ZART : A Multifunctional Itemset Miner Algorithm
Identifieur interne : 006181 ( Main/Merge ); précédent : 006180; suivant : 006182ZART : A Multifunctional Itemset Miner Algorithm
Auteurs : Laszlo Szathmary ; Amedeo Napoli ; Sergei O. KuznetsovSource :
English descriptors
- KwdEn :
Abstract
In this paper we address two main problems in data mining, namely the extraction of frequent patterns and the generation of interesting association rules. The most well-known algorithm for finding frequent patterns is the Apriori algorithm. The main problem with Apriori and most of its variations is the high number of operations required for counting pattern supports. To overpass this drawback, a new algorithm called Pascal was proposed, that introduced the notion of pattern counting inference. Using this technique, the support of a pattern can be determined without accessing the database if its so-called key patterns are already known. We propose a new algorithm, called Zart, which is based on Pascal and extends Pascal in certain ways : it can identify frequent closed itemsets, and in parallel it can find minimal generators of closed patterns. At present, there are no algorithms that propose to extract at the same time frequent itemsets, frequent closed itemsets, and minimal generators. We show how to use Zart to generate informative association rules. The reduced informative rules represent a minimal, non-redundant set of most relevant association rules without any loss of information. With the Zart algorithm these rules can be identified very quickly and easily.
Links toward previous steps (curation, corpus...)
- to stream Crin, to step Corpus: 004027
- to stream Crin, to step Curation: 004027
- to stream Crin, to step Checkpoint: 000406
Links to Exploration step
CRIN:szathmary05aLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="261">ZART : A Multifunctional Itemset Miner Algorithm</title>
</titleStmt>
<publicationStmt><idno type="RBID">CRIN:szathmary05a</idno>
<date when="2005" year="2005">2005</date>
<idno type="wicri:Area/Crin/Corpus">004027</idno>
<idno type="wicri:Area/Crin/Curation">004027</idno>
<idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">004027</idno>
<idno type="wicri:Area/Crin/Checkpoint">000406</idno>
<idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Checkpoint">000406</idno>
<idno type="wicri:Area/Main/Merge">006181</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">ZART : A Multifunctional Itemset Miner Algorithm</title>
<author><name sortKey="Szathmary, Laszlo" sort="Szathmary, Laszlo" uniqKey="Szathmary L" first="Laszlo" last="Szathmary">Laszlo Szathmary</name>
</author>
<author><name sortKey="Napoli, Amedeo" sort="Napoli, Amedeo" uniqKey="Napoli A" first="Amedeo" last="Napoli">Amedeo Napoli</name>
</author>
<author><name sortKey="Kuznetsov, Sergei O" sort="Kuznetsov, Sergei O" uniqKey="Kuznetsov S" first="Sergei O." last="Kuznetsov">Sergei O. Kuznetsov</name>
</author>
</analytic>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>association rules</term>
<term>itemset mining</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en" wicri:score="4367">In this paper we address two main problems in data mining, namely the extraction of frequent patterns and the generation of interesting association rules. The most well-known algorithm for finding frequent patterns is the Apriori algorithm. The main problem with Apriori and most of its variations is the high number of operations required for counting pattern supports. To overpass this drawback, a new algorithm called Pascal was proposed, that introduced the notion of pattern counting inference. Using this technique, the support of a pattern can be determined without accessing the database if its so-called key patterns are already known. We propose a new algorithm, called Zart, which is based on Pascal and extends Pascal in certain ways : it can identify frequent closed itemsets, and in parallel it can find minimal generators of closed patterns. At present, there are no algorithms that propose to extract at the same time frequent itemsets, frequent closed itemsets, and minimal generators. We show how to use Zart to generate informative association rules. The reduced informative rules represent a minimal, non-redundant set of most relevant association rules without any loss of information. With the Zart algorithm these rules can be identified very quickly and easily.</div>
</front>
</TEI>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 006181 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Merge/biblio.hfd -nk 006181 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Lorraine |area= InforLorV4 |flux= Main |étape= Merge |type= RBID |clé= CRIN:szathmary05a |texte= ZART : A Multifunctional Itemset Miner Algorithm }}
![]() | This area was generated with Dilib version V0.6.33. | ![]() |