Searching the Wikipedia with contextual information
Source:
Conference on Information and Knowledge Management (CIKM) (2008)
Abstract:
We propose a framework for searching the Wikipedia with con-
textual information. Our framework extends the typical keyword
search, by considering queries of the type q, p , where q is a set
of terms (as in classical Web search), and p is a source Wikipedia
document. The query terms q represent the information that the
user is interested in finding, and the document p provides the con-
text of the query. The task is to rank other documents in Wikipedia
with respect to their relevance to the query terms q given the con-
text document p. By associating a context to the query terms, the
search results of a search initiated in a particular page can be made
more relevant.
We suggest a number of features that extend the classical query-
search model so that the context document p is considered. We then
use RankSVM (Joachims 2002) to learn weights for the individual
features given suitably constructed training data. Documents are
ranked at query time using the inner product of the feature and the
weight vectors. The experiments indicate that the proposed method
considerably improves results obtained by a more traditional ap-
proach that does not take the context into account.
Download:
