Note:
Prodotto software

Descrizione sintetica del prodotto:
jP3D is a data mining software that generates frequent itemsets that are also k-anonymous. They are also called k-Anonymous Patterns. Any information inferable from frequent itemsets generated with jP3D are therefore guaranteed to refer to at least k transactions (which may represent sensitive data) of the input dataset. Notice that using a large value for minimum support does not guarantee pattern k-anonymity, since some hidden and unfrequent patterns may be inferred anyway. Please refer to this VLDB Journal paper for further information about this privacy-preserving data mining technology.

Ulteriori informazioni sul prodotto:
Requirements: - An updated Java Runtime Environment (JRE) installed in your system (freely available from Sun Microsystems for most used architectures including Windows and Linux) - a dataset containing transactions (set of items). You can test jP3D by using the following text files obtained from the UCI Mushrooms dataset: db (8K tuples). - an implementation of the well-known Apriori mining algorithm. Please download Christian Borgelt's Apriori in order to have syntax compatibility of the output (make sure to assign executable permissions). Notice that Apriori does not guarantee anonymity in the result. - MathCollection, an extension for the Java Collection Framework by Dietmar Lippold, is also required. This mathCollection.jar file must be placed in a directory called lib - jP3D software, which is boundled as a jP3D.jar file (click to download). This file must be placed in the same directory together with the lib (sub)directory.