Abstract (EN):
Link analysis is a technique frequently used in the ranking
of web sites. On the web, we often encounter content that
is organized by entries, sorted from recent to old, and generally
follows the structure of a blog. In this paper we explore
and evaluate the usage of a bibliometrics measure, called hindex,
for the task of blog ranking, in an information retrieval
context. We base our experiments on the TREC Blogs08
collection, which comprises over 28 million posts. The results
obtained indicate that the h-index is a robust metric
that allows for an improved relevance discrimination between
blogs, when compared to the in-degree. Additionally, tests
performed using distinct versions of the post graph, indicate
that this metric might tolerate a certain level of link clutter.
Language:
English
Type (Professor's evaluation):
Scientific
No. of pages:
4
License type: