JP2022175661A - CALIBRATION SUPPORT DEVICE, CALIBRATION SUPPORT METHOD, AND PROGRAM - Google Patents

Abstract

To provide a proofreading assisting device, a proofreading assisting method and a program capable of extracting a letter string that has a possibility of an orthographical variant without increasing a necessary time for proofreading even if a sentence is a long sentence.SOLUTION: A proofreading assisting device includes: an obtaining unit that obtains a target sentence as a target of proofreading; a dividing unit that generates divided sentences obtained by dividing the target sentence into at least three or more; a connecting unit that generates a connected sentence which is obtained by connecting different divided sentences from each other among the divided sentences and is shorter than the target sentence; and an evaluating unit that extracts a letter sting which is a candidate for an orthographical variant in the connected sentence.SELECTED DRAWING: Figure 2

Description

The present invention relates to a proofreading support device, a proofreading support method, and a program.

Sentences are corrected for spelling errors. For example, Patent Literature 1 discloses a technique for extracting character strings that are candidates for spelling variations from a sentence to be proofread, comparing the extracted candidates with the spellings of other candidates, and determining the presence or absence of spelling variations. It is

JP-A-3-184162

The method described in Patent Document 1 compares a character string extracted from a sentence with all other candidate notations. For example, if the number of character strings extracted from a sentence is K (K is any natural number), the number of comparisons is K×(K−1). Therefore, if the sentence to be proofread is a short sentence of about 100 words, the number of times of comparison is about 10,000. As a result, the number of times increases at an accelerating rate. For this reason, there is a problem that it may take several days to check for spelling variations in long sentences, which is not practical.

The present invention has been made in view of such circumstances, and is capable of extracting character strings that may have spelling variations without increasing the time required for proofreading even long sentences. An object is to provide a support device, a proofreading support method, and a program.

The proofreading support apparatus of the present invention includes an acquisition unit that acquires a target sentence to be proofread, a division unit that generates split sentences by dividing the target sentence into at least three or more, and the split sentences. Among them, a connection unit that generates a connected sentence that is a connected sentence formed by connecting different divided sentences and is shorter than the target sentence, and an evaluation unit that extracts a character string that is a candidate for spelling variation in the connected sentence. Prepare.

A proofreading support method of the present invention is a proofreading support method performed by a computer, wherein an obtaining unit obtains a target sentence to be proofread, and a dividing unit divides the target sentence into at least three or more. A connecting unit generates a connected sentence formed by connecting different divided sentences among the divided sentences and is shorter than the target sentence, and an evaluating unit generates the connected sentence Extract character strings that are candidates for spelling variations in .

A program according to the present invention is a program for causing a computer to operate as the proofreading support apparatus described above, and for causing the computer to function as each unit provided in the proofreading support apparatus.

According to the present invention, there is a low risk of overlooking spelling variations, and it is possible to prevent an increase in the time required for proofreading even for long sentences.

4A and 4B are diagrams for explaining processing performed by the proofreading support apparatus 10 according to the embodiment; FIG. 1 is a block diagram showing a configuration example of a proofreading support device 10 according to an embodiment; FIG. FIG. 12 illustrates an example of list information 120 according to an embodiment; FIG. 12 illustrates an example of list information 120 according to an embodiment; FIG. 12 illustrates an example of list information 120 according to an embodiment; FIG. 12 illustrates an example of list information 120 according to an embodiment; FIG. 12 illustrates an example of list information 120 according to an embodiment; FIG. 12 illustrates an example of list information 120 according to an embodiment; 4 is a flow chart showing the flow of processing performed by the proofreading support apparatus 10 according to the embodiment; FIG. 10 is a diagram illustrating processing for extracting spelling variation candidates; FIG. 10 is a diagram illustrating processing for extracting spelling variation candidates; FIG. 10 is a diagram illustrating processing for extracting spelling variation candidates; FIG. 10 is a diagram illustrating processing for extracting spelling variation candidates;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings.

First, a method of extracting candidates for spelling variations (candidates for spelling variations) from a sentence will be described. 10A to 10D are diagrams for explaining the process of extracting spelling variation candidates.

An example of a sentence is shown in FIG. 10A. As shown in the example of FIG. 10A, a case of extracting spelling variation candidates from a sentence stating "There is handling. There is handling. There is handling. There is no handling." .

FIG. 10B shows an example of sentences divided by words (morphemes). As shown in the example of FIG. 10B, if the sentence in FIG. /./handling/was/no/desu/ne/.". "/" here is a symbol indicating a delimiter.

Processing for extracting spelling variation candidates will be described with reference to FIGS. 10C and 10D. FIG. 10C shows processing for extracting spelling variation candidates. FIG. 10D shows an example of the list used for the evaluation of FIG. 10C.

The table in FIG. 10C shows items such as words, the number of registered words in the list, evaluation results, and presence/absence of registration in the list. As for the words, the words included in the sentence are shown in order. The number of words registered in the list indicates the number of words registered in the list. The list is a list used when evaluating words contained in sentences, as shown in the example of FIG. 10D. Evaluation here means determining whether or not the words included in the sentence match or are similar to the words described in the list.

The evaluation result indicates the result of evaluating the words contained in the sentence. The evaluation result here indicates the result of determining whether or not the words included in the sentence match (or are similar to) the words registered in the list.

Whether or not the word is registered in the list indicates whether or not the word included in the sentence is registered in the list. Here, when it is determined that the word included in the sentence and the word registered in the list do not match (or are similar) based on the evaluation result, the word included in the sentence is added to the list. be done.

The table in FIG. 10D shows items such as list number, word, element 1, element 2, and so on. The list No. is identification information such as a number that uniquely identifies the words registered in the list. A word indicates a word specified by the list No. The element indicates words determined to be similar to the word specified in the list No.

First, evaluation is started with no words registered in the list. That is, it is determined whether or not the word "handling" shown at the top of the table in FIG. 10C matches or is similar to the words registered in the list. No words are registered in the list at this point. Therefore, the word "handling" is evaluated as "no match/similarity". Words for which an evaluation result of "no match/similarity" is obtained are registered in the list. In the example of FIG. 10C, list No. 1 indicates that the word "handling" has been registered. Then, No. in the list of FIG. 10D. 1 shows an example in which the word "handling" is registered.

Next, it is determined whether the word "ga" matches or is similar to the words registered in the list. Since only the word "handling" is registered in the list, the word "ga" is evaluated as "no match/similarity". Words for which an evaluation result of "no match/similarity" is obtained are registered in the list. In the example of FIG. 10C, list No. 2 shows that the word "ga" has been registered. Then, No. in the list of FIG. 10D. 2 shows an example in which the word "ga" is registered.

Next, it is determined whether the word "with" matches or is similar to the words registered in the list. Only the words "handle" and "ga" are registered in the list. For this reason, the word "yes" is evaluated as "no match/similarity". Words for which an evaluation result of "no match/similarity" is obtained are registered in the list. In the example of FIG. 10C, list No. 3 shows that the word "yes" has been registered. Then, No. in the list of FIG. 10D. 3 shows an example in which the word "with" is registered.

Next, it is determined whether the word "masu" matches or is similar to the words registered in the list. Only the words "handling", "ga" and "with" are registered in the list. For this reason, the word “masu” is evaluated as “no match/similarity”. Words for which an evaluation result of "no match/similarity" is obtained are registered in the list. In the example of FIG. 10C, list No. 4 shows that the word "masu" has been registered. Then, No. in the list of FIG. 10D. 4 shows an example in which the word "masu" is registered.

Next, it is determined whether the word "." matches or is similar to the words registered in the list. Only the words "handle", "ga", "ari" and "masu" are registered in the list. For this reason, the word "." is evaluated as "no match and similarity". Words for which an evaluation result of "no match/similarity" is obtained are registered in the list. In the example of FIG. 10C, list No. 5 indicates that the word "." has been registered. Then, No. in the list of FIG. 10D. 5 shows an example in which the word "." is registered.

Next, it is determined whether the word "handling" matches or is similar to the words registered in the list. Only the words "handle", "ga", "ari", "masu" and "." are registered in the list. Therefore, the word "handling" is evaluated as "matching No. 1". Words that are evaluated as "matched" are not newly registered in the list.

Next, it is determined whether the word "wa" matches or is similar to the words registered in the list. Only the words "handle", "ga", "ari", "masu" and "." are registered in the list. For this reason, the word "ha" is evaluated as "no match/similarity". Words for which an evaluation result of "no match/similarity" is obtained are registered in the list. In the example of FIG. 10C, list No. 6 shows that the word "ha" has been registered. Then, No. in the list of FIG. 10D. 6 shows an example in which the word "wa" is registered.

Next, it is determined whether the word "yes" matches or is similar to the words registered in the list. In the list, words such as "handle", "ga", "ari", "masu", "." and "ha" are registered. Therefore, the word "yes" is evaluated as "similar" to "yes" in list No. 3. Words evaluated as "similar" are not newly registered in the list. A word evaluated as “similar” is added to the elements of words determined to be similar. Here, No. of the list of FIG. 10D. An example in which the word "yes" is added to element 1 of 3 is shown.

It should be noted that whether words are similar to each other may be determined arbitrarily. For example, consider a case in which the proofreading support apparatus 10 extracts candidates for spelling variations, and the user determines whether or not to proofread them so that the spellings are unified. In this case, it is preferable that words to be corrected for spelling variation be determined to be “similar” to each other in the sentence.

For example, in the sentence of FIG. 10A, it is preferable that "handling", "handling", and "handling" are determined to be similar words. If the descriptions of "handling", "handling", and "handling" are mixed in the text, the proofreader will judge these descriptions as notational variations and proofread so that they are unified to one of the descriptions. This is because it is usually considered that In addition, it is preferable that the words "with" and "with" are determined to be similar words. If the text contains both "Yes" and "Yes" statements, the proofreader usually judges these statements as notational variations and corrects them so that they are consistent with one of the statements. This is because it is conceivable.

Note that the example of FIG. 10C illustrates a case in which ``with'' and ``not'' are determined to be dissimilar words. However, it is not limited to this. It may be determined as a word similar to "with" and "not". For example, if there is a mixture of "yes" and "no" in the text, the proofreader will determine that the descriptions of "yes" and "no" are notational variations, and unify them with one of them. , the proofreading support apparatus 10 is programmed to determine that "yes" and "no" are similar words.

As shown in the example of FIG. 10C, in extracting spelling variations, words in a sentence are evaluated in order. Then, based on the evaluation results, words are registered in the list. When evaluating a word in a sentence, it is necessary to perform a process of comparing with all the words registered in the list. Therefore, the time required for evaluation increases according to the number of words registered in the list. For example, as shown in the item "number of words registered in list" in FIG. 10C, the number of words registered in the list is finally "9" in the sentence shown in the example of FIG. 10A.

If the sentence is short and the number of extracted words is small, the evaluation does not take much time. However, if the sentence is long and the number of words to be extracted is large, the evaluation will take an enormous amount of time, which is not realistic.

As a countermeasure against this, in the present embodiment, the sentence is divided. By dividing the text, the text becomes shorter, and the time required for evaluation can be shortened. However, by only evaluating the divided sentences, it is not possible to match the notation variations in the entire sentence. For example, when a sentence is divided into two, the first half of the sentence contains the word “handling” without any spelling variations, and the second half of the sentence contains the word “handling” without any spelling variations. There may be cases where In this case, there is no notation variation in the sentences divided into the front and back. However, in the entire text, there are variations in the notation of "handling" and "handling". There is a risk of overlooking spelling variations only by evaluating the divided sentences.

As a countermeasure against this, in the present embodiment, divided sentences are connected and the connected sentences are evaluated. This ensures that spelling variations in the concatenated sentences are consistent. In addition, by making a round-robin combination for the combination of the connected sentences, the notation variations in the entire sentences are matched.

FIG. 1 is a diagram illustrating processing performed by a proofreading support apparatus 10 according to an embodiment. The upper part of FIG. 1 shows target sentences ABCD. A target sentence ABCD is a sentence to be proofread. In the example shown in this figure, the target sentence ABCD is a sentence that reads, "There is handling. There is handling. There is handling. There is no handling."

First, the proofreading support apparatus 10 generates divided sentences by dividing the target sentence. In the example of this figure, an example in which the target sentence ABCD is divided into four divided sentences A to D is shown.

Next, the proofreading support device 10 generates a connected sentence by connecting the divided sentences. This figure shows an example in which connected sentences AB, AC, AD, BC, BD, and CD are generated by connecting two different divided sentences among divided sentences A to D. FIG.

Then, the proofreading support apparatus 10 evaluates each of the concatenated sentences AB, AC, AD, BC, BD, and CD, and generates respective lists (list information 120 described later). If there is a word registered redundantly in each list, the proofreading support apparatus 10 merges the word with its element group. The merged list matches the list that would be produced if the entire sentence were evaluated.

Here, the proofreading support apparatus 10 may process the evaluation of each of the concatenated sentences AB, AC, AD, BC, BD, and CD in parallel. For example, the proofreading support device 10 processes the evaluation of each connected sentence in parallel. In this case, a number of evaluations corresponding to the number of divisions are processed in parallel. By increasing the number of divisions, it is possible to shorten the time required for evaluation. On the other hand, it is conceivable that the total number of calculations increases compared to the case of evaluating the entire sentence. However, due to recent improvements in the performance of computers, the increase in the total number of calculations does not significantly affect the processing time. That is, even if the total amount of calculation increases, the effect of shortening the processing time by performing parallel processing is great. Therefore, even a long sentence does not increase the time required for proofreading.

It should be noted that the number of divided sentences into which the proofreading support apparatus 10 divides the target sentence may be arbitrarily determined. Considering concatenation, it does not make sense to divide the target sentence into two and concatenate the two divided sentences. Therefore, the proofreading support apparatus 10 may divide the target sentence into at least three parts.

FIG. 2 is a block diagram showing a configuration example of the proofreading support device 10 according to the embodiment. The proofreading support device 10 is a computer device that extracts character strings that are candidates for spelling variation in a sentence to be proofread (target sentence). As the proofreading support device 10, for example, a server device, a cloud, a PC (Personal Computer), etc. can be applied.

The proofreading support device 10 includes, for example, a communication unit 11, a storage unit 12, and a control unit 13. The communication unit 11 communicates with an external device via a communication network or the like. The communication unit 11 receives, for example, text information indicating a target sentence from an external server device or the like. In addition, the communication unit 11 transmits, as proofreading results, character strings that are candidates for spelling variations extracted from the target text to a server device or the like provided outside.

The storage unit 12 is a storage medium such as a HDD (Hard Disk Drive), flash memory, EEPROM (Electrically Erasable Programmable Read Only Memory), RAM (Random Access read/write Memory), ROM (Read Only Memory), or a combination thereof. Consists of The storage unit 12 stores programs for executing various processes of the proofreading support apparatus 10 and temporary data used when performing various processes. The storage unit 12 stores list information 120, for example. The list information 120 is information about candidates for spelling variation extracted from the target sentence.

The control unit 13 is realized by causing a CPU (Central Processing Unit) provided as hardware in the proofreading support apparatus 10 to execute a program. The control unit 13 comprehensively controls the proofreading support device 10 . The control unit 13 includes, for example, an acquisition unit 130, a division unit 131, a connection unit 132, an evaluation unit 133, and a device control unit .

Acquisition unit 130 acquires a target sentence. For example, the acquisition unit 130 acquires information indicating the target sentence from an external server device or the like via the communication unit 11 . The acquiring unit 130 outputs information indicating the acquired target sentence to the dividing unit 131 .

The dividing unit 131 generates divided sentences based on the target sentence. For example, the dividing unit 131 generates divided sentences such that the number of words included in each divided sentence is approximately the same. In this case, the dividing unit 131 divides the target sentence into words and calculates the number of words described in the target sentence. The dividing unit 131 searches for a boundary at which the target sentence is divided using character strings before and after the description corresponding to the number of words obtained by dividing the calculated number of words by the number of divisions.

The dividing unit 131 uses, for example, periods, symbols, and commas as boundaries for dividing the target sentence. Symbols are, for example, an exclamation mark "!", a question mark "?", a musical note "♪", a colon ":", and the like. The reason why punctuation marks, symbols, and commas are used as boundaries is that they are less likely to be subject to notational variations. The dividing unit 131 at least does not set a boundary in the middle of a word. For example, if a boundary is set between ``handle'' and ``i'' in the word ``handle'', the word ``handle'' that was originally ``handle'' becomes ``handle'' This is because there is a possibility that it will be evaluated as a word "handled" and an erroneous evaluation will be made.

The dividing unit 131 determines boundaries for dividing the target sentence, and generates a plurality of divided sentences by dividing at the determined boundaries. The dividing unit 131 outputs information indicating the generated divided sentences to the connecting unit 132 .

The linking unit 132 generates linked sentences. The connecting unit 132 generates, as connected sentences, all patterns that can be considered as a combination of two divided sentences that are different from each other, for example. The linking unit 132 outputs information indicating the generated linked sentences to the evaluation unit 133 .

The evaluation unit 133 evaluates each of the connected sentences. For example, the evaluation unit 133 divides the connected sentence into words, and sequentially determines whether the divided words match or are similar to the words registered in the list (list information 120). If the word neither matches nor resembles a word registered in the list, the evaluation unit 133 registers the word in the list. Moreover, when a word is similar to a word registered in the list, the evaluation unit 133 adds the word as an element of the word registered in the list (similar word).

The device control unit 134 controls the proofreading support device 10 in an integrated manner. For example, the device control unit 134 outputs text information indicating the target sentence received by the communication unit 11 to the acquisition unit 130 . The device control unit 134 merges the list (list information 120) generated as a result of evaluation by the evaluation unit 133. FIG. The device control unit 134 transmits the merged list to an external server device as a list indicating candidates for spelling variations.

3-8 are diagrams illustrating examples of list information 120 according to embodiments. 3 to 8 show items such as list number, word, element 1, element 2, and so on. Since these items are the same as the items shown in the table of FIG. 10D, the description thereof is omitted.

FIG. 3 shows an example of the list generated corresponding to the concatenated text AB as list information 120A. In the list information 120A, the word "with" described in the 6th and 7th characters and the word "with" described in the 16th and 17th characters in the concatenated sentence AB are candidates for spelling variations. An extracted example is shown.

FIG. 4 shows an example of the list generated corresponding to the concatenated text AC as list information 120B. In the list information 120B, the word "handling" described in the 1st to 4th characters and the word "handling" described in the 21st to 23rd characters in the concatenated sentence AC are candidates for spelling variations. An extracted example is shown.

FIG. 5 shows an example of the list generated corresponding to the concatenated text AD as list information 120C. In the list information 120C, the word "handling" described in the 1st to 4th characters and the word "handling" described in the 30th to 31st characters in the concatenated sentence AD are candidates for spelling variations. An extracted example is shown.

FIG. 6 shows an example of the list generated corresponding to the concatenated text BC as list information 120D. In the list information 120D, the word "handling" described in the 11th to 14th characters and the word "handling" described in the 21st to 23rd characters in the concatenated sentence BC are candidates for spelling variations. An extracted example is shown. Also, in the list information 120D, the word "Yes" described in the 16th and 17th characters and the word "Yes" described in the 25th and 26th characters in the concatenated sentence BC are notation variations. An example extracted as a candidate for is shown.

FIG. 7 shows an example of the list generated corresponding to the linked text BD as list information 120E. In the list information 120E, the word "handling" described in the 11th to 14th characters and the word "handling" described in the 30th to 31st characters in the concatenated sentence BD are candidates for spelling variations. An extracted example is shown.

FIG. 8 shows an example of the list generated corresponding to the concatenated text CD as list information 120F. In the list information 120F, the word "handling" described in the 21st to 23rd characters and the word "handling" described in the 30th to 31st characters in the concatenated sentence CD are candidates for spelling variations. An extracted example is shown.

Here, a method for merging the lists (list information 120) by the device control unit 134 will be described. First, the device control unit 134 identifies the positions where the candidates for the spelling variation shown in each list are described, and merges the words having the duplicate identified positions.

For example, when list information 120A to 120F as shown in FIGS. For the spelling variants of "" and "aru", identify the position where the word is written in each list.

The position where the word “Yes” is written in the list information 120A is the 6th or 7th character. The position where the word “present” is written in the list information 120A is the 16th to 17th characters. The position where the word “present” is written in the list information 120D is the 16th to 17th characters. The position where the word “Yes” is written in the list information 120D is the 25th to 26th characters.

In each list, the position where the word "presence" is described overlaps at the 17th and 18th characters. In this case, the device control unit 134 merges the spelling variations of "yes" and "yes" in the list information 120A and the spelling variations of "yes" and "yes" in the list information 120D as the same group. As a result, the 7th and 8th characters "Ari", the 16th and 17th characters "Ari", and the 25th and 26th characters "Ari" are candidates for notation variation in the target sentence.

In addition, the device control unit 134 controls the notation variation of "handling" and "handling" in the list information 120B, the notation variation of "handling" and "handling" in the list information 120C, and the notation variation of "handling" and "handling" in the list information 120D. With regard to spelling variations, spelling variations of "handling" and "handling" in the list information 120E, and spelling variations of "handling" and "handling" in the list information 120F, the positions where the words are described in each list are specified.

The position where the word “handling” is written in the list information 120B is the 1st to 4th characters. The position where the word “handling” is written in the list information 120B is the 21st to 23rd characters. The position where the word “handling” is written in the list information 120C is the 1st to 4th characters. The position where the word “handling” is written in the list information 120C is the 30th to 31st characters.

The position where the word “handling” is written in the list information 120D is the 11th to 14th characters. The position where the word “handling” is written in the list information 120D is the 21st to 23rd characters. The position where the word “handling” is written in the list information 120E is the 11th to 14th characters. The position where the word “handling” is written in the list information 120E is the 30th to 31st characters.

The position where the word “handling” in the list information 120F is written is the 21st to 23rd characters. The position where the word “handling” is written in the list information 120F is the 30th to 31st characters.

The device control unit 134 merges the lists in which the position where the word “handling” is described overlaps in the 1st to 4th characters. The device control unit 134 merges the notation variation of "handling" and "handling" in the list information 120B and the notation variation of "handling" and "handling" in the list information 120C as the same group. As a result, the 1st to 4th characters "handling", the 21st to 23rd characters "handling", and the 30th to 31st characters "handling" are candidates for notation variations in the target sentence.

In addition, the device control unit 134 merges the 21st to 23rd characters in which the position where the word “handling” is written overlaps in each list. The device control unit 134 controls the notation variation of "handling" and "handling" in the list information 120B, the notation variation of "handling" and "handling" in the list information 120D, and the notation variation of "handling" and "handling" in the list information 120F. The shakes are regarded as the same group and merged. As a result, the 1st to 4th characters "handling", the 11th to 14th characters "handling", the 21st to 23rd characters "handling", and the 30th to 31st characters "handling" It is a candidate for notational variation in

In addition, the device control unit 134 merges the 11th to 14th characters in which the position where the word “handling” is described overlaps in each list. The device control unit 134 merges the notation variation of "handling" and "handling" in the list information 120D and the notation variation of "handling" and "handling" in the list information 120E as the same group. As a result, the 11th to 14th characters "handling", the 21st to 23rd characters "handling", and the 30th to 31st characters "handling" are candidates for notation variations in the target sentence.

From the above, the 1st to 4th characters "handling", the 21st to 23rd characters "handling", and the 30th to 31st characters "handling" are candidates for spelling variation in the target sentence (first list) is generated. In addition, the 1st to 4th letters “handling”, the 11th to 14th letters “handling”, the 21st to 23rd letters “handling”, and the 30th to 31st letters “handling” are A list (second list) of candidates for spelling variation is generated. The 11th to 14th characters "Handling", the 21st to 23rd characters "Handling", and the 30th to 31st characters "Handling" are candidates for spelling variation in the target sentence (third list). generated. In this case, the respective elements of the first to third lists overlap each other. In this case, the device control unit 134 merges the first to third lists into one list. As a result, the 1st to 4th characters "handling", the 11th to 14th characters "handling", the 21st to 23rd characters "handling", and the 30th to 31st characters "handling" It is a candidate for notational variation in

FIG. 9 is a flow chart showing the flow of processing performed by the proofreading support apparatus 10 according to the embodiment. The proofreading support apparatus 10 acquires a proofreading target (target sentence) (step S10), divides the acquired target sentence (step S11), and generates divided sentences. The proofreading support device 10 connects the divided sentences (step S12) to generate a connected sentence. The proofreading support device 10 acquires the concatenated sentence (step S13), and by evaluating the acquired concatenated sentence, extracts candidates for spelling variations in the concatenated sentence (step S14). The proofreading support apparatus 10 determines whether or not all the connected sentences have been evaluated (step S15), and if there are any connected sentences that have not yet been evaluated, the process returns to step S13. When all the connected sentences have been evaluated, the proofreading support device 10 merges words registered in duplicate in the list (list information 120) generated along with the evaluation (step S16). The proofreading support apparatus 10 transmits the merged list to, for example, an external server apparatus as a result of extracting spelling variation candidates.

As described above, the proofreading support apparatus 10 of the embodiment includes the acquisition unit 130 , the division unit 131 , the connection unit 132 and the evaluation unit 133 . Acquisition unit 130 acquires a target sentence. The dividing unit 131 generates divided sentences by dividing the target sentence into at least three or more. The concatenating unit 132 generates a concatenated sentence that is formed by concatenating different divided sentences among the divided sentences and that is shorter than the target sentence. The evaluation unit 133 extracts character strings that are candidates for spelling variation in the concatenated sentence. As a result, the proofreading support apparatus 10 of the embodiment can extract character strings that are candidates for spelling variations in concatenated sentences that are shorter than the target sentence. Therefore, even if the target sentence is long, a connected sentence shorter than the target sentence can be corrected. Therefore, it is possible to extract character strings that are candidates for spelling variation without increasing the time required for proofreading even in long sentences.

Further, in the proofreading support apparatus 10 of the embodiment, the evaluation unit 133 compares the target word included in the connected sentence with the registered word registered in the list information 120 showing the list of words for determining the presence or absence of spelling variations. The evaluation unit 133 registers the target word in the list information 120 when the target word does not match or is not similar to the registered word. When the target word is similar to the registered word, the evaluation unit 133 adds the target word to the elements of registered words similar to the target word in the list information 120 . Of the registered words registered in the list information 120, the evaluation unit 133 sets the registered words and the words added to the elements of the registered words as candidates for spelling variation. As a result, the proofreading support apparatus 10 of the embodiment compares the target word included in the connected sentence with the list information 120, registers it if it does not match or is similar, and adds it to the element if it is similar. , it is possible to extract candidates for spelling variation.

Further, the proofreading support device 10 of the embodiment further includes a device control section 134 . Based on the list information 120 generated corresponding to each of the linked sentences, the device control unit 134 selects, among the registered words registered in each of the list information 120, the registered words that are redundantly registered in the plurality of list information 120. Merge registered words. As a result, the proofreading support apparatus 10 of the embodiment can extract spelling variations in the target sentence without duplication based on the list information 120 of each of the linked sentences.

Now consider a comparative example. As a measure to reduce the time required for evaluation, it is possible to prepare a list of words that are likely to have spelling variations, and check only the words registered in the list for spelling variations to speed up proofreading. Conceivable. However, this method cannot check spelling variations of words that are not registered in the list in advance. For this reason, there is a possibility of overlooking notation variations.

On the other hand, in this embodiment, list information 120 corresponding to connected sentences is generated. The list information 120 is created according to whether the word described in the connected sentence matches or is similar to another word. Therefore, a corresponding connected sentence is generated. Therefore, it is possible to reduce the possibility of overlooking spelling variations and reduce the risk of overlooking spelling variations.

In addition, in the proofreading support apparatus 10 of the embodiment, the linking unit 132 generates linked sentences corresponding to all combinations of two divided sentences that are different from each other. As a result, with the proofreading support apparatus 10 of the embodiment, even if the notation standardized in one of the divided sentences corresponds to the notation variation in the other divided sentence, Variation can be extracted, and the risk of overlooking notation variation can be reduced.

Further, in the proofreading support apparatus 10 of the embodiment, the evaluation unit 133 executes, in parallel, a process of extracting character strings that are candidates for spelling variation for each of the concatenated sentences. As a result, the proofreading support apparatus 10 of the embodiment can shorten the time required for evaluation.

In addition, in the proofreading support apparatus 10 of the embodiment, the dividing unit 131 divides the target sentence using any of the periods, symbols, and commas in the target sentence as boundaries. As a result, in the proofreading support apparatus 10 of the embodiment, boundaries are not set in the middle of words, and the risk of overlooking spelling variations can be reduced.

All or part of the proofreading support apparatus 10 in the above-described embodiment may be implemented by a computer. In that case, a program for realizing this function may be recorded in a computer-readable recording medium, and the program recorded in this recording medium may be read into a computer system and executed. It should be noted that the "computer system" referred to here includes hardware such as an OS and peripheral devices. The term "computer-readable recording medium" refers to portable media such as flexible discs, magneto-optical discs, ROMs and CD-ROMs, and storage devices such as hard discs incorporated in computer systems. Furthermore, "computer-readable recording medium" means a medium that dynamically retains a program for a short period of time, like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. It may also include something that holds the program for a certain period of time, such as a volatile memory inside a computer system that serves as a server or client in that case. Further, the program may be for realizing a part of the functions described above, or may be capable of realizing the functions described above in combination with a program already recorded in the computer system. It may be implemented using a programmable logic device such as FPGA.

Although the embodiment of the present invention has been described in detail with reference to the drawings, the specific configuration is not limited to this embodiment, and design and the like are included within the scope of the gist of the present invention.

DESCRIPTION OF SYMBOLS 10... Proofreading support apparatus 130... Acquisition part 131... Division part 132... Connection part 133... Evaluation part

Claims (9)

an obtaining unit that obtains a target sentence to be proofread; a dividing unit that generates divided sentences by dividing the target sentence into at least three or more;
a connecting unit for generating a connected sentence formed by connecting different divided sentences among the divided sentences, the connected sentence being shorter than the target sentence;
an evaluation unit that extracts a character string that is a candidate for spelling variation in the concatenated sentence;
A calibration support device. The evaluation unit
Comparing the target word contained in the linked sentence with the registered word registered in the list information indicating the list of words for determining the presence or absence of spelling variations,
if the target word does not match or is similar to the registered word, registering the target word in the list information;
when the target word is similar to the registered word, adding the target word to the elements of the registered word similar to the target word in the list information;
The calibration support device according to claim 1. The evaluation unit regards, among the registered words registered in the list information, the registered words and the words added to the elements of the registered words as candidates for spelling variations,
The calibration support device according to claim 2. Based on the list information generated corresponding to each of the linked sentences, among the registered words registered in each list information, the registered words duplicately registered in a plurality of the list information are merged. further comprising a device control unit that causes
The calibration support device according to claim 3. The connecting unit generates the connected sentences corresponding to all combinations of the two divided sentences that are different from each other.
The proofreading support device according to any one of claims 1 to 4. The evaluation unit executes, in parallel, a process of extracting character strings that are candidates for spelling variation for each of the concatenated sentences.
The proofreading support device according to any one of claims 1 to 5. The dividing unit divides the target sentence using any one of a period, a symbol, or a comma in the target sentence as a boundary.
The proofreading support device according to any one of claims 1 to 6. A proofreading support method performed by a computer, wherein an acquisition unit acquires a target sentence to be proofread,
The dividing unit generates divided sentences by dividing the target sentence into at least three or more,
generating a connected sentence in which the connecting part is a connected sentence formed by connecting different divided sentences among the divided sentences, the connected sentence being shorter than the target sentence;
The evaluation unit extracts a character string that is a candidate for spelling variation in the concatenated sentence;
Proofreading assistance method. A program for causing a computer to operate as the proofreading support device according to any one of claims 1 to 7, the program causing the computer to function as each unit provided in the proofreading support device.

Images