Summary: Finding All Maximal Frequent Sequences in Text
Helena Ahonen­Myka
Wilhelm­Schickard­Institut f¨ur Informatik,
University of T¨ubingen,
Sand 13,
D­72076 T¨ubingen, Germany
helena.ahonen@acm.org
Abstract
In this paper we present a novel algorithm
for discovering maximal frequent sequences
in a set of documents, i.e., such sequences
of words that are frequent in the document
collection and, moreover, that are not con­
tained in any other longer frequent sequence.
A sequence is considered to be frequent if it
appears in at least oe documents, when oe is
the frequency threshold given. Our approach
combines bottom­up and greedy methods,
and, hence, is able to extract maximal se­
quences without considering all the frequent

Source: Ahonen, Helena - Department of Computer Science, University of Helsinki

Collections: Computer Technologies and Information Sciences