the operators. For example, Operator 1 needs to visit, 1 plat, 2 gold, 3 silver and 4 bronze.
Here the data looks like the following:
As to the nature of the data, I cannot really share this due to the sensitivity of the data. The end result would look like the following:
1st Line;
Line 2:
Line 3:
So if there are 1000 stores/outlets and 100 operators; each would be allocated 10 stores to visit, 1 plat, 2 gold, 3 silver and 4 bronze.
Unfortunately I have no idea where to begin, so far all I have is the operator data which includes the names of the operators. And the initial data which contain shop information like pricing and names and locations.
I do not know how I would link the data in the manner which I described.
Upvotes: -3
Views: 25