At the same time, many of the current commercial search engines use word frequency (tf-idf) to assist in ranking.
[2] Modern query expansion methods either imply document collection analysis (global or local)[3] or are dictionary- or ontology-based.
[5] Rocchio proposed to judge manually some of the retrieved documents and use this feedback information to expand the query.
Since collecting users' judgment can be challenging, only the first top retrieved documents are considered as relevant.
[6] Pseudo-relevance feedback is efficient in average but can damage results for some queries,[7] especially difficult ones since the top retrieved documents are probably non-relevant.