Matlab implementation of ROCK(RObust Clustering using linKs) clustering algorithm[1].
Open test_rock in MATLAB environment and test according to comments.
test_data is the data set from the original paper.
congressive-votes is the data set from the UCI Machine Learning Repository. It is the United States Congressional Voting Records in 1984. Each record corresponds to one Congress man's votes on 16 issues (e.g., education spending, crime). All attributes are boolean with Yes (that is, 1) and No (that is, 0) values, and very few contain missing values. A classification label of Republican or Democrat is provided with each data record. The data set contains records for 168 Republicans and 267 Democrats.
[1]. Guha, Sudipto, Rajeev Rastogi, and Kyuseok Shim. "ROCK: A robust clustering algorithm for categorical attributes." Information systems 25.5 (2000): 345-366, doi: 10.1016/S0306-4379(00)00022-3