In this model, the first retrieval is executed based on the
keywords extracted from user natural language query. By
using classifier, classification information is extracted from
the upper ranked n (30) documents obtained from the first
retrieval. On the basis of the extracted classification
information, the term cluster (m) that represents each group
is generated, and then the model allows user to select the
appropriate candidate term cluster. Relevance feedback is
conducted by using the selected candidate term cluster, the
retrieval system re-retrieves the original documents, and then
provides the final results to user.