JP2000057144A - Document editing device and document editing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a document editing device which can easily perform a processing for making a 2nd language document obtained as translation results of a 1st language document natural as a document of the 2nd language by translating the 1st language document into the 2nd language document and copying prescribed attributes on natural language processing constructing the 2nd language document into another sentence. SOLUTION: An original in Japanese sentences made in a 1st language is read from an inputting part 101. A translation managing part 102 stores the document data of the original of the received 1st language in a translation database 104. A translation engine 103 performs machine translation of the original sent from the translation managing part 102 and stores translation results together with syntax analysis results in the translation database 104. Finally, the results are outputted by an outputting part 105 and the processing is finished. Thus, it is possible to rewrite a desired sentence, etc., into a desired style or tone without being troubled by a small operation.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a document editing apparatus and a document editing method for editing a document in a natural language.

[0002]

2. Description of the Related Art As the internationalization and borderlessness of business have progressed, the network society has progressed, and digital information exchange has become commonplace such as the Internet and e-mail. With the spread of such a network environment, general users frequently acquire information on the Internet through WWW.

[0003] In addition, recently, as a response to the information thus obtained, attention has been paid to the role of an information sender to transmit information from here. In the case of the recipient, the ability to read and listen was sufficient,
As a sender, a more aggressive "writing / speaking" operational ability is indispensable. Therefore, in a language other than the native language, it is very difficult to speak or write a language that is natural for the native speaker.

On the other hand, at present, translation software whose quality and accuracy are not perfect is facing the same situation as humans. Translation software basically translates each sentence without looking at the surrounding text, so if you read the translation result as a whole document, the style and tone are consistent even if the objective facts are correctly communicated Not so often awkward. From the standpoint of information dissemination, it is necessary to increase persuasiveness and impact by changing the style to a sharp one and adding techniques. As mentioned above, such editing ability is particularly difficult in the case of foreign languages, but there is already a method of supporting it from the machine side at the word level, but the stylistic level has not been realized yet. .

[0005]

Conventionally, the translation is performed one sentence and one sentence uniformly without considering the style and tone of phrases and sentences constituting the document, so that the resulting document is read throughout. And tended to lack nature. However, a method of (semi-) automatically assisting document editing in this aspect has not been established, and users have had to ask native speakers for correction. In addition, when a user corrects a sentence on the document editing apparatus, he / she is burdened with accurate operations such as deletion, insertion, and replacement, thereby imposing a heavy load.

SUMMARY OF THE INVENTION The present invention has been made in consideration of the above circumstances, and is intended to make a second language document obtained as a translation result of a first language document natural as the second language document. It is an object of the present invention to provide a document editing apparatus and a document editing method that can easily perform the above processing.

[0007]

A document editing apparatus according to the present invention (claim 1) translates a document in a first language into a document in a second language by natural language processing (eg, syntax analysis). A predetermined attribute (e.g., syntax attribute, pronunciation information, etc.) of one sentence constituting the second language document or a part thereof in the natural language processing is changed to another sentence constituting the second language document or the other sentence thereof. And copying means for copying a portion.

[0008] Preferably, a means for restoring all or a part of the predetermined attribute to the other sentence or a part thereof whose predetermined attribute has been changed by the copying means may be further provided. For example, information obtained at the time of syntax analysis may be added or canceled.

Preferably, when the copying unit performs a change process for changing the attribute of the other sentence or a part thereof, a unit for designating which attribute in the one sentence or a part thereof is focused on. And the copying unit may perform the change processing based on the designation.

The document editing apparatus according to the present invention (claim 4) has means for designating a first character string as a copy source of a predetermined feature, and a second character as a copy destination of a predetermined feature. Means for designating a string, and transforming the second character string so that the designated second character string has predetermined characteristics of the designated first character string. And copying means for processing.

The document editing apparatus according to the present invention (Claim 5) translates a document in a first language into a document in a second language by natural language processing (for example, syntax analysis) and constructs the document in the second language. A predetermined attribute (for example, syntax attribute, pronunciation information, etc.) of one sentence or a part thereof in natural language processing is copied to another sentence or a part thereof constituting the second language document. I do.

The present invention (claim 6) provides a computer
A first language document is translated into a second language document by natural language processing (for example, syntax analysis), and a predetermined attribute in the natural language processing for one sentence or a part thereof constituting the second language document A gist of the present invention is a computer-readable recording medium storing a program for copying (for example, syntax attributes, pronunciation information, and the like) into another sentence or a part of the second language document.

Note that the present invention relating to the apparatus is also realized as an invention relating to a method, and the present invention relating to a method is also realized as an invention relating to an apparatus. Further, the present invention according to an apparatus or a method has a function of causing a computer to execute a procedure corresponding to the present invention (or causing a computer to function as means corresponding to the present invention, or a computer having a function corresponding to the present invention). The present invention is also realized as a computer-readable recording medium on which a program (for realizing the program) is recorded.

According to the present invention, a desired sentence or phrase can be defined without using explicit or grammatical terms in the grammatical terms and without troublesome operations such as deletion, insertion, and replacement. The style can be easily changed.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. Embodiment FIG. 1 shows a configuration example of a document editing apparatus according to an embodiment of the present invention.

As shown in FIG. 1, this document editing apparatus includes an input unit 101, a translation management unit 102, a translation engine 1
03, a translation database 104, and an output unit 105.

The input unit 101 receives original texts and commands through communication means such as the Internet, input devices such as a keyboard, and the like. Translation management unit 10
Reference numeral 2 is for managing translation requests. More specifically, job management, document management, document comparison, and the like are performed.

The translation engine 103 includes a translation management unit 102
Machine translation of the original text according to the content sent from
Output the result. The machine translation method is not particularly limited, and any currently known method may be used.

The translation database 104 is for temporarily or long term storage of original sentences, translation results, and the like. The output unit 105 is for presenting the translation result stored in the translation database 104 to the user.

It is more preferable that the document editing apparatus has a general editing function for a user to perform operations such as addition, deletion, and replacement of words and the like. Next, a document editing process performed by the document editing apparatus according to the present embodiment will be described.

First, the flow of a translation process for a document input from the input unit 101 will be described using a specific example in which the first language is Japanese and the second language is English. In FIG.
An example of the procedure of the translation process will be described.

Here, as a specific example, a case where the document shown in FIG. 3 is translated will be described. First, the input unit 1
From 01, the original sentence of the Japanese sentence of the first language is read. The translation management unit 102 stores the received data of the original document in the first language in the translation database 104 (step S1).

Next, the translation engine 103 machine-translates the original sent from the translation management unit 102 (step S).
2) The translation result is stored in the translation database 104 together with the syntax analysis result (step S3), and finally the result is output by the output unit 105 (step S4), and the process ends.

Next, a post-editing process for a translation result will be described. FIG. 4 shows an example of the procedure of the post-editing process. Here, as a specific example, a case where post-editing work is performed on the document shown in FIG. 5 obtained as a translation result of the document shown in FIG. 2 will be described. Note that (1), (2),..., (8) shown in FIG. 5 are provided for reference of description, and are not document data.

As a post-edit, the user first enters the sentence (7) "The planse to c" at the end of the document in FIG.
all ... "in the first sentence (1)" It seedstha
The following operation is performed to rewrite the style as "t ...".

For the sake of simplicity, the sentence (7) will be described in a shorter "Theplan seeds".
to call for aboringmon
oplies and other privacy
ges. It is described as "."

As a method of designating a sentence or the like, for example, there is the following method. The document shown in FIG.
The portion corresponding to the sentence (1) is inverted using a mouse or the like, and then, in the state of being inverted, from the menu of the edit command key as shown in FIG. Click the "" item. Next, a copy destination of the style, here, a portion corresponding to the sentence (7) is selected using a mouse or the like. It should be noted that various other methods can be used as the operation method.

