ZURÜCK

3.4.3.1: Local Weighting Strategies

3.4.3.1.1: Term frequency

Number of appearances of a term in a document.

Rationale: the main topic of a document should cover most of its text. In this text important terms should be used frequently.

Method:

wi,j=h(i,j)

with h(i,j) denoting the frequency of term tj in document di and K[0,1]

3.4.3.1.2: Using document structure

Terms can be weighted according to the part of document they occur in.

Terms from the title or the free keyword section should be more important than terms from the body of an article.


ZURÜCK

© 1998 / HTML-Version 17. 11. 1998: R. Ferber