JPH0740262B2 - Kana-Kanji converter - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a kana-kanji conversion device for converting a Japanese sentence input in kana into a kana-kanji mixed sentence.

2. Description of the Related Art In recent years, a kana-kanji conversion device has been widely used in various Japanese processing devices such as a Japanese word processor as the most general means for inputting Japanese.

By the way, since there are many homophones in Japanese words, it may be difficult to uniquely convert a Japanese sentence input in kana to a kana-kanji mixed sentence using a kana-kanji conversion device. It is well known that there are many.

Therefore, in order to reduce the selection operation of homonyms by the user, by providing a means for storing the usage information of the homonyms selected by the user in the past, the homonyms are displayed in descending order of frequency of use. There have been devised methods for displaying and displaying homonyms in order from the most recently used one.

Problems to be Solved by the Invention However, since the above method is basically based on performing Kana-Kanji conversion on a word-by-word basis, the following problems occur when attempting Kana-Kanji conversion on a phrase-by-phrase basis. Had. That is, in a conventional kana-kanji conversion device that can convert a kana string into a kana-kanji mixed sentence in units of words or clauses, the kana-kanji mixed sentence obtained by performing kana-kanji conversion on the same input kana string is a grammar. There are cases in which there are two types of conversion results: one that consists of independent words and one that consists of independent words attached to adjuncts.

For example, two types of conversion results can be obtained for the input Kana sequence of "Kareha": "dead leaf" independent word (dead leaf) "he is" independent word (he) + adjunct word (ha).

At this time, according to the conventional method, when the user selects “dead leaf”, “dead leaf” is included in the homonym of “kaleha”.
The word "he" has a higher priority, and when "he is" is selected, the word "he" has a higher priority among the homonyms of the reading "kare". That is, in the conventional method, when the conversion results of the above two types are conceivable, it is possible to determine the priority order of the homonyms between the conversion results belonging to the same type. It is not possible to judge whether or not to display with priority. Therefore, even if the user always selects the conversion result of "Kareha" for the input Kanaha "Kareha", the next time he tries to convert the input Kanaha of "Kareha" again, "He is" There is a problem that the conversion result is not always displayed with priority.

The present invention has been made to solve the above problems.

Means for Solving Problems According to the present invention, “independent words” and “independent words +
When two types of conversion results with "adjunct" are obtained,
Not only the priority of homonyms is determined only between the conversion results that belong to the same type, but also the priority between independent words that belong to two types and have different reading lengths is considered. is there.

Operation With the above configuration, it is possible to display a plurality of obtained conversion results in descending order of priority in order that is most suitable for the user.

Embodiment An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a kana-kanji conversion device according to an embodiment of the present invention. In FIG. 1, reference numeral 1 denotes a means for inputting a Japanese kana string, and a desired output from a plurality of kana-kanji mixed output character strings obtained by performing kana-kanji conversion on the input kana string. It is an input device having means for selecting a character string. An input kana string storage unit 2 stores a kana string input from the input device 1. A kana-kanji conversion control unit 3 reads a kana string from the input kana string storage unit 2 and generates a character string containing kana-kanji characters. Reference numeral 4 denotes a word storage unit which stores words that can be independent words other than auxiliary particles and auxiliary verbs in association with their readings, Kanji notation, and part-of-speech information. Reference numeral 5 denotes a grammatical information storage unit that stores information about a set of adjunct words including particles and auxiliary verbs and connection information between words. Reference numeral 6 indicates the reading and kanji notation of a word used by the user among the words stored in the word storage unit, and auxiliary word addition information indicating whether or not an auxiliary word has been added to the word at the time of use. Is a word use information storage unit for storing The word information is stored in the word use information storage unit 6 by moving the already stored word information backward by one word and storing the word information used at the latest point in the resulting empty area. The more recently used words are stored closer to the beginning. 7 is a priority order of a pair of an input kana string generated by the kana-kanji conversion control unit 3 in which a delimiter is inserted at the boundary between an independent word and an adjunct word and an output character string mixed with kana-kanji obtained by kana-kanji conversion. It is an output character string storage unit for storing in descending order. The output character string storage unit has an area for storing priority information for each output character string. Reference numeral 8 denotes a display device for displaying the output character strings stored in the output character string storage unit in descending order of priority. 9
Stores the word information of the selected word in the word use information storage unit 6 by selecting one set of output character strings from the set of output character strings displayed on the display device 8 via the input device 1. This is a word usage information updating unit.

The operation of the kana-kanji conversion device configured as described above will be described below with reference to the flowchart shown in FIG.

(B) ... The kana string input by the user from the input device 1 is stored in the input kana string storage unit 2.

(B) ... The Kana-Kanji conversion control unit 3 executes the following operation. The Kana string is read from the input Kana string storage unit 2, the words are read from the word storage unit 4 so that the word reading matches a part or all of the input Kana string, and these words are stored in the grammar information storage unit 5. After determining whether or not a correct output character string can be constructed by using the information on the set of attached auxiliary words and the connection information between words, regarding the grammatically correct output character string, An input kana string in which a delimiter is inserted at the boundary and an output character string mixed with kana-kanji are paired and sequentially stored in the output character string storage unit 7.

(C) ... In the Kana-Kanji conversion control unit 3, the set of output character strings stored in the output character string storage unit 7 in (B) is used to determine the priority using the word information stored in the word use information storage unit 6. Sort from highest to lowest. The sorting procedure will be described in detail below. First, the reading and kanji notation of the independent word portion of the output character string are extracted for all the output character strings stored in the output character string storage unit 7. Here, the reading of the independent word part is obtained from the input kana string up to immediately before the delimiter, and the kanji notation of the independent word is taken from the output character string mixed with kana-kanji to just before the delimiter. Further, it is checked whether or not there is an adjunct word after the retrieved independent word, by checking whether or not kana remains after the delimiter symbol in the input kana string. Next, with respect to all the output character strings, the reading of the independent word extracted by the above operation, the word information consisting of three terms of the kanji notation and the additional information of the attached word, and the word information stored in the word use information storage unit 6 Perform a comparison to see if there is a match. At this time, the comparison of the word information is started from the beginning of the word use information storage unit 6, and if the word information does not match, the word information at the next position is sequentially compared. If the word information matches, the word information is compared. After storing the stored position in the word use information storage unit 6 in the priority information storage area of the corresponding output character string of the output character string storage unit 7, the word information of the next output character string is compared. To do so. Further, when it is found that all the word information has been compared and there is no matching word information, a specific numerical value indicating the lowest priority is set in the priority information storage area of the corresponding output character string in the output character string storage unit 7. After storing, proceed to compare word information for the next output string. In this way, when the match check of word information is completed for all output character strings, the output character string stored in the output character string storage unit at the end has the higher priority according to the priority information stored in the match check of word information. Sort in order. The rearrangement is performed such that the priority information values are arranged in ascending order.

(D) The display device 8 displays the output character strings containing the kana-kanji characters stored in the output character string storage unit 7 in order from the first stored character string without the delimiter.

(E) ... Position information in the output character string storage unit of a set of output character strings that the user has given a selection instruction using the input device 1 from the set of output character strings displayed on the display device 8. Is transmitted to the word usage information updating unit 9.

(F) ... The word usage information update unit 9 recognizes the output character string selected by the user based on the position information in the output character string storage unit 7 obtained in (E), and the output character string (C The word information including the reading of the independent word, the kanji notation, and the auxiliary word additional information similar to the one extracted in () is stored at the head of the word use information storage unit 6. When storing the word information, the previously stored word information is moved backward by one word. At this time, if word information is already stored in all areas of the word use information storage unit 6, the last word information is erased.

The word information used in the word usage information storage unit 6 is accumulated by repeating the execution of the above Kana-Kanji conversion from (A) to (F), and an output character string obtained in the subsequent Kana-Kanji conversion. It will be used effectively when determining the priority of.

The operation has been described above with reference to the flowchart of FIG. 2. However, in order to make the operation more concrete, the operation performed by the Kana-Kanji control unit 3 will be mainly described with reference to FIGS. 3 to 7. An example will be described.

FIG. 3 is a diagram showing a storage state of the word storage unit 4, which is “Say”.
For words having the pronunciations of "and", "Seika", the reading, the kanji notation, and the part of speech information are stored. FIG. 4 is a diagram showing a storage state of the word use information storage unit 6, which also shows a portion in which word information about reading independent words “say” and “seika” is stored. FIG. 5 to FIG. 7 are diagrams showing the storage state of the output character string storage unit 7, and the storage state of the output character string storage unit 7 when Kana-Kanji conversion conversion is performed on the input Kana string “SEIKA” Shows the change. FIG. 5 shows the stored contents of the output character string storage unit 7 immediately after the processing of the flowchart (b) is executed. Here, an output character string determined to be grammatically correct, which is obtained by referring to the word information shown in FIG. 3 and the grammatical information storage unit 5 from the word storage unit 4, is stored. FIG. 6 shows the output character string storage unit at the time when the matching check between the word information stored in the word use information storage unit 6 and the word information in the output character string is completed in the execution of the process of the flowchart (c). 7 is a diagram showing a storage state of No. 7, which shows that the priority order information obtained by the above-mentioned matching check is newly stored in the priority information storage area. FIG. 7 is a diagram showing the storage state of the output character string storage unit 7 immediately after the execution of the process of the flowchart (C) is finished, and shows that the output character strings are rearranged in the order of high priority.

