JPH0563831B2 - - Google Patents

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention converts a long sequence of continuous kana characters, such as those input as a sentence unit, into phrases while converting them into kana-kanji characters. This invention relates to a kana-kanji conversion device that can effectively create Japanese sentences.

[Technical background of the invention and its problems]

In conventional Japanese word processors, etc., the unit of Kana-Kanji input is generally limited to a single phrase, and in most cases, the input of compound nouns is also limited to several units at most. be. Therefore, when inputting and creating Japanese sentences using this type of device, the operator must always be aware of the units of words or phrases, which places a heavy burden on the operator. Therefore, recently, various studies have been attempted to input a sequence of pronunciations of kana in sentence units, so-called solid sentences, without restricting the units of text input, and to perform kana-kanji conversion processing on the solid sentences. Specifically, they achieve their purpose by, for example, recursively performing clause analysis processing. However, this requires a considerably long processing time and requires a large amount of buffer memory, which is a problem. Furthermore, although it has been considered to simplify the algorithm for phrase analysis processing by limiting the processing time and memory amount, it cannot be denied that the accuracy of the conversion processing deteriorates. In addition, a major issue has been how to display and output the conversion results obtained in this manner so that instructions for selecting homophones can be easily handled.

For example, if you enter the character string ``Zandaka wo motoru'' in kana, it is expected that it will be able to be mechanically divided into clauses as ``Zandaka wo / motoru''.
Alternatively, the division form ``Zandaka wo mo/stop'' is also grammatically possible. In this case, it is generally considered empirically that the so-called longest match is most likely, but based only on such empirical rules, clause analysis is always performed starting from the beginning of the input character string, for example, If only "stop" is extracted, the conversion accuracy will be significantly degraded. Therefore, it becomes necessary to extract a plurality of conversion candidates such as "also/stop balance" and "to/calculate balance" and leave the selection to the operator's discretion.

Also, for the input "Keisanshiki",
If the word "calculation formula" is not registered in the dictionary within the device, a plurality of conversion results will similarly occur, such as "calculation/formula" and "calculation/tree." Furthermore, conversion results such as "hair/heritage/formula" are also generated. Furthermore, if the word ``rukei'' is registered in the dictionary,
The following candidates also appear: ``To find the balance/exile/formula.''

By the way, as an evaluation process for outputting the most probable one in the first order among various conversion results for such an input, for example, priority is determined in order of the number of clauses or words that make up the whole. I have thought of a way to do that. Specifically, for example, in response to the input of "Kogakushi Yotoku", "high amount/income" is determined to be more likely than "term ga/phrase/income" or "term ga/kukoto/phrase". It is.
In this case, it is natural and preferable to output homophones with priority given to words that are used more frequently. However, even if the number of phrases that make up the whole phrase is the same, the way they are separated may be different, and when outputting analysis results that include a large number of phrases to reduce conversion omissions, the data to be handled is The structure becomes complicated. Moreover, the problem is how to display and output the plurality of conversion results in order to make it easier for the operator to perform selection operations. To solve this problem, for example, a dictionary search unit with multiple words registered in the dictionary can be used to extract a series of clause units from a series of input character examples, and each of these clauses can be associated with the pronunciation of that clause. When converting to kana-kanji mixed notation and outputting, the forwardmost bunsetsu start point among the mutually different break positions of the plural bunsetsu series for the input character string is used as a reference point, and the next clause that exists after this reference point is A method has been considered in which kana-kanji conversion is performed using the independent word part of the clause and the dependent word part of the previous clause separated by the reference point as units of conversion candidates, and the conversion candidates are displayed and output in order.

