JP2008027290A - Japanese sentence creation support method and apparatus - Google Patents

Abstract

It is possible to create an appropriate Japanese sentence including a noun for which a Japanese sentence has not been sufficiently created in the past, such as a newly created technical term.
A verb type corresponding to an input verb is acquired from a verb dictionary by a verb type acquiring unit, and a noun corresponding to the verb type and a role of the noun are combined with a noun and noun role acquiring unit. Is obtained from the verb type list 4, and the noun complements that can be combined for each particle are obtained from the noun dictionary 3 by the noun dictionary 3 using the particle and the role as a key, and the candidate sentence creating means 10 is used for each particle. The candidate strings including the noun candidates acquired in the above are arranged in a prescribed order together with the corresponding particles, and the input verbs are arranged last to create a candidate sentence, which is presented on the display unit 2, by the Japanese sentence output means 11. Of the noun candidates in the candidate column for each particle, only the selected noun candidate is combined with the corresponding particle to create and output a Japanese sentence.
[Selection] Figure 1

Description

  The present invention relates to a technique for creating a Japanese sentence (electronic data thereof) in an interactive format based on an input operation or a selection operation from a user in a computer system or a predetermined apparatus.

  Conventionally, the simplest method for creating a Japanese sentence in a computer system (computer) is a method in which a user inputs characters corresponding to all the Japanese sentences to be created using a keyboard or a well-known kana-kanji conversion program. However, there is a problem that the burden on the user is extremely large.

  Therefore, a large amount of Japanese (character strings) is stored separately for each of the basic part-of-speech verbs, nouns, and particles that make up the Japanese sentence, and part or all of these Japanese parts are part of the part-of-speech. A method of creating a Japanese sentence by presenting to a user via a display unit in a prescribed order according to the type, allowing the user to select any Japanese for each part of speech via a keyboard, etc., and combining them was suggested.

  However, in this method, although the order of each Japanese that composes a Japanese sentence can be the prescribed order according to the type of part-of-speech described above, the combination of nouns and verbs and the combination of nouns and particles Since there was no information or restrictions, it was possible to select and combine them arbitrarily, so the burden on users who need to select appropriate Japanese is not small, and it is actually inappropriate Japanese sentences could be created.

In the above-described method, in order to reduce the burden on the user and create only an appropriate Japanese sentence, it is necessary to give some information and restrictions to support the combination of nouns and verbs and the combination of nouns and particles. One such technique is to statistically process the relationship between nouns and verbs based on a natural language corpus created in the past, and to preferentially present complements that are likely to be linked. Has been proposed (see Patent Document 1).
JP 2002-149653 A "Document database creation method and storage medium storing document database creation program"

  However, in the case of the above-mentioned proposal, sufficient samples cannot be obtained for nouns for which Japanese sentences have not been sufficiently prepared in the past, such as newly created technical terms. There was a drawback that it could not be done.

  The present invention has been made in view of the above circumstances, and the purpose of the present invention is to add a verb or noun to a noun for which a Japanese sentence has not been sufficiently created in the past, such as a newly created technical term. It is possible to provide support for combinations with particles, and to create appropriate Japanese sentences including such nouns.

  The Japanese sentence creation support method according to claim 1, wherein, together with the input part and the display part, for each noun used in the Japanese sentence, at least one particle that can be combined with the noun and at least one role of the noun Categorization of noun dictionaries registered in Japanese, verbs used in Japanese sentences, according to the number of nouns required in the Japanese sentence containing the verb, the particles that are combined with each noun, and the role of each noun For a plurality of verb types that are the types of verbs, the verb type list that registers the roles of each noun and each noun in the prescribed order, and for each verb used in a Japanese sentence, A computer having a verb dictionary in which at least one verb type is registered is used, and the computer searches the verb dictionary using a verb input from the user via the input unit as a key. The verb type corresponding to the input verb is acquired, the particle and the role of each noun corresponding to each noun corresponding to the acquired verb type are acquired from the verb type list, and the acquired particle and role are used as keys. The noun dictionary is searched, and the noun complements that can be combined with the particles are obtained for each particle, and the candidate strings including all the noun candidates obtained for each particle are arranged in the prescribed order together with the corresponding particles. Arranging the input verbs at the end to create a candidate sentence and presenting it to the user via the display unit; among the noun candidates in the candidate string for each presented particle, from the user via the input unit Each of the selected noun candidates is combined with the corresponding particle, and the input verb is combined at the end to create and output a Japanese sentence.

  With this configuration, it is determined whether the input verb is one of multiple verb types classified according to the number of nouns required in the Japanese sentence, the particles combined with each noun, and the role of each noun. Identify and acquire the role of each noun corresponding to each noun corresponding to the specified verb type, obtain the complement of the noun that can be combined with the relevant particle for each particle, and acquire the noun obtained for each particle Candidate strings containing all of the candidates are arranged together with the corresponding particles and entered verbs to create candidate sentences and present them to the user, so Japanese sentences are sufficiently created in the past like newly created technical terms. Even if the noun is not a noun, it is not considered as a noun candidate for a non-conforming verb in view of its role and connectable particles, and the creation of inappropriate Japanese sentences can be suppressed.

  In addition, the Japanese sentence creation support method according to claim 2 includes, together with the input unit and the display unit, for each noun used in the Japanese sentence, at least one particle that can be combined with the noun and at least one that the noun has. A noun dictionary that registers roles and at least one field in which the noun is used, verbs used in Japanese sentences, the number of nouns required in the Japanese sentence containing the verb, and Verb type that registers the particle and the role of each noun in the prescribed order for the plural verb types that are classified according to the particle that is combined with the noun and the role of each noun. A list, a verb dictionary in which at least one verb type is registered for each verb used in a Japanese sentence, and field information indicating the use frequency of each noun in the Japanese sentence are stored. Using a computer provided with a field information storage unit, the computer searches the verb dictionary using the verb input from the user via the input unit as a key, and acquires the verb type corresponding to the input verb, The noun corresponding to the acquired verb type and the role of each noun are acquired from the verb type list, the noun dictionary is searched using the acquired particle and role as a key, and nouns that can be combined with the noun Accompaniment is acquired for each particle along with its field, candidate sequences including all of the noun candidates acquired for each particle are arranged in the prescribed order together with the corresponding particles, and the input verb is arranged last. A candidate sentence is created, and the display of each candidate noun for each candidate string in the candidate sentence is in a form according to the field of the noun candidate and the field information stored in the field information storage unit. The above-mentioned noun candidates are presented to the user via the display unit, and among the noun candidates in the presented candidate strings, only the noun candidates respectively selected from the user via the input unit and corresponding nouns respectively. The combined verbs are finally combined to create and output a Japanese sentence, and the field information stored in the field information storage unit is updated based on the field of the selected noun.

  With this configuration, when presenting the created candidate sentence to the user, the display of each candidate noun for each candidate column is in a form according to the field of the candidate noun candidate and the field information stored in the field information storage unit, for example, When the field information includes content indicating that the frequency of use of nouns in a specific field is high, the noun candidates for the specific field are presented to the user in a form that emphasizes the specific field. Japanese sentences can be created by preferentially selecting nouns.

  Further, the Japanese sentence creation support method according to claim 3 is the Japanese sentence creation support method according to claim 1 or 2, wherein a plurality of verb types are obtained from a verb dictionary for a verb input from a user via an input unit. If acquired, similarly for each verb type, the role of particles and nouns are acquired, noun candidates are acquired, candidate sentences are respectively created and presented to the user via the display unit, Among the noun candidates in the candidate string for each particle of each candidate sentence, only the noun candidates in the one candidate sentence each selected from the user via the input unit are combined with the corresponding particle and inputted. It is characterized by creating and outputting Japanese sentences by combining verbs at the end.

  With this configuration, even if there are a plurality of verb types corresponding to the input verb, a candidate sentence for each verb type can be created and presented to the user, and the nouns in one candidate sentence selected by the user can be displayed. A Japanese sentence can be created and output by combining only candidates with corresponding particles.

  Further, the Japanese sentence creation support method according to claim 4 is the Japanese sentence creation support method according to any one of claims 1 to 3, wherein two or more noun candidates are selected from noun candidates corresponding to one particle. When selected, each noun candidate is connected using a character string “to” and combined with the particle to create and output a Japanese sentence.

  With this configuration, a Japanese sentence can be created and output even when two or more nouns are selected for one particle.

  According to the Japanese sentence creation support apparatus described in claims 5 to 8, it is possible to provide an apparatus capable of executing Japanese sentence creation support by the above-described Japanese sentence creation support method described in claims 1 to 4.

  According to the present invention, a noun that can be combined, its role, or in addition to a noun for which a Japanese sentence has not been sufficiently created in the past, such as a newly created technical term, is used. By creating a noun dictionary that registers the subject areas, you can automatically create a Japanese sentence by automatically presenting appropriate supplements for noun candidate extraction and particle setting when creating a Japanese sentence. It is possible to support and improve the efficiency of creating Japanese sentences.

  Further, in the present invention, the combination of nouns and verbs and the combination of nouns and particles is not statistical, but strictly defined based on the typology of verbs. Therefore, the present invention is more reliable than a method using a conventional corpus. Highly supportive is possible.

  Hereinafter, the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows an example of an embodiment of a Japanese sentence creation support apparatus according to the present invention. In the figure, 1 is an input unit, 2 is a display unit, 3 is a noun dictionary, 4 is a verb type list, and 5 is a verb dictionary. , 6 is a field information storage unit, 7 is a verb type acquisition means, 8 is a particle and noun role acquisition means, 9 is a noun candidate acquisition means, 10 is a candidate sentence creation means, 11 is a Japanese sentence output means, and 12 is a field information update. Means.

  The input unit 1 includes a known keyboard and mouse, and sends information based on an input operation and a selection operation from the user to each means. It is assumed that the input unit 1 has a Japanese input function including kanji by a known kana-kanji conversion program. The display unit 2 includes a well-known liquid crystal or CRT display, and displays (presents) information input or selected from the input unit 1 and results of processing by other means.

  As shown in FIG. 2, the noun dictionary 3 includes, for each noun used in a Japanese sentence, at least one particle that can be combined with the noun, at least one role (attribute) that the noun has, At least one field to be used is registered. For example, the noun “operator” can be combined with the particles “ga” and “ha”, has the role “person”, and is used in the field “communication”.

  As shown in FIG. 3, the verb type list 4 includes the verbs used in the Japanese sentence, the number of nouns required in the Japanese sentence including the verb, the particles combined with each noun, For a plurality of verb types, which are types classified according to roles, the particle and the role of each noun that are combined with each noun are specified in the prescribed order, in this case “noun 1 (subject)”, “noun 2 (object) 1) "," noun 3 (object 2) ", and so on. For example, the verb “TYPE: 1” requires one noun, “noun 1 (subject)”, which indicates that the noun must be combined with the particle “ga” and have the role “thing”. ing.

  As shown in FIG. 4, the verb dictionary 5 registers at least one verb type for each verb label used in a Japanese sentence, here, a verb label that is a final character string. For example, it is indicated that the verb type of the verb label “walk” is “TYPE: 1”.

  The noun dictionary 3, the verb type list 4 and the verb dictionary 5 are created in advance by hand and stored in a storage device (not shown). The contents of each dictionary and the like are updated manually, that is, contents are added or changed as necessary.

  The field information storage unit 6 stores field information indicating the frequency of use of each noun field in a Japanese sentence, here a Japanese sentence created by the device, for example, a numerical value “1” every time a noun is used (selected). The count value for each field to which is added is stored.

  The verb type acquisition means 7 searches the verb dictionary 5 using the verb (verb label) input from the user via the input unit 1 as a key, and acquires the verb type corresponding to the input verb. The particle and noun role acquisition means 8 acquires from the verb type list 4 the particle and the role of each noun that are combined with each noun corresponding to the verb type acquired by the verb type acquisition means 7. The noun candidate acquisition means 9 searches the noun dictionary 3 using the particle and role acquired by the particle and noun role acquisition means 8 as a key, and acquires the complement of the noun that can be combined with the particle together with its field.

  The candidate sentence creation means 10 includes a candidate sequence including all of the noun candidates acquired for each particle by the noun candidate acquisition means 9 together with the corresponding particles acquired by the particle and the noun role acquisition means 8 in a prescribed order, here {Noun 1 candidate 1 | Noun 1 candidate 2 | Noun 1 candidate 3 ...} [Noun 1 noun 1], {Noun 2 candidate 1 | Noun 2 candidate 2 | Noun 2 candidate 3 ...} [Noun 2 Noun 3 candidate 1 | noun 3 candidate 2 | noun 3 candidate 3 ...} [noun 3 particle 3], ... in this order, and the input verbs are arranged last to create a candidate sentence. Of the candidate sentences, the display of each candidate noun for each candidate string is displayed in a form corresponding to the field of the noun candidate and the field information stored in the field information storage unit 6, for example, the field with the highest use frequency. The noun candidate is included in the top of the candidate column and displayed in a larger character. It presented to the user via the display section 2 above.

  The Japanese sentence output unit 11 corresponds only to the noun candidates selected from the user via the input unit 1 among the noun candidates in the candidate string for each particle presented to the user via the display unit 2. Each is combined with a particle, and the input verb is finally combined to create and output a Japanese sentence. When two or more noun candidates are selected from the noun candidates corresponding to one particle via the input unit 1, the noun candidates are connected using the character string “to”. Then, combine with the particle and create and output a Japanese sentence.

  The field information update unit 12 updates the field information stored in the field information storage unit 6 based on the field of the noun selected by the user via the input unit 1 in the Japanese sentence output unit 11, for example, the selected noun “1” is added to the above-mentioned count value for each field corresponding to the field.

  When there are a plurality of verb types acquired by the verb type acquiring means 7, the particle and noun role acquiring means 8, the noun candidate acquiring means 9, and the candidate sentence creating means 10 are similar to each verb type. The role of the noun is acquired, the noun candidates and their fields are acquired, candidate sentences are respectively created and presented to the user via the display unit 2, and the Japanese sentence output means 11 provides the user with the display unit 2. Of the noun candidates in the candidate strings for each particle of each candidate sentence presented, only the noun candidates in one candidate sentence respectively selected from the user via the input unit 1 are combined with the corresponding particles, The input verb is finally combined to create and output a Japanese sentence.

  FIG. 5 shows the flow of processing in the above-described Japanese sentence creation support apparatus. Hereinafter, the operation will be described in detail with an example.

  First, when a user inputs a verb used in a Japanese sentence from the input unit 1 with the verb label (s1), the verb type acquisition means 7 searches the verb dictionary 5 using the input verb label as a key, The verb type corresponding to the made verb label is acquired (s2).

  For example, when the input verb label is “acquire”, the verb type “TYPE: 4, 5” is acquired as can be understood from FIG. 4 showing the contents of the verb dictionary 5.

  At this time, if no verb type can be acquired because the record corresponding to the input verb label is not in the verb dictionary 5 or the like, the process ends as an error (s3, s4).

  Next, the particle and noun role acquisition unit 8 acquires from the verb type list 4 the particle and the role of each noun that are combined with each noun corresponding to the verb type acquired by the verb type acquisition unit 7 (s5).

  Here, as can be understood from FIG. 3 showing the contents of the verb type list 4, the nouns corresponding to the noun corresponding to the verb type “TYPE: 4” and the role of each noun are “noun 1: particle [ ”] Role [People]”, “Noun 2: Particle [O] Role [Things]”, “Noun 3: Particle [From] Role [Place]” are acquired, and corresponds to the verb type “TYPE: 5” The noun 1: particle [ga] role [person], “noun 2: particle [o] role [thing]”, “noun 3: particle [de]” Role [Location] "is acquired.

  Next, the noun acquisition unit 9 searches the noun dictionary 3 using the particle and role acquired by the particle and noun role acquisition unit 8 as a key, and acquires the complement of the noun that can be combined with the particle together with the field. (S6).

  Here, as can be understood from FIG. 2 showing the contents of the noun dictionary 3, noun candidates corresponding to the verb type “TYPE: 4” and corresponding to “noun 1: particle [ga] role [person]”. In addition, “noun [operator] field [communication]”, “noun [section manager] field [distribution]”, “noun [distribution system] field [distribution]” are acquired, and “noun 2: particle” [No] Nouns corresponding to the role [thing] and their fields include “noun [package] field [distribution]”, “noun [logistics system] field [distribution]”, “noun [setting value] field [ Communication, distribution] "is acquired, and nouns corresponding to" noun 3: particle [kara] role [place] "and their fields include" noun [warehouse] field [distribution] "," noun [database] ] And [communication, distribution] ”are acquired.

  Similarly, noun candidates corresponding to the verb type “TYPE: 5” corresponding to “noun 1: particle [ga] role [person]” and their fields are “noun [operator] field [communication]”, “Noun [Manager Tanaka] field [Distribution]”, “Noun [Logistics system] field [Distribution]” is acquired, and noun candidates corresponding to “Noun 2: Particle [O] Role [thing]” and its As the fields, “noun [package] field [distribution]”, “noun [logistics system] field [distribution]”, “noun [setting value] field [communication, distribution]” are acquired, and “noun 3: particle” “Noun [warehouse] and field [distribution]”, “noun [database] and field [communication, distribution]” are acquired as noun candidates and fields corresponding to [de] and role [place].

  Next, the candidate sentence creating means 10 includes candidate strings including all of the noun candidates acquired for each particle by the noun candidate acquiring means 9 together with the corresponding particles acquired by the noun and noun role acquiring means 8, respectively. Candidate sentences are arranged in order and the input verbs are arranged last (s7), field information is obtained from the field information storage unit 6 (s8), and each noun candidate for each candidate string in the candidate sentence. Is displayed in a form corresponding to the field of the noun candidate and the field information, in which the noun candidate including the most frequently used field is displayed at the top of the candidate string and displayed in a larger character. It is presented to the user via the display unit 2 (s9).

  For example, if the count value of the field “communication” is the maximum among the count values for each field of the field information stored in the field information storage unit 6, the candidate sentence corresponding to the verb type “TYPE: 4” ( For Candidate 1), among the noun candidates “Operator”, “Director Tanaka” and “Logistics system” corresponding to “Noun 1: particle [ga] role [person]”, the field includes “communication”. Among the “nouns”, the noun candidates “package”, “distribution system”, and “setting value” corresponding to “noun 2: particle [a] role [thing]” are “communication” in the field. Only the “set value” is included, and among the noun candidates “warehouse” and “database” corresponding to “noun 3: particle [kara] role [place]”, the field includes “communication”. Is only a “database”, so it appears in large letters at the beginning of each candidate column. Which is given to, as shown in FIG. 6, "Operator", "Settings", and "database". Similarly, for the candidate sentence (candidate 2) corresponding to the verb type “TYPE: 5”, large characters are displayed at the top of each candidate column as shown in FIG. Set value ”and“ database ”.

  If there are multiple noun candidates that contain the most frequently used fields, they are displayed in the order they were acquired from the top, and there are no noun candidates that contain the most frequently used fields, or field information When there is no difference in the count value for each field of the field information stored in the storage unit 6, the noun candidates are displayed as they are in the order of acquisition.

  Next, when the user selects one of the noun candidates in the candidate string for each particle of each candidate sentence presented on the display unit 2 from the input unit 1 (s10), the Japanese sentence output unit 11 However, only the noun candidates in the selected one candidate sentence are combined with the corresponding particles, respectively, and the input verb is finally combined to create and output a Japanese sentence (s11).

  For example, when the user selects “operator”, “setting value”, and “database” in the candidate sentence (candidate 1) corresponding to the verb type “TYPE: 4”, “the operator acquires the setting value from the database”. Japanese sentences are created and output.

  Finally, the field information update unit 12 updates the field information stored in the field information storage unit 6 based on the field of the noun selected by the user via the input unit 1 in the Japanese sentence output unit 11. Then, “1” is added to the count value for each of the above-mentioned fields corresponding to the field of the designated noun (s12).

  That is, here, the count value of the field “communication” is added by “1” to the selection of the noun “operator”, and the fields “communication” and “distribution” are added to the selection of the noun “setting value”. The count value is incremented by “1”, and the count values of the fields “communication” and “distribution” are incremented by “1” for the selection of the noun “database”.

[Other embodiments]
In the embodiment described above, the configuration and processing related to the field in which nouns are used may be omitted.

  That is, in the apparatus configuration of FIG. 1, information corresponding to the field is deleted from the noun dictionary 3, and the field information storage unit 6 and the field information update unit 12 are deleted. Further, the noun dictionary 3 is searched by using the particle and role acquired by the particle and noun role acquisition means 8 as a key, and the noun complement that can be combined with the particle is acquired for each particle. Only. Further, the processing in the candidate sentence creation means 10 is defined with candidate particles including all of the noun candidates acquired for each particle by the noun candidate acquisition means 9 together with the corresponding particles acquired by the noun and noun role acquisition means 8. The candidate verbs are created by arranging them in order and arranging the inputted verbs last, and presenting them to the user via the display unit 2.

  In the flowchart of FIG. 5, steps s8 and s12 are deleted, only noun candidates are acquired from the noun dictionary in step s6, and the candidate sentences created in step s7 are displayed as they are in step s9.

  Thereby, based on the type of the input verb, a candidate sentence created from a character string extracted as a noun candidate and a noun that can be combined with the candidate sentence is selected from the candidate sentences for each noun candidate for each candidate string. Support the creation of a Japanese sentence exactly as in the case of the above-described embodiment, except that the display cannot be presented in a form corresponding to the field of the noun candidate and the field information stored in the field information storage unit. Can do.

  The present invention installs a program for realizing the functions shown in the block diagram of FIG. 1 or a program having the procedure shown in the flowchart of FIG. 5 via a medium or a communication line in a known computer. Is also feasible.

The block diagram which shows an example of embodiment of the Japanese sentence creation assistance apparatus of this invention Explanatory diagram showing an example of a noun dictionary Explanatory diagram showing an example of verb type list Explanatory diagram showing an example of a verb dictionary The flowchart which shows the flow of the process in the Japanese sentence creation assistance apparatus of this invention Explanatory drawing which shows an example of the display screen of a candidate sentence

Explanation of symbols

  1: input unit, 2: display unit, 3: noun dictionary, 4: verb type list, 5: verb dictionary, 6: field information storage unit, 7: verb type acquisition means, 8: particle and noun role acquisition means, 9 : Noun candidate acquisition means, 10: candidate sentence creation means, 11: Japanese sentence output means, 12: field information update means.

Claims (8)

Along with the input part and display part
For each noun used in a Japanese sentence, a noun dictionary formed by registering at least one particle that can be combined with the noun and at least one role of the noun,
Verbs used in Japanese sentences are classified into the number of nouns required in the Japanese sentence that contains the verbs, the particles combined with each noun, and the role of each noun. For each type, a verb type list formed by registering a particle and a role of each noun to be combined with each noun in a prescribed order;
For each verb used in a Japanese sentence, using a computer having a verb dictionary formed by registering at least one of the above verb types,
The computer
Search the verb dictionary using the verb input from the user via the input unit as a key, and obtain the verb type corresponding to the input verb;
Acquire the particle and the role of each noun that binds to each noun corresponding to the acquired verb type from the verb type list,
Search the noun dictionary using the acquired particle and role as a key, and obtain a noun complement that can be combined with the particle for each particle.
Candidate strings including all of the noun candidates acquired for each particle are arranged in the prescribed order together with the corresponding particles, the input verbs are arranged last, a candidate sentence is created, and a user is displayed via the display unit. To present
Of the noun candidates in the presented candidate strings for each particle, only the noun candidates selected from the user via the input unit are combined with the corresponding particles, respectively, and the input verb is combined finally. Japanese sentence creation support method characterized by creating and outputting Japanese sentences. Along with the input part and display part
For each noun used in a Japanese sentence, a noun dictionary formed by registering at least one particle that can be combined with the noun, at least one role of the noun, and at least one field in which the noun is used When,
Verbs used in Japanese sentences are classified into the number of nouns required in the Japanese sentence that contains the verbs, the particles combined with each noun, and the role of each noun. For each type, a verb type list formed by registering a particle and a role of each noun to be combined with each noun in a prescribed order;
A verb dictionary in which at least one verb type is registered for each verb used in a Japanese sentence;
Using a computer with a field information storage unit that stores field information indicating the frequency of use of each noun field in Japanese sentences,
The computer
Search the verb dictionary using the verb input from the user via the input unit as a key, and obtain the verb type corresponding to the input verb;
Acquire the particle and the role of each noun that binds to each noun corresponding to the acquired verb type from the verb type list,
Search the noun dictionary using the acquired particle and role as a key, and obtain a noun complement that can be combined with the particle together with its field for each particle.
Candidate strings including all of the noun candidates acquired for each particle are arranged in the prescribed order together with the corresponding particles, and the input verb is arranged last to create a candidate sentence, and among the candidate sentences, The display of each candidate noun for each candidate column is presented to the user via the display unit in a form corresponding to the field of the noun candidate and the field information stored in the field information storage unit,
Among the noun candidates in the presented candidate strings for each particle, only the noun candidates respectively selected from the user via the input unit are combined with the corresponding particles, and the input verb is combined finally. Create and output Japanese sentences,
Updating the field information stored in the field information storage unit based on the field of the selected noun, a Japanese sentence creation support method. In the Japanese sentence creation support method according to claim 1 or 2,
When multiple verb types are acquired from the verb dictionary for a verb input from the user via the input unit, the roles of particles and nouns are acquired for each verb type, and noun candidates are acquired. Each candidate sentence is created and presented to the user via the display unit, and the noun candidates in the candidate strings for each particle of the presented candidate sentences are selected from the user via the input unit, respectively. A Japanese sentence creation support method characterized in that only noun candidates in a single candidate sentence are combined with corresponding particles, and the input verb is finally combined to create and output a Japanese sentence. The Japanese sentence creation support method according to any one of claims 1 to 3,
When two or more noun candidates are selected from the noun candidates corresponding to one particle, the noun candidates are connected with each other using the character string “to” and combined with the particle in the Japanese sentence. Japanese sentence creation support method characterized by creating and outputting. An input section;
A display unit;
For each noun used in a Japanese sentence, a noun dictionary formed by registering at least one particle that can be combined with the noun and at least one role of the noun,
Verbs used in Japanese sentences are classified into the number of nouns required in the Japanese sentence that contains the verbs, the particles combined with each noun, and the role of each noun. For each type, a verb type list formed by registering a particle and a role of each noun to be combined with each noun in a prescribed order;
A verb dictionary in which at least one verb type is registered for each verb used in a Japanese sentence;
A verb type acquisition means for searching a verb dictionary using a verb input from a user via an input unit as a key, and acquiring a verb type corresponding to the input verb;
Particle and noun role acquisition means for acquiring a particle and a role of each noun from a list of verb types combined with each noun corresponding to the acquired verb type;
Noun candidate acquisition means for searching a noun dictionary using the acquired particle and role as a key, and acquiring a noun complement that can be combined with the particle for each particle;
Candidate strings including all of the noun candidates acquired for each particle are arranged in the prescribed order together with the corresponding particles, the input verbs are arranged last, a candidate sentence is created, and a user is displayed via the display unit. Candidate sentence creation means presented in
Of the noun candidates in the presented candidate strings for each particle, only the noun candidates selected from the user via the input unit are combined with the corresponding particles, respectively, and the input verb is combined finally. Japanese sentence creation support device characterized by comprising a Japanese sentence output means for creating and outputting Japanese sentences. An input section;
A display unit;
For each noun used in a Japanese sentence, a noun dictionary formed by registering at least one particle that can be combined with the noun, at least one role of the noun, and at least one field in which the noun is used When,
Verbs used in Japanese sentences are classified into the number of nouns required in the Japanese sentence that contains the verbs, the particles combined with each noun, and the role of each noun. For each type, a verb type list formed by registering a particle and a role of each noun to be combined with each noun in a prescribed order;
A verb dictionary in which at least one verb type is registered for each verb used in a Japanese sentence;
A field information storage unit that stores field information indicating the frequency of use of each noun field in the Japanese sentence;
A verb type acquisition means for searching a verb dictionary using a verb input from a user via an input unit as a key, and acquiring a verb type corresponding to the input verb;
Particle and noun role acquisition means for acquiring a particle and a role of each noun from a list of verb types combined with each noun corresponding to the acquired verb type;
Noun candidate search means for searching a noun dictionary using the acquired particle and role as a key, and acquiring a complement of a noun that can be combined with the particle together with its field for each particle;
Candidate strings including all of the noun candidates acquired for each particle are arranged in the prescribed order together with the corresponding particles, and the input verb is arranged last to create a candidate sentence, and among the candidate sentences, Candidate sentence creating means for presenting each candidate noun for each candidate string in a form corresponding to the field of the noun candidate and the field information stored in the field information storage unit and presenting it to the user via the display unit When,
Among the noun candidates in the presented candidate strings for each particle, only the noun candidates respectively selected from the user via the input unit are combined with the corresponding particles, and the input verb is combined finally. Japanese sentence output means for creating and outputting Japanese sentences,
Field information communication means for updating the field information stored in the field information storage unit based on the field of the selected noun is provided. In the Japanese sentence creation support device according to claim 5 or 6,
When multiple verb types are acquired from the verb dictionary for a verb input from the user via the input unit, the roles of particles and nouns are acquired for each verb type, and noun candidates are acquired. And a noun role acquisition means, a noun candidate acquisition means, and a candidate sentence creation means for creating candidate sentences and presenting them to the user via the display unit, and candidates for each of the presented candidate sentences Of noun candidates in the column, only the noun candidates in one candidate sentence each selected by the user via the input unit are combined with the corresponding particles, respectively, and the input verb is combined finally to the Japanese sentence. Japanese sentence creation support device, characterized by comprising a Japanese sentence output means for creating and outputting a sentence. The Japanese sentence creation support device according to any one of claims 5 to 7,
When two or more noun candidates are selected from the noun candidates corresponding to one particle, the noun candidates are connected with each other using the character string “to” and combined with the particle in the Japanese sentence. Japanese sentence creation support device, characterized by comprising a Japanese sentence output means for creating and outputting a sentence.

Images