Sciweavers

DAWAK
2009
Springer

TidFP: Mining Frequent Patterns in Different Databases with Transaction ID

13 years 7 months ago
TidFP: Mining Frequent Patterns in Different Databases with Transaction ID
Since transaction identifiers (ids) are unique and would not usually be frequent, mining frequent patterns with transaction ids, showing records they occurred in, provides an efficient way to mine frequent patterns in many types of databases including multiple tabled and distributed databases. Existing work have not focused on mining frequent patterns with the transaction ids they occurred in. Many applications require finding strong associations between transaction id (e.g., certain drug) and the itemsets (e.g., certain adverse effects) to help deduce some pertinent lacking information (like how many people use this product in total) and information (like how many people have the adverse effects). This paper proposes a set of algorithms TidFPs, for mining frequent patterns with their transaction ids in a single transaction database, in a multiple tabled database, and in a distributed database. The proposed technique scans the database records only once even with level-wise Apriori-bas...
C. I. Ezeife, Dan Zhang
Added 16 Aug 2010
Updated 16 Aug 2010
Type Conference
Year 2009
Where DAWAK
Authors C. I. Ezeife, Dan Zhang
Comments (0)