A first attempt to extract association rules from a database frequently yields a significant number of rules, which may be rather difficult for the user to browse in searching interesting information. However, powerful languages allow the user to specify complex mining queries to reduce the amount of extracted information. Hence, a suitable rule set may be obtained by means of a progressive refinement of the initial query. To assist the user in the refinement process, we identify several types of containment relationships between mining queries that may lead the process. Since the repeated extraction of a large rule set is computationally expensive, we propose an algorithm to perform an incremental recomputation of the output rule set. This algorithm is based on the detection of containment relationships between mining queries.
Incremental Refinement of Mining Queries / Baralis, ELENA MARIA; Psaila, G.. - STAMPA. - 1676:(1999), pp. 173-182. (Intervento presentato al convegno Data Warehousing and Knowledge Discovery First International Conference, DaWaK’99 tenutosi a Florence (ITA) nel August 30 – September 1, 1999) [10.1007/3-540-48298-9_19].
Incremental Refinement of Mining Queries
BARALIS, ELENA MARIA;
1999
Abstract
A first attempt to extract association rules from a database frequently yields a significant number of rules, which may be rather difficult for the user to browse in searching interesting information. However, powerful languages allow the user to specify complex mining queries to reduce the amount of extracted information. Hence, a suitable rule set may be obtained by means of a progressive refinement of the initial query. To assist the user in the refinement process, we identify several types of containment relationships between mining queries that may lead the process. Since the repeated extraction of a large rule set is computationally expensive, we propose an algorithm to perform an incremental recomputation of the output rule set. This algorithm is based on the detection of containment relationships between mining queries.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/1408132
Attenzione
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo