HIERARCHICAL CLUSTERING STUDY ON BUSES MAINTANANCE COST
Cluster as one of method from data mining is a method that grouping variable or object then we could know what the characteristic of objects or variable is. We could use various method in cluster analysis. Like hierarchical method has different method for distance method and linkage method. The dist...
Saved in:
| Main Author: | |
|---|---|
| Format: | Final Project |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/22100 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | Cluster as one of method from data mining is a method that grouping variable or object then we could know what the characteristic of objects or variable is. We could use various method in cluster analysis. Like hierarchical method has different method for distance method and linkage method. The distance method for qualitative data is Eucliden, Mahalanobis, Chebyshev, and Cosinus distance. The linkage method single, complete, and average linkage. There is no method that known as best method in cluster analysis. In this final project we will choosing the best method from those metods. We will use using cophenetic correlation coefficient index for choosing the best method. The object that will be clustered is buses based on its maintenance cost for 15 months. Based on the advantage and deficiency of distance method, Euclidean method is the appropriate method for counting distance between the busses. The best method for linkage method is average linkage. |
|---|