ABSTRAKSI: Data mining is interesting patterns and trend finding process in large database. Clustering is one of data mining functionality used for grouping objects into clusters, in which objects in the same cluster have high of similarity and high dissimilarity in different clusters.
Clustering problem is well known in the database literature for their use in numerous applications such as customer segmentation, classification and trends analysi ch such pecific subspace may also vary. Hence, the subspace clustering or projected . This final project studies and analyzes how ROCLUS algorithm clusters projected cluster in high dimensional data. eywords : data mining, PROCLUS algorithm, subspace clustering, projected lustering, high dimensional data s. In high dimensional spaces not all dimensions may be relevant to a given cluster. One way to handling this is to pick the closely correlated dimensions and find clusters in the corresponding subspace. Traditional feature selection algorithm attempts to achieve this.Kata Kunci : data mining, PROCLUS algorithm, subspace clustering, projectedABSTRACT: Data Mining adalah proses pencarian pola dan kecenderungan yang menarik dari basisdata berukuran besar. Klasterisasi merupakan salah satu fungsio ua dimensi relevan dengan kan suatu konsep yang disebut ubspace atau projected clustering, dimana subset-subset dimensi yang dipilih PROCLUS melakukan klasterisasi pada kasus ubspace clustering untuk data berdimensi tinggi. ata Kunci : data mining, algoritma PROCLUS, subspace clustering, projected lustering, data berdimensi tinggi. nalitas data mining yang dimanfaatkan untuk mengelompokkan objek-objek ke dalam klaster-klaster dimana objek-objek di dalam satu klaster yang sama mempunyai kesamaan yang tinggi dan mempunyai perbedaan yang tinggi pada objek-objek antar klaster yang berbeda.Keyword: