摘要:
A named entity extracting apparatus that extracts a named entity suitable for a user by enabling an order to be set in which the named entity is extracted from texts includes: an extraction order reading unit 103 that acquires a named entity pattern name stored in association with an extraction order in an extraction order storage unit 102; a named entity extracting unit 105 that extracts the named entity from input texts using a named entity pattern corresponding to the named entity pattern name acquired by the extraction order reading unit 103; and an extraction end judging unit 106 which outputs, in the case where extraction has not ended, a text on which the extraction is in progress to the extraction order reading unit 103, and continues the named entity extraction processing.
摘要:
To provide a content searching device which can efficiently present to the user a topical related keyword.A content searching device (100), which searches content from a content database with a use of a related keyword, includes: a related segment calculating unit (106) which calculates, for each content attribute, a related segment which is defined in order for first content and second content to be included in a same time segment, the related segment being calculated based on whether or not a degree of difference, for each content attribute, calculated out of a plurality of first keywords and a plurality of second keywords meets a predetermined reference value, the plurality of the first keywords each describing the first content to be stored in the content database (101), and the plurality of the second keywords each describing the second content having been stored in the content database (101); and a dictionary updating unit (107) which updates a degree of relevance stored in a dictionary database (102), the degree of relevance being updated with a use of the related segment, and the degree of relevance, among the plurality of keywords, being calculated for each content attribute.
摘要:
A named entity extracting apparatus that extracts a named entity suitable for a user by enabling an order to be set in which the named entity is extracted from texts includes: an extraction order reading unit 103 that acquires a named entity pattern name stored in association with an extraction order in an extraction order storage unit 102; a named entity extracting unit 105 that extracts the named entity from input texts using a named entity pattern corresponding to the named entity pattern name acquired by the extraction order reading unit 103; and an extraction end judging unit 106 which outputs, in the case where extraction has not ended, a text on which the extraction is in progress to the extraction order reading unit 103, and continues the named entity extraction processing.
摘要:
A related word presentation device includes a program information storage unit that stores program information of each program; and an information dividing unit that generates, for each of the attributes of the words included in the program information, at least one group which includes a reference word belonging to the attribute and a set of words which co-occur with the reference word in a program. A degree-of-relevance calculating unit stores attribute-based association dictionaries each of which indicates, for the corresponding attribute of words, (i) the words and (ii) the degrees of relevance between the words calculated based on the frequency of co-occurrence in each of groups. A search condition obtaining unit obtains the search word and the attribute; a substitute word obtaining unit selects substitute words from the attribute-based association dictionary for the obtained attribute; and an output unit presents the selected substitute word.
摘要:
An information retrieval apparatus, which can present to a user only a related word matching a user search intent, includes: an associative dictionary storage unit for storing words included in plural pieces of text to be searched and relevance degrees between the words; an appearance frequency storage unit for storing an appearance frequency that is the number of pieces of text in which the words stored in the associative dictionary storage unit appear, among the plural pieces of text to be searched; and a related word obtaining unit that obtains a related word to be presented to the user, from the relevance degree between the search word entered by the user and another word among the words, the appearance frequency, and the user search intent.
摘要:
A contents retrieval device (100) presenting an appropriate related keyword to a user even when an object user wishes to retrieve dynamically changes. The contents retrieval device (100) includes a contents estimation unit (107) retrieving contents according to a search keyword, a document space database (103) storing document spaces according to an occurrence frequency of the keyword, a document space selection unit (104) selecting a the narrowing-down document space and an expansion document space from the document space database (103) according to the search keyword and the occurrence frequency of the document space indicating a degree of relevance with the contents according to the search keyword, a related keyword estimation unit (108) selecting keywords corresponding to the narrowing-down document space and the expansion document space as a narrowing-down keyword and an expansion keyword, respectively, and an output unit displaying the selected narrowing-down and expansion keywords.
摘要:
An information retrieval apparatus, which can present to a user only a related word matching a user search intent, includes: an associative dictionary storage unit (102A) for storing words included in plural pieces of text to be searched and relevance degrees between the words; an appearance frequency storage unit (102B) for storing an appearance frequency that is the number of pieces of text in which the words stored in the associative dictionary storage unit (102A) appear, among the plural pieces of text to be searched; and a related word obtaining unit (104) that obtains a related word to be presented to the user, from the relevance degree between the search word entered by the user and another word among the words, the appearance frequency, and the user search intent.
摘要:
In an interactive program search apparatus (100) which presents search condition candidates for expanding or narrowing down search results, reason words indicating the reason why the search condition candidates are presented are adaptively determined based on user's preference, search actions, and watching actions. An association-source word extracting unit (109) extracts an association-source word from the program search results, and an associated word extracting unit (110) extracts associated words associated with the association-source word, from an association dictionary storage unit (103). A reason word extracting unit (111) extracts reason words illustrating the relationships between the association-source word and the associated words, using the association-source word, the associated words, and the obtainment history information composed of words included in the program information of the programs selected by the user in the past and selected words among the words.
摘要:
A related word presentation device (100) for appropriately performing omission prevention search includes: a program information storage unit (101) which stores program information (101a) of each program; an information dividing unit (103a) which generates, for each of the attributes of the words included in the program information (101a), at least one group which includes, as a unit, a reference word which is a word belonging to the attribute and a set of words which co-occur with the reference word in a program; a degree-of-relevance calculating unit (103b) which stores, in an association dictionary storage unit (102), attribute-based association dictionaries (102a, 102b, 102c) each of which indicates, for the corresponding attribute of words, (i) the words and (ii) the degrees of relevance between the words calculated based on the frequency of co-occurrence in each of groups; a search condition obtaining unit (104) which obtains the search word and the attribute; a substitute word obtaining unit (105) which selects substitute words from the attribute-based association dictionary for the obtained attribute; and an output unit (106) which presents the selected substitute word.
摘要:
In an interactive program search apparatus (100) which presents search condition candidates for expanding or narrowing down search results, reason words indicating the reason why the search condition candidates are presented are adaptively determined based on user's preference, search actions, and watching actions. An association-source word extracting unit (109) extracts an association-source word from the program search results, and an associated word extracting unit (110) extracts associated words associated with the association-source word, from an association dictionary storage unit (103). A reason word extracting unit (111) extracts reason words illustrating the relationships between the association-source word and the associated words, using the association-source word, the associated words, and the obtainment history information composed of words included in the program information of the programs selected by the user in the past and selected words among the words.