Book search apparatus, book search database generation apparatus, book search method, book search database generation method, and program
View Patent ↗Provided is an invention for efficiently searching for books used in training to reduce metathesis. A book search database storage unit 110 stores a book search database in which information identifying a book, information regarding a frequency of a proximity number for a mora character notation pair appearing in forward order in the book, and information regarding a frequency of a proximity number for the mora character notation pair appearing in reverse order are associated with each other. A display unit 300 displays a book search screen including a correctly pronounced phrase input region and an incorrectly pronounced phrase input region. A subject pronunciation designation unit 120 determines two mora character notations and a proximity number to use as search conditions based on an input of the correctly pronounced phrase input region and an input of the incorrectly pronounced phrase input region. A subject book information generation unit 130 generates search results including information regarding books matching the search conditions from the book search database.
1. A book search database generation apparatus comprising a processor configured to execute operations comprising:
generating, using a book text database in which information identifying a book and a full text of the book are associated with each other for each of a plurality of books as input, a string of morphemes through a morphological analysis of the text for each of the plurality of books;
acquiring information regarding a frequency of a proximity number for a mora character notation pair appearing in forward order and information regarding a frequency of a proximity number for the mora character notation pair appearing in reverse order from the string of morphemes of the text; and
generating a book search database associated with information identifying a book, information regarding a frequency of a proximity number for a mora character notation pair appearing in forward order in the book, and information regarding a frequency of a proximity number for the mora character notation pair appearing in reverse order in the book for each of the plurality of books.
2. A book search database generation method comprising:
generating, using a book text database in which information identifying a book and a full text of the book are associated with each other for each of a plurality of books as input, a string of morphemes through a morphological analysis of the text for each of a plurality of books; and
acquiring information regarding a frequency of a proximity number for a mora character notation pair appearing in forward order and information regarding a frequency of a proximity number for the mora character notation pair appearing in reverse order from the string of morphemes of the text; and
generating a book search database associated with information identifying a book, information regarding a frequency of a proximity number for a mora character notation pair appearing in forward order in the book, and information regarding a frequency of a proximity number for the mora character notation pair appearing in reverse order in the book for each of the plurality of books.
3. A non-transitory tangible medium having recorded thereon a program for causing a computer to function as the book search database generation apparatus according to claim 1 .