Approaches to High Accuracy Retrieval: Phrase-Based Search Experiments in the HARD Track

Our main research focus this year was on the use of phrases (or multi-word units) in query expansion. Multi-word units (MWUs) comprise a number of lexical units, such as named entities (“United Nations”), nominal compounds (“amusement park”) and phrasal verbs (‘check in’). Although MWUs can belong to different parts of speech, our focus was on nominal MWUs. We used a combined syntactico-statistical approach for selecting nominal MWUs. In the first selection pass we obtained valid noun phrases, and in the second pass we used statistical measures to select strongly bound MWUs. We have experimented with using two statistical measures of selecting MWUs from text: the C-value (Frantzi and Ananiadou 1996, Vintar 2004) and the Log-Likelihood ratio (Dunning 1993). Selected MWUs were then suggested to the user for interactive query expansion. Two main research questions were studied in these experiments: