Cluster: A fast tool to identify groups of similar programs. cluster is a tool to partition a large pool of C programs into groups according to structural similarity. Its method involves calculating an alignment score for each program against a mosaic made of randomly selected code fragments of fixed size from the pool. The scores are then grouped together so that the distance between two adjacent members of a group is at most some threshold value. cluster is effective in identifying tight clusters of similar programs and is capable of distributing its workload over a network of workstations to achieve very fast running times. As a tool, cluster is highly configurable: the user can adjust its alignment scoring scheme and clustering threshold as well as obtain visual alignments of programs suspected to be similar.

References in zbMATH (referenced in 1 article , 1 standard article )

Showing result 1 of 1.
Sorted by year (citations)

  1. Carter, Casey; Tran, Nicholas: Cluster: A fast tool to identify groups of similar programs (2002)