The DRMM(deep relevance matching model)1 employs a joint deep architecture at the query term level for relevance matching. By using matching histogram mapping, a feed forward matching network, and a term gating network, the model can effectively deal with the three relevance matching factors mentioned above. Here LCH indicates the matching histogram mapping network is log-count histogram. and IDF indicates that the term gating network use word idf to calculate the term importance weight. enter image description here

1 Jiafeng Guo, Yixing Fan, Qingyao Ai, W. Bruce Croft. A Deep Relevance Matching Model for Ad-hoc Retrieval. 2016, CIKM


DRMM.tar.bz2     Downloads  192  times