| 摘要 |
PROBLEM TO BE SOLVED: To suppress the registration of unnecessary information associated with the reading of an unknown work at the registering of the unknown work which is not registered on a dictionary on the dictionary together with its way of reading. SOLUTION: A document from which an unknown word is to be extracted is inputted from an input section 40 and a document access section 68 fetches a text one statement by one statement. A morpheme analyzing section 50 performs morphological analysis on the text fetched one statement by one statement and an unknown work extracting section 56 extracts an unknown word from the text, based on the analyzed results of the section 50 and, at the same time, estimates the part of speech of the unknown work and derives the way of reading of the unknown word. The way of reading is derived in such a way that morphological analysis is performed on the extracted unknown work and a possible way of reading is retrieved from a dictionary at every divided morpheme, and then, the ways of reading obtained at every morpheme are combined together. |
