SIMILAR DOCUMENT RETRIEVAL SYSTEM, SIMILAR DOCUMENT RETRIEVAL METHOD AND RECORDING MEDIUM

摘要

It is an object to provide an accurate retrieval result at high speed by using a query based on a layout of a partial region in the retrieval of a desired document from a large quantity of electronic documents. A similar document retrieval system includes a storage unit (1) that stores a plurality of items and a plurality of electronic documents set in advance based on characteristics as for sizes and positions of partial regions in the electronic documents, and further stores partial region information indicative of the partial regions made to correspond to items of the partial regions out of the plurality of items and the electronic documents including the partial regions for every partial region in the plurality of electronic documents; a reception unit (106) that receives partial regions for retrieval; an extraction unit (104) that extracts characteristic information as for sizes and positions of the partial regions for retrieval; and a specification unit (109) that specifies retrieval items corresponding to the partial regions for retrieval out of a plurality of items by means of the characteristic information, and further specifies electronic documents provided with partial regions shown by the partial region information corresponding to the items for retrieval with reference to the storage unit (1).