Finding authorities and hubs from link structures on the World Wide Web 论文

摘要

Recently, ther have been a number of algorH#-9 prC osed for analyzing hyperDCC link str5#TDso as todeterCDthe best "authorODHUM for a given topic or quer . While such analysis is usually combined with content analysis, ther is a sense in which some algorUO-9 ar deemed to be"mor balanced" andother "mor focused". WeunderDe a compar --C e study of hyperDTD link analysisalgor-CDCO Guided by some experD#C tal quer5CD we prD ose somefor-C crC5O r for evaluating and comparTlink analysisalgor-CUH5 Keywords link analysis, websear hing, hubs, author-9TD5 SALSA, KleinberMU algor9T#5 thrCHHE-9 Bayesian 1.