Referring again to FIG. 5, as the output character string for the input kana string "SEIKA", there are two types of output characters, "independent word (SEI) + adjunct word (KA)" and "independent word (SEIKA)". You can see that the columns are obtained. On the other hand, from the stored contents of the word use information storage unit 6 in FIG. 4, "surname" and "correct" are obtained as the words of "sei" used in the type to which the attached word is added, and the attached word is added. It can be seen that the reading used in the type consisting of only independent words that are not given gives the word "birth house" as the word "seika". Based on these information, the output character strings having the contents shown in FIG. 5 are rearranged in descending order of priority, and the state shown in FIG. 7 is obtained. Comparing FIG. 5 and FIG. 7 here, in the state of FIG. 5, the output character strings of the types “independent word + adjunct word” and “independent word” are stored separately, but FIG. In the state of, the output character strings are stored in descending order of priority, regardless of their type.

As described above, according to this embodiment, the word use information storage unit 6
Kana-Kanji conversion is performed for the input Kana sequence by storing the words used in the past in the input Kana string by distinguishing whether they were used in the form of "independent word + adjunct" or in the form of "independent word". Even when two types of output character strings, "independent word + adjunct word" and "independent word" are obtained when performing, the output character string with the highest priority is output in order regardless of the difference between these types. It can be displayed on the display device 8.

In the present embodiment, the word use information storage unit 6 is configured such that recently used words are prioritized. However, the word use information storage unit 6 indicates that the word stored in the word storage unit 4 is “independent”. A configuration may be used in which a word having a high frequency of use is prioritized such that two types of a word + adjunct word and an independent word are distinguished and the frequency of use in each type is stored.

EFFECTS OF THE INVENTION As described above, according to the present invention, even when two types of “independent word + adjunct word” and “independent word” are obtained as an output character string for kana-kanji conversion for the same kana sequence, The set of output character strings can be displayed in an order that is most suitable for the user regardless of the difference in the type, and its practical effect is great.

[Brief description of drawings]

FIG. 1 is a block diagram of a kana-kanji conversion device in one embodiment of the present invention, FIG. 2 is a flowchart showing the operation of the kana-kanji conversion device in one embodiment of the present invention, and FIG. 3 is a storage state of a word storage unit. FIG. 4, FIG. 4 is a diagram showing stored information in a word use information storage unit, and FIGS. 5, 6, and 7 are
It is a figure which shows the change of the memory state of an output character string memory | storage part. 1 ... input device, 2 ... input kana string storage unit 3 ... kana-kanji conversion control unit 4 ... word storage unit 5 ... grammar information storage unit 6 ... word usage information storage unit 7 ... output character string Storage unit, 8 ... Display device

Claims (1)

[Claims] 1. An input device for inputting a Japanese kana string, a word storage unit for storing words that can be independent words other than particles and auxiliary verbs, associated kanji notations and their part-of-speech information, and particles and A grammatical information storage unit that stores pairing information of auxiliary words consisting of auxiliary verbs and connection information between words, and a kana string input from the input device to the information stored in the word storage unit and the grammatical information storage unit. A kana-kanji conversion control unit that performs kana-kanji conversion based on the kana-kanji character and generates a character string candidate of an independent word consisting only of kanji that does not include a particle and auxiliary verb and / or a character string candidate consisting of that kanji and particle or auxiliary verb; A display unit that displays a plurality of character string candidates converted by the Kanji conversion control unit, and selects one of the plurality of character string candidates displayed on the display unit by the input device. A kana-kanji conversion device configured to read the independent word included in the character string candidates selected by the input device, the kanji notation of the independent word, and an adjunct word indicating the presence or absence of an adjunct word at the time of selection. By selecting one of the word use information storage units that store additional information in pairs according to a predetermined priority and the character string candidates displayed on the display unit by the input device, the selected character strings The word use information updating unit that updates the information stored in the word use information storage unit regarding the, and the Kana-Kanji conversion control unit reads the independent word for each of all the generated character string candidates, The kanji notation and its presence / absence are compared with corresponding information of each pair stored in the word use information storage unit, and character string candidates in which the above information match each other are stored in the word use information storage. Kana-Kanji conversion apparatus characterized by according to the priority of the matched pair of inner to be displayed on the display unit.