| R. Sibson. Slink: An optimally e#cient algorithm for the single-link cluster method. The Computer Journal, 16:30--34, 1973. |
....T 1 and T 2 . Other measures are used and they have a big e#ect on the computational load. After the full generation of the dendrogram, it is pruned down to a desired level. The single link algorithm computes all the distances but does not require to store them all. The classical SLINK [69] algorithm requires O(N time and O(N) space. Figure 5.3 shows the dendrogram obtained using hierarchical agglomerative clustering for the iris data set. One can see that again the clusters obtained are almost identical with the classes of the iris. In the case of a database which was ....
R. Sibson. Slink: An optimally e#cient algorithm for the single-link cluster method. The Computer Journal, 16:30--34, 1973.
....that if all the weights are strictly positive, a tree with the maximum number of edges and p = 1 connected components will result. If some of the weights W uv are zero, it is possible to obtain trees with more than one connected component. More sophisticated MWST algorithms exist (see for example [10, 63, 25, 20, 58]) and they improve on Kruskal s algorithm on both running time and memory requirements. However, the running time of all published algorithms is at least proportional to the number of candidate edges ( E ) In our case, this number is equal to n(n 1) 2 since all pairs of variables have to be ....
R. Sibson. SLINK: an optimally e#cient algorithm for the single-link cluster method. The Computer Journal, 16, 1973.
No context found.
R. Sibson. Slink: An optimally e#cient algorithm for the single-link cluster method. The Computer Journal, 16:30--34, 1973.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC