Levelwise search of frequent patterns with counting inference
Identifieur interne : 002C92 ( Crin/Corpus ); précédent : 002C91; suivant : 002C93Levelwise search of frequent patterns with counting inference
Auteurs : Yves Bastide ; Rafik Taouil ; Nicolas Pasquier ; Gerd Stumme ; Lotfi LakhalSource :
English descriptors
- KwdEn :
Abstract
In this paper,we address the problem of the efficiency of the main phase of most data mining applications : The frequent pattern extraction. This problem is mainly related to the number of operations required for counting pattern supports in the database, and we propose a new method called pattern counting inference, that allows to perform as few support counts as possible. Using this method, the support of a pattern is determined without accessing the database whenever possible, using the supports of some of its sub-patterns called key patterns. This method was implemented in the Pascal algorithm that is an optimization of the simple and efficient Apriori Algorithm. Experiments comparing Pascal to the Apriori, Close and Max-Miner algorithms, each one representative of a frequent patterns discovery strategy, show that Pascal improves the efficiency of the frequent pattern extraction from correlated data and that it does not induce additional execution times when data is weakly correlated.
Links to Exploration step
CRIN:bastide00aLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="120">Levelwise search of frequent patterns with counting inference</title>
</titleStmt>
<publicationStmt><idno type="RBID">CRIN:bastide00a</idno>
<date when="2000" year="2000">2000</date>
<idno type="wicri:Area/Crin/Corpus">002C92</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">Levelwise search of frequent patterns with counting inference</title>
<author><name sortKey="Bastide, Yves" sort="Bastide, Yves" uniqKey="Bastide Y" first="Yves" last="Bastide">Yves Bastide</name>
</author>
<author><name sortKey="Taouil, Rafik" sort="Taouil, Rafik" uniqKey="Taouil R" first="Rafik" last="Taouil">Rafik Taouil</name>
</author>
<author><name sortKey="Pasquier, Nicolas" sort="Pasquier, Nicolas" uniqKey="Pasquier N" first="Nicolas" last="Pasquier">Nicolas Pasquier</name>
</author>
<author><name sortKey="Stumme, Gerd" sort="Stumme, Gerd" uniqKey="Stumme G" first="Gerd" last="Stumme">Gerd Stumme</name>
</author>
<author><name sortKey="Lakhal, Lotfi" sort="Lakhal, Lotfi" uniqKey="Lakhal L" first="Lotfi" last="Lakhal">Lotfi Lakhal</name>
</author>
</analytic>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>algorithmsa</term>
<term>data mining</term>
<term>frequent patterns extraction</term>
<term>key patterns</term>
<term>pattern counting inference</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en" wicri:score="2477">In this paper,we address the problem of the efficiency of the main phase of most data mining applications : The frequent pattern extraction. This problem is mainly related to the number of operations required for counting pattern supports in the database, and we propose a new method called pattern counting inference, that allows to perform as few support counts as possible. Using this method, the support of a pattern is determined without accessing the database whenever possible, using the supports of some of its sub-patterns called key patterns. This method was implemented in the Pascal algorithm that is an optimization of the simple and efficient Apriori Algorithm. Experiments comparing Pascal to the Apriori, Close and Max-Miner algorithms, each one representative of a frequent patterns discovery strategy, show that Pascal improves the efficiency of the frequent pattern extraction from correlated data and that it does not induce additional execution times when data is weakly correlated.</div>
</front>
</TEI>
<BibTex type="inproceedings"><ref>bastide00a</ref>
<crinnumber>A00-R-434</crinnumber>
<category>3</category>
<equipe>LIMOS</equipe>
<author><e>Bastide, Yves</e>
<e>Taouil, Rafik</e>
<e>Pasquier, Nicolas</e>
<e>Stumme, Gerd</e>
<e>Lakhal, Lotfi</e>
</author>
<title>Levelwise search of frequent patterns with counting inference</title>
<booktitle>{Bases de Données Avancées - BDA'00, Blois}</booktitle>
<year>2000</year>
<month>Oct</month>
<url>http://www.loria.fr/publications/2000/A00-R-434/A00-R-434.ps</url>
<keywords><e>data mining</e>
<e>frequent patterns extraction</e>
<e>pattern counting inference</e>
<e>key patterns</e>
<e>algorithmsa</e>
</keywords>
<abstract>In this paper,we address the problem of the efficiency of the main phase of most data mining applications : The frequent pattern extraction. This problem is mainly related to the number of operations required for counting pattern supports in the database, and we propose a new method called pattern counting inference, that allows to perform as few support counts as possible. Using this method, the support of a pattern is determined without accessing the database whenever possible, using the supports of some of its sub-patterns called key patterns. This method was implemented in the Pascal algorithm that is an optimization of the simple and efficient Apriori Algorithm. Experiments comparing Pascal to the Apriori, Close and Max-Miner algorithms, each one representative of a frequent patterns discovery strategy, show that Pascal improves the efficiency of the frequent pattern extraction from correlated data and that it does not induce additional execution times when data is weakly correlated.</abstract>
</BibTex>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Crin/Corpus
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 002C92 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Crin/Corpus/biblio.hfd -nk 002C92 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Lorraine |area= InforLorV4 |flux= Crin |étape= Corpus |type= RBID |clé= CRIN:bastide00a |texte= Levelwise search of frequent patterns with counting inference }}
![]() | This area was generated with Dilib version V0.6.33. | ![]() |