Index Weighting Based on the Weight of the Index of Each Tag

Abstract

There is a study in progress that extracts an index using the structure and contents of a document for achieving more effective index extraction and index weighting. Most studies calculate the relevance of the context and determine the index according to the weights of each tag. But, these previous studies are determining the relevance according to simple parameters from general theory, rather than proving relevance via objective experiments. This study classifies a thesis in which the main paragraphs were converted into XML, via XML structural information, and proposes a technique that determines the weight of the final index while updating the weight of the index according to the weight of terms, subsequent to calculating the weight for the term in each paragraph.