03.04.2014 Views

Information Extraction and Retrieval - IIIT Hyderabad

Information Extraction and Retrieval - IIIT Hyderabad

Information Extraction and Retrieval - IIIT Hyderabad

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Vector-based Model<br />

• Define:<br />

– w ij > 0 whenever k i ∈ d j<br />

– w iq >= 0 associated with the pair (k i ,q)<br />

– vec(d j ) = (w 1j , w 2j , ..., w tj )<br />

– vec(q) = (w 1q , w 2q , ..., w tq )<br />

– To each term k i is associated a unitary vector vec(i)<br />

– The unitary vectors vec(i) <strong>and</strong> vec(j) are assumed to be orthonormal (i.e.,<br />

index terms are assumed to occur independently within the documents)<br />

• The t unitary vectors vec(i) form an orthonormal basis for<br />

a t-dimensional space<br />

• In this space, queries <strong>and</strong> documents are represented as<br />

weighted vectors<br />

25

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!