In which wi is the weight assigned for the ithcomponent
of document d, i is the index of that component and the set
of the index of all components in which k appear defined
asA={x/nx (k,d)>0}. Parameter a=max(wi | i∈A) is the
weight of the most important component where k appears,
also serves as the predefined minimum value for ip(k, d.).
The number of a document’s component and the weight for
each component is different for each type of document. A paper, for example, contains title, abstract, keyword, main
content and reference, with title and abstract often have the
largest weight. With this new formula, we correct a
problem in previous work and ensure that a keyphrase
which appears in title as well as abstract will have higher ip
than a keyphrase appear only in title.
In which wi is the weight assigned for the ithcomponentof document d, i is the index of that component and the setof the index of all components in which k appear definedasA={x/nx (k,d)>0}. Parameter a=max(wi | i∈A) is theweight of the most important component where k appears,also serves as the predefined minimum value for ip(k, d.).The number of a document’s component and the weight foreach component is different for each type of document. A paper, for example, contains title, abstract, keyword, maincontent and reference, with title and abstract often have thelargest weight. With this new formula, we correct aproblem in previous work and ensure that a keyphrasewhich appears in title as well as abstract will have higher ipthan a keyphrase appear only in title.
การแปล กรุณารอสักครู่..
