Load on a facility or a cluster center, defined to be the aggregation of all population densities of the points the cluster houses, is an important perceivable measure for facility-location problem.
Moreover, after knowing the loads on different facilities, other values like average load on a facility and standard deviation of load on each facility can be calculated to compare how different clustering algorithms are performing.
To do
For Vanilla Kmeans, the output log files are already present which contain the points in each cluster. A code needs to be written to parse through the output files and calculate the load on each facility
Description
Load on a facility or a cluster center, defined to be the aggregation of all population densities of the points the cluster houses, is an important perceivable measure for facility-location problem. Moreover, after knowing the loads on different facilities, other values like average load on a facility and standard deviation of load on each facility can be calculated to compare how different clustering algorithms are performing.
To do
For Vanilla Kmeans, the output log files are already present which contain the points in each cluster. A code needs to be written to parse through the output files and calculate the load on each facility
Example
Point -> Pop_density 1 -> 5 2 -> 10 3 -> 15 4 -> 20 5 -> 25
Cluster 1 - [3, 4, 5] Cluster 2 - [1, 2]
Load on cluster 1 = 60 Load on cluster 2 = 15