Rocchio relevance feedback

Creator

Created

2025 Mar 10 15:18

Editor

Edited

2025 Mar 10 15:27

Refs

Optimum query Seeks to find a query that maximizes similarity to elevant documents while minimizing similarity with non-relevant documents

Let

C_r

denote set of relevant docs, Let

C_{nr}

denote set of non-relevant docs.

q_{opt} = \argmax_p [sim(q, C_r) - sim(q, C_{rn})]

q_{opt} = \frac{\sum_{d\in C_r} d_i}{|C_r|} - \frac{\sum_{d\in C_{rn}} d_i}{|C_{rn}|}

Rocchio relevance feedback is the first relevance feedback algorithm which is the part of the SMART IR system

\vec{q_m} = \alpha\, \vec{q_0} + \frac{\beta}{|D_r|} \sum_{\vec{d_j} \in D_r} \vec{d_j} - \frac{\gamma}{|D_{nr}|} \sum_{\vec{d_j} \in D_{nr}} \vec{d_j}