For example, in the above example, for the input ``Keisanshiki to find Zandaka'', create phrase series candidates such as ``Also/stop the balance'' or ``Determine the balance'', and convert the conversion candidate result. It asks for things like "[stop] [balance]", "[requires] [balance]", etc., and displays and outputs these in order, which greatly simplifies the process of selecting the next candidate with the same sound. be able to.
That is, for example, regarding the above part,
Although the word ``Modome'' does not actually exist, there are other homophone word information and other written words in the pronunciation of ``Moto''. In other words, it is shown that there are other candidates corresponding to the same pronunciation, and similarly to the selection of homophones within a bunsetsu, the selection of a differently structured bunsetsu string becomes possible by instructing the switching of the next homophone candidate.

However, if the different structure clause string consists of one independent word, the selection operation is easy, but if it consists of a combination of multiple independent words, many homophones will be output. For example, in a solid sentence, you could write, ``When considering the state of research and development activities from a modern perspective, the authors' remorseful opinion is that there was a lack of effort in recording and publishing research results.'' When input, {research and/development, research/evening party/departure} {gathering/state, activity-like/nest, activity-like/vinegar, activity/state} of {today/target, today/enemy, today/
Drops, dark blue/intellectual {branch, point of view, starting point, fulcrum}
From {thought, intuition, view, can}, {research / production, research / system, research / voice, research / life, research / world, research / sex, research / star , research/results, research/midsummer, research/net price, research/fresh flowers, research/confectionery, research/fruit and vegetables, matter/acute, cum/
Acute, ticket/acute, sword/acute, prefecture/acute,
Case / sudden death, cum / sudden death, ticket / sudden death, sword / sudden death, prefecture / sudden death, case / old system, cum / old system, ticket /
Emergency system, sword/old system, prefecture/old system, case/maiden name,
cum/maiden name, ticket/maiden name, sword/maiden name, prefecture/maiden name, research/meaning, research/below, research/medical, research/stomach, research/made/day, research/ System/
day, research/voice/day, research/life/day, research/world/
day, research/star/day, research/production/section, research/system/
Division, Research/Voice/Division, Research/Student/Division, Research/World/
Division, Research/Star/Division, Research/Production/Cation, Research/System/
transformation, research / voice / transformation, research / life / transformation, research / world /
, research / star / , research / production / department, research / system /
department, research/voice/department, research/student/department, research/world/
department, research / star / department, research / production / home, research / system /
Home, Research/Voice/Home, Research/Life/Home, Research/World/
Home, Research/Star/Home, Research/Made/Bottom, Research/System/
lower, research/voice/lower, research/student/lower, research/world/
{Seki, Crown} for the record of {Research/Star/Bottom} and {favorable reception, publication, comments}. Writing with insufficient effort {to print, tosu, other to, rice} {car, person,
The opinion of the company and the building is that it is unconstitutional. Yes, the ticket is
Noiken, Noi Prefecture, opinion group, unconstitutional faction, opinion wave, unconstitutional wave, Noiken faction, Noiken faction, Noiken faction, Noiken faction, Noikaneha , Noi Kenha, Noi Kenha,
Conversion candidates for ``Noi Prefectural Wave'' are obtained. Here, more than 60 homophones were generated for the input sentence ``Kenki Yuuseikano'', and selection took a lot of time and caused many problems.

[Purpose of the invention]

The present invention has been made in consideration of these circumstances, and its purpose is to provide a kana-kanji conversion device that can efficiently select a large number of homophones and improve the efficiency of document creation. be.

[Summary of the invention]

When selecting homophones, the present invention not only stores confirmed confirmation information for each phrase, but also stores combination information with adjacent phrases, and a conversion candidate that includes this stored confirmation information appears. When this happens, priority is given to output.

〔Effect of the invention〕

Thus, according to the present invention, if "research/results" is selected in response to the input of "kenkiyuuseika" in the above example, the next time the same input is received,
"Research/Results" is output with priority over "Research/Production" and only needs to be confirmed. That is, for example, by selecting the above-mentioned "research/results", "research" and "results" are stored as confirmation information for each phrase, and "research-results" is stored as adjacent combination information. and is subject to priority output. When all of the above example sentences are selected, the following combination information is stored.

Research-development t0 Research-results t0 Development-activities t1 Results-records t3 Activities-status t0 Records-publication t2 Status-today t1 Announcement-related t1 Today-target t0 Related-effort t3 Target-perspective t1 Effort-deficiency × Perspective- Thoughts t1 Author t2 Author t1 Reflections t2 Reflections t3 Opinions t0 written after the combination information in the above example
~t3 indicates the storage timing. The timing is different depending on whether there is a homophone opposition word in adjacent clauses or when there is no homophone opposition word before or after it.

t0...When the homophone is selected, it is a compound word.

t1… 〃 The previous clause is a homophone antonym.

t2… 〃 The previous clause is not a homophone antonym.

t3... 〃 The following clause is not a homophone antonym.

x... Adjacent clauses are not homonyms.

Here, × does not need to store combination information. Also, ignoring particles and auxiliary verbs between clauses,
It is shown in the example. Depending on the timing of storing the confirmation information, it may not be necessary to store it as combination information.

For example, storing only t0 also increases selection efficiency. Therefore, there is no need to provide a special instruction key to select another candidate in a string of differently structured phrases.
Great effects such as the ability to efficiently perform selection processing using conventional homophone next candidate keys can be achieved.

[Embodiments of the invention]

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a schematic configuration diagram of an embodiment device. The input device 1 includes a keyboard device, a voice recognition device, a kana character reading device, and the like. The character string (first character string) representing the reading input via this input device 1 is:
For example, after being converted into a kana character code, it is given to the kana-kanji conversion unit 2. Note that the first character string representing the above-mentioned pronunciation is indicated as, for example, hiragana, katakana, romaji, or the like.

The kana-kanji conversion unit 2 is, for example, a block extraction unit 2.
a, a total phrase series extraction section 2b, a phrase extraction section 2c, and a conversion dictionary 3. This kana-kanji conversion unit 2 converts the first
For the character string , a second character string consisting of a corresponding display character string mixed with kanji is obtained, and this is provided to the output control section 4. Block extractor 2a
is provided to efficiently analyze particularly long input character strings for the first character string, and is set to a preset value N, for example, N.
= 4, the phrase analysis result to be associated as a phrase series within N phrases is obtained, and when there is no corresponding phrase analysis result, the first input character string is divided into several block sections. Then, the pronunciation kana strings of each block divided in this way are sent to the total bunsetsu series extraction section 2b, and the kana-kanji conversion results for each block obtained by the total bunsetsu series extraction section 2b, that is, the second character. The columns are sequentially sent to the output control section 4.

The above-mentioned total phrase series extraction section 2b includes a phrase extraction section 2c.
is used to divide the first character string into divisible phrase series, and output the conversion results of kana-kanji mixed notation obtained for each of these phrase series to the block extraction section 2a. The phrase series is obtained for all combinations of the input character string that can be divided into phrases, and after ranking the most likely candidates by, for example, priority evaluation, the phrase series is output in order starting from the first ranked candidate. Ru.
In addition, when evaluating the priority, in general, there is a higher tendency for phrases with a small number of phrases to correspond to sentences containing kanji that are the input target, so for example, phrase sequences with a small number of phrases are given priority for output. will be carried out.

The phrase extraction unit 2c performs a collation search between the input character code string and character strings (words) registered in advance in the conversion dictionary 3, and extracts written characters containing kanji that correspond to the first character string. We are looking for a second string of characters. The conversion dictionary 3 has an input heading table area 3, as shown in FIG.
a, an output heading table area 3b and a part of speech area 3c, a first character string representing a reading is stored in the input heading table area 3a, and a first character string representing a reading is stored in the input heading table area 3a; The second character string is stored in the output heading table area 3b. The part of speech area 3c stores part of speech information for the first and second character strings.

Therefore, the clause extracting unit 2c searches for a character string (word) registered in advance in the input heading table area 3a of the conversion dictionary 3 using, for example, a known longest forward match method for the given input character string, By analyzing the conjugated endings and adjuncts, the part of the input character string that matches the initial character of the input character string for the longest length is obtained as the branch extraction result. At this time, the above-mentioned analysis of the conjugated ending is performed based on the part of speech item stored in the part of speech area 3c. Then, a second character string consisting of written characters mixed with kanji corresponding to the phrase extraction result found through this analysis is read out from the output heading table area 3b and output. Furthermore, at this time, the total phrase series extraction unit 2b inputs the input unit (phrase extraction result) to the phrase extraction unit 2c into various combinations that can be divided into phrases in the block section determined for the input character string. We are trying to find the most probable phrase sequence among them.

The kana-kanji converter 2 converts the second kana-kanji mixture of kana-kanji notation obtained for such a phrase series.
The character string is transferred to the output data memory 5 of the output control section 4. The output control unit 4 converts these data into a predetermined display output format and stores it in the document display memory 6.

These data stored in the document display memory 6 are sent to the confirmation information priority section 8, and by referring to the confirmation information storage section 9, if a conversion candidate that has already been confirmed is found, that conversion is performed. The order is changed so that candidates are prioritized and output.

In this embodiment, the confirmation information storage unit 9
As shown in Figure d, it consists of information for each phrase and information on combinations of adjacent phrases. Information for each clause is
Words in ROM and words in FDD/RAM are memorized separately. The words in the ROM memorize standard words commonly used in kana-kanji conversion. The words stored in the FDD/RAM include words registered by users and words by field.
With this configuration, confirmation information can be easily replaced when the field is changed.

The combination information of adjacent phrases stores the relative position of the information for each phrase, and whether the word is a standard word, a word registered by the user, or a field-specific word.

FIG. 3e shows an example in which data is stored at timing t0 in the above example.

Here, the output data memory 5 includes a combination table 5a, a mapping table 5b, and a headword table 5c as shown in FIG. It is stored together with the structure. This example represents the data storage structure of the Japanese sentence example shown in FIG. ing. This table 5
The rows of a indicate the interpretational types of bunsetsu structures, and the columns describe the connections of clauses in each bunsetsu structure in order. That is, it is shown that there is one type of sequence candidate in the first block, two types in the second block, and three types in the third block. Further, each of these numerical values indicates an element number of the mapping table 5b. In addition, headwords of homophone dissonance words that exist for each phrase unit are grouped and stored in a table 5c, and each group in the headword table 5c is described using each element of the mapping table 5b as a pointer. In this way, the correspondence with the input reading kana position is also described at the same time.

FIG. 3b shows a table structure of the document display memory 6 that stores character display information for outputting the contents of the output data memory 6 to the display device 7.
Based on the contents of the output data memory 5, this memory 6 compares and collates homophone meaning words and different structure clause strings located at the same pronunciation kana position. The input character string is divided into parts, and the common adjunct character part and the uniquely obtained conversion result part are each separated from other parts and described. In addition, for conversion candidate results that have the same reading kana in the adjunct part for the headword character at the beginning of a certain bunsetsu, it is separated according to the beginning of the previous bunsetsu, and the adjunct part after the separation is The string is written by joining it to the next clause. In other words, even if it is a letter of an adjunct word in a clause of a certain bunsetsu series, if the corresponding part with the same reading is part of the beginning of an independent word in another clause series, then the letter of that adjunct word is written in conjunction with the conversion candidate for the next clause.

FIG. 3c is an example in which the contents of FIG. 3b are changed in order by the confirmation information priority unit 8. In this example, if "resource development" is stored in the confirmation information storage unit 9 as combination information, "resource development"
is changed to the first place in the display ranking.

As will be described later, these homophones are displayed and output with their display attributes different from normal display attributes. FIGS. 5 a to 5 e are display examples, and the shaded areas indicate different display attributes. This change in display attributes is performed, for example, by inverting, blinking, changing brightness, underlining, etc. the displayed characters.

Next, the kana-kanji conversion process in the kana-kanji conversion section 2 will be explained using the example of kana character input shown in FIG.

The block extraction unit 2a obtains as many as possible a sequence of N phrases from the beginning of the input character sequence.
Assuming that N is 4, for example, in the example shown in FIG. 4a, the entire input sequence is first input to the clause extractor 2c, and "and" is obtained as the first clause result by the longest match method. Next, the longest matching result is obtained in the same manner, using the point after this bunsetsu break as the starting point (the position of the next bunsetsu start character), to obtain the bunsetsu ``Kongonoto''. Such processing is repeated in order to obtain the first phrase series candidate ``And/Kongonoto/Ushiha/Kaiteishi'' as shown in item ``A'' in FIG. 4B. Next, in order to obtain a phrase series different from the phrase series obtained for this item "A", we will use the last character of the third phrase result "Ushiha", that is, "' is deleted and this is added to the phrase extraction section 2c.
Similarly, the longest-matching phrase ``Ushi'' is obtained, and ``HA'' is found as the following phrase starting with ``HA''. Thereafter, each time a short clause is obtained from the third, second, and first clauses, another series of clauses following these clauses is sequentially obtained. In this way, all four-clause sequences that can be divided into phrases for the input character string are obtained as shown in FIG. 4b. At this time, a corresponding headword candidate (second character string) containing kanji is also obtained.

Next, among these sequences, attention is focused only on the candidate whose total length is the longest (the bunsetsu sequence). This is based on the fact that, as shown above, the sequence with the minimum number of input clauses has a high tendency to match the conversion result intended for input. This means that the number of clauses in one block should be the minimum, and if the number of clauses in one block is the same,
This means that the block has a longer length.

However, among the results shown in FIG. 4b,
The longest bunsetsu series are those shown in item "A" and item "U". Next, we find locations where these phrase series have phrase breaks in common. In this example, "and/~" and "~
は／〜”し is required as a common clause break point. The block extracting unit 2a determines these two positions as delimiters of blocks, and defines the first block section as "and" and the second block section as "congo no toshiwa." Then, the character strings in these sections are sequentially analyzed by the total clause string analysis section 2b, and the conversion results are sent to the output control section 4.

As a result, the first block section above is "and"
This information is first sent to the output control unit 4.
sent to. In this case, since there are no other homophones, the text is displayed as character data (conversion result) in the document in its normal form. Thereafter, the second block section is analyzed.

Here, the total phrase sequence analysis unit 2b calculates character sequences corresponding to the pronunciations of a given block section by brute force, but in reality, as shown in FIG. 4b, the phrase sequence has already been determined. Therefore, it is sufficient to select only those that correspond to the range of the specified interval. As a priority evaluation, for example, if only the candidates with the minimum number of clauses are selected, the analysis result will be as in the item "" in FIG. 4d. Of course, other clause candidate sequences may also be stored in the output data memory 5. For example, “Future / ask /
It is also possible to add "magazine" to the output results.

Next, the block extracting unit 2a extracts the unit of block in the same way for the remaining portion of the input character sequence for which no block has been determined yet, that is, the character string "Kaitishige...". Then, the conversion result as shown in the item "" in FIG. 4d is obtained. Subsequently, the conversion result ``limited'' is obtained as shown in the item ``'' in FIG. 3d, and the conversion process for the entire input series is completed.

The above conversion results are output to the output controller 4 for each block.
sent to. The output control section 4 has an output data memory 5
The data for each block stored in the data block is converted, sequentially stacked in the document display memory 6, the output priority is changed, and the data is output to the display device 7.
That is, the output control section 4 creates document display data as shown in FIG. 3b, and first outputs this to the display device 7 as shown in FIG. 4a. In the above example, for the phrase candidate "Koku no to", the independent word part is separated as "Kitai" and the attached word part is separated as "noto". Similarly, the bunsetsu candidate "Ushiwa" is separated into "Ushi" and "Ha". Furthermore, in the second clause series candidate, the independent word part is separated into "Shikaku" and the attached word part is separated into "no", and "Investment wa" is separated into "Investment" and "Ha". Therefore, in this case, both "no" and "wa" are common adjunct characters, and since "future" has no other homophones, these character parts are displayed in the normal display form. . Furthermore, since there are a plurality of conversion candidates (homophones and allographs) for other characters, they are displayed with high brightness, for example, to emphasize these candidates. In other words, although "to" in the bunsetsu "Koku no to" is an adjunct, it is treated together with the next bunsetsu word "ushi".

Therefore, the display device 7 will display, ``And from now on, [Togyo] will be limited to [revised and limited].'' Note that characters in brackets [ ] indicate characters displayed with high brightness.

In order to select the desired headword here, the input device 1 is provided with a next candidate key, as shown, for example, at 1a in FIG. This next candidate key 1a
For example, as shown in FIG. 5, when the cursor 1b is placed on the [and cow] part and the next candidate key 1a is operated in this state, the function is to change the display to the next candidate. be. Therefore, in this case, instead of the structure in the first row of table 5a shown in FIG. 4a, the phrase string structure shown in the second row,
That is, "Future [Investment]" is output and displayed as shown in FIG. 5b. Further, the next candidate key 1
As you manipulate a, [Investment] becomes [Fighting Spirit] and [Clairvoyance].
The word will be changed to other homophones one after another, and then it will return to its original form. Note that these operations are performed using the document display memory 6 shown in FIG.
This is done based on data stored in .

By the way, the document correspondence table 6a shown in FIG.
1 to i7 and output data (conversion results) are shown in correspondence. The display word table 6b shows the contents of the output data. In displaying the conversion candidates, first, "toushi" in the display word table 6b is designated by data i3 in the document correspondence table 6a, and this is displayed. Thereafter, when the next candidate key 1a is operated, the pointer is advanced to the next candidate "Investment" in that block based on the data i3, and the display is switched. In this way, the pointer is advanced and the display is switched to ``frozen to death,'' and then it is controlled to return to ``cattle.'' As shown in the displayed word table 6b, the multiple conversion candidate results analyzed based on the bunsetsu unit as described above are organized according to homophone and allograph relationships, and the unit of separation is partially changed. ing.

When the next candidate key 1a is operated and the desired word is displayed, the selection key 1b is operated to confirm. By this confirmation operation, the output control section 6 instructs the confirmation information storage section 9 to store the confirmation information.
In the example, when it is specified that "investment" is the target word, for example, the dictionary number of "investment" is stored in the confirmation information storage section 9.

In this "investment", since there is not a plurality of clauses that have been confirmed, only information for each clause is registered.

Similarly, for the character string "Kaitishige...", the independent word part is treated as "revised", the adjunct word part is treated as "shi", and the "shi" part is treated as "resource".
Since there is a phrase word ``shi''
is combined with the next clause ``Limit'' and output. Since there is no break between the words that all candidates have in common until the following "Hatsu ni", these words are output together. Note that in machine processing, each word of an independent word is treated as a clause.

Since the character string "Kaiteishige..." has a homophone structure made up of multiple clauses, in addition to the information for each clause, information on combinations with adjacent clauses is also stored in the confirmation information storage unit 9. Remember. In the embodiment, "undersea - resources" and "resources - development" are stored. Here, the priority relationship between confirmation information and combination information for each phrase will be explained. For example, if "revision" is selected alone after "seafloor-resources" is selected, the priority relationship in phrase units is revision>seafloor. Therefore, when "Kaitei" is input, the output order is revised → seabed → revised → expanded. However, when "Kaiteishigen" is input, the combination information is prioritized, and the output order is: seabed → revision → revision → expansion.

