Open AccessOFGIM: A New Algorithm to Mine Generalized-Itemsets
Author(s) -
Hussein Khafaji,
Noora Mula
Publication year - 2021
Publication title -
magallaẗ kulliyyaẗ al-rāfidayn al-ǧāmi'aẗ al-'ulūm/maǧallaẗ kulliyyaẗ al-rāfidayn al-ǧāmiʻaẗ li-l-ʻulūm
Language(s) - English
Resource type - Journals
eISSN - 2790-2293
pISSN - 1681-6870
DOI - 10.55562/jrucs.v32i2.320
Subject(s) - apriori algorithm , association rule learning , data mining , intersection (aeronautics) , pruning , computer science , affinity analysis , database transaction , hash function , algorithm , extension (predicate logic) , process (computing) , database , engineering , agronomy , computer security , biology , programming language , aerospace engineering , operating system
Most databases mined by a single layer/ crisp association rules algorithms are not flat but contain data in hierarchal/ generalized format. In spite of this fact, a few algorithms available to mine generalized itemsets to produce generalized association rules, GAR, which escort to mine more specific and concrete knowledge for decision makers. This research presents a new algorithm, (Optimized Frequent Generalized Itemsets Miner (OFGIM)), to mine generalized itemsets. Simply, it depends on extending the transactions of a database. The extension is done by adding the parent of an item to the transaction containing the item. The mining process is accomplished by the union of itemsets and the intersection of the tidsets. The algorithm requires two database scans only; the first one is for extending operation and the second scan is for mining process. The proposed algorithm does not need a specified data structure such as hash tree and prunes the apriori-based pruning steps.OFGJM was tested by using six synthetic databases. OFGJM overcomes apriori based algorithm in a ratio of ¼ in all the experiments, but it exhibitsits RAMappealing.