This paper presents the implementation of DCI++, an enhancement of DCI, a scalable algorithm for discovering frequent sets in large databases. The main contribution of DCI++ resides on a novel counting inference strategy, inspired by previously known results by Basted et al. Moreover, multiple heuristics and efficient data structures are used in order to adapt the algorithm behavior to the features of the specific dataset mined and of the computing platform used. DCI++ turns out to be effective in mining both short and long patterns from a variety of datasets. We conducted a wide range of experiments on synthetic and real-world datasets, both in-core and out-of-core. The results obtained allow us to state that DCIpp performances are not over-fitted to a special case, and its high performance is maintained on datasets with different characteristics.
kDCI: a Multi-Strategy Algorithm for Mining Frequent Sets
Orlando S;Lucchese C;Palmerini P;Perego R;Silvestri F
2003
Abstract
This paper presents the implementation of DCI++, an enhancement of DCI, a scalable algorithm for discovering frequent sets in large databases. The main contribution of DCI++ resides on a novel counting inference strategy, inspired by previously known results by Basted et al. Moreover, multiple heuristics and efficient data structures are used in order to adapt the algorithm behavior to the features of the specific dataset mined and of the computing platform used. DCI++ turns out to be effective in mining both short and long patterns from a variety of datasets. We conducted a wide range of experiments on synthetic and real-world datasets, both in-core and out-of-core. The results obtained allow us to state that DCIpp performances are not over-fitted to a special case, and its high performance is maintained on datasets with different characteristics.File | Dimensione | Formato | |
---|---|---|---|
prod_91022-doc_123609.pdf
solo utenti autorizzati
Descrizione: kDCI: a Multi-Strategy Algorithm for Mining Frequent Sets
Tipologia:
Versione Editoriale (PDF)
Dimensione
236.89 kB
Formato
Adobe PDF
|
236.89 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.