6 to 11 show the flow of the above-described processing of this apparatus. In the control flow shown in FIG. 6, input key codes obtained from the input device 1 are constantly checked, and if the input code is a kana character code that corresponds to the reading of a Japanese sentence, it is stored in the stack in sequence. . If the input code indicates a conversion request, the kana-kanji conversion process shown in FIG. 7 is performed. This conversion request is made when the input device has a conversion request key.
This is generated by the operator inputting a character string of an appropriate length and then pressing the conversion request key.
Further, regardless of whether the input device has a conversion request key or not, it is desirable to automatically generate the conversion request when it is detected that the input character is a code indicating a punctuation mark, for example.

If the input code corresponds to the next candidate key 1a, the homophone selection process shown in FIG. Editing processing is performed on the text. Further, FIG. 9 shows the editing and output processing of the conversion candidates in FIG. 7 according to the present invention.

Generally speaking, word processors are known that select homophones sequentially for each conversion result, or select them all at once after inputting, for example, one page of character strings. The device may be of either type.

As explained above, according to this device, when a relatively long input kana character string is analyzed and converted into a character string containing kana and kanji to create a Japanese sentence, a large number of phrases are generated as a result. The column candidates can be displayed simply and clearly, and the order in which they are displayed is such that confirmation information is prioritized for complex phrases, thereby increasing the efficiency of the operator's homophone selection operation. In other words, it is possible to select homophones not only for one phrase candidate, but also for candidates in a series of different phrases with different phrase breaks as described above. Therefore, it is possible to easily select the target kana-kanji conversion characters from the conversion candidate characters of homophone dissonance words and homophone dissonance phrases, and to create a Japanese sentence extremely efficiently. Furthermore, the burden on the operator can be significantly reduced, and its practical advantages are enormous.

Note that the present invention is not limited to the above embodiments. For example, even if the input is not a solid sentence, but is input in units of clauses, the output priority order can be changed by storing information that combines the units of clauses and adjacent clauses, and the selection can be encouraged. It's okay. Further, the types of adjunct words and the adjunct words themselves may be stored in combination. In short, the present invention can be implemented with various modifications without departing from the gist thereof.

[Brief explanation of the drawing]

FIG. 1 is a schematic configuration diagram of a device showing an embodiment of the present invention, FIG. 2 is a diagram showing the configuration of a memory of a conversion dictionary,
FIG. 3 is a diagram showing the configuration of the output data memory and document display memory and confirmation information, FIG. 4 is a diagram showing input character strings and their conversion processing, FIG. 5 is a diagram showing an example of displaying conversion candidates, 6 to 11 are diagrams showing an example of a control flow of conversion processing. 1... Input device, 2... Kana-kanji conversion section, 3... Conversion dictionary, 4... Output control section, 5... Output data memory,
6... Control table, 7... Display device, 1a... Next candidate key, 1b... Selection key, 2a... Block extractor,
2b... Total clause series extraction unit, 2c... Clause extraction unit, 3
a... Input heading table area, 3b... Output heading table area, 6a... Document correspondence table, 6b... Display word table, 8... Confirmation information priority section, 9... Confirmation information storage section.

Claims (1)

[Claims] 1. An input device for obtaining a series of input character strings;
a dictionary storage unit that stores a plurality of words; a dictionary search unit that searches this dictionary storage unit; a clause sequence extraction unit that uses the dictionary search unit to extract a series of clause units from the input character string; A result output unit that performs grammatical evaluation on each clause found by the sequence extraction unit and outputs a notation containing kana and kanji, and provides confirmation information for the conversion results output to this result output unit. means, a confirmation information storage section that stores the confirmed information, and a confirmation information priority section that changes the priority order when a conversion candidate having the confirmation information stored in the confirmation information storage section appears; The kana-kanji conversion device is characterized in that the confirmation information stores confirmation information for each phrase and information combining adjacent phrases. 2. The kana-kanji conversion device according to claim 1, wherein the confirmation information priority section outputs information combining adjacent phrases with priority over confirmation information for each phrase. 3. The information on the combination of adjacent phrases stored in the confirmation information storage section is stored in the result output section when the phrase is a compound phrase with no single segmentation. The kana-kanji conversion device described in item 1.