Term Weighting Method Using Tag On Web Document

摘要

PURPOSE: A method for allocating a weight using a tag of a web document is provided to increase the efficiency of a web document processing application system by automatically selecting a tag weight, and by allocating the tag weight to terms found in a web document using the tag weight selected. CONSTITUTION: A weight by tags is calculated in order to allocate a term weight(501). A frequency of each tag is considered. After that, the weight by tags is adjusted using (expression 3)(502). In order to calculate a weight of a term having an HTML tag in an arbitrary document, a term weight calculation expression of (expression 2) is combined with a tag weight calculation expression of (expression 3)(503).