The web is rapidly vanishing along with our collective memory. At
least 68 web archives have been developed to cope with the web's
ephemerally problem, but despite their technology having achieved a
good maturity level, the retrieval effectiveness they provide still
presents unsatisfactory results. As result, web archives are still
"data silos" with a great potential to unfold.
In this presentation, I will discuss two different approaches to
improve the search effectiveness of web archives by exploiting the
temporal dimension of their data. The first approach models the
long-term persistence of web documents to better estimate their
relevance to a query. The second approach adapts the learning-to-rank
framework to consider the variance of web characteristics over time.