Then, the translation management unit 102 extracts each stored syntax analysis result from the translation database 105 (step S11). FIG. 8 shows an example of the tree structure of the sentence (1). FIG. 9 shows an example of the tree structure of the sentence (7).

Next, the translation management unit 102 compares the extracted syntax analysis results (step S1).
2). For example, referring to the syntax dictionary shown in FIG.
By comparing the parsing results of the two, it can be seen that the former sentence is a "sem to ..." syntax and the latter sentence is an "it seeds + clause" syntax having a formal subject of it.

Then, using the syntax conversion rule shown in FIG. 11, the “seem to.
+ Clause "syntax. FIG. 12 shows an example of the tree structure after the conversion.

Next, a rewriting process is performed while using the grammar of the generation unit of the translation engine 103 (that is, the function of generating a sentence from a tree structure) as necessary (step S1).
3). For example, first, the former sentence is formed by combining the semantic subject "the plan" with its predicate "call" and the like. The tense is copied from "sees", and "call" is also a third person singular present. The detailed operation of utilizing such a verb is an existing technique, and a description thereof will be omitted.
Finally, before this section, connect the main sentence “It seeds” to the desired “It seeds that th
e plan calls forablishin
g monopolies and otherpri
villages that have made ○○
○○ 's family and associates
weaky over years. "
Get.

Then, the newly obtained English sentence is displayed at the position where the original English sentence (7) was located (step S1).
4), end the process. Next, as another example of rewriting, the sentence “On the control” in (5) of FIG.
y, "The people's life isbe
coming more difficult, "sa
id thepresident. ”To“ ○ ”in (6).
○○ did notspecify what re
form changes hereto to
keep the economy going, bu
tsid new measures cord
beintroduced. ", And a request for rewriting is issued.

In this case, due to the use of quotation marks and upper / lower case, the system is a direct speech method in which the former sentence transmits the speaker's word as it is during parsing, and the latter sentence uses the speaker's word in parsing. We judge that it is an indirect speech style that conveys the contents in the words of the person who communicates. More specifically, the former is "subject indicating the transmitting subject + transmitting verb + comma + quote +
Transmitted part + quote + period "," quote + transmitted part +
Comma + quote + communication verb + subject + communication verb "or" quote + transported part + comma + quote + communication verb + subject subject ". The latter is "subject indicating the transmitting subject + transmitting verb + (tha
t) + transmitted part ".

Therefore, using these pieces of information, the quotation mark in the former sentence and the comma at the end thereof are first deleted, and
Next, the tense of the transmitted part is adjusted according to the tense of the transmission verb. In this case, the transfer verb "say" is replaced by the past tense "sai
d ", so in direct speech," is beco
ming "and what was currently in progress is" wasbec
oming ”and the past progressive form. In this intermediate stage, "On the control, The people"
e's life wasbecoming more
diffult Said thepressed
ent. To get. Here, "the people's
life was becoming more di
"ficulult" is a transmission part, "said" is a transmission verb,
Since “the president” is the subject indicating the transmitting entity, the order is rearranged in the specified order and “On the
control, the present sa
id that The People's Life
was becoming moredifficul
t. To get. Finally, the necessary uppercase and lowercase characters are processed, and the final result is “On the control, t
he presidentsid that that the
people's life wasbecomin
g more difficult. ".

Next, it is assumed that there is no sentence having a desired style in the machine translation result by another rewriting method. In such a case, the user may be able to refer to another document and instruct to reflect the style. In this case, since parsing is not performed for the referenced statement, the minimum parsing necessary for rewriting is performed before automatic rewriting (however, there is a parsing result that can be used for the referred statement. In that case, the result may be used).

FIG. 13 shows an example of the processing procedure in this case. Here, as a specific example, in a document whose original sentence is in English, “We woodnot have gone,
had I known this worldhap
pen. , And want to rewrite the English sentence (8) being edited later shown in FIG. 5 to this sentence pattern.

First, when the above request is received, an English sentence that is a source of imitation is parsed (step S21).
"Had I know this woodwha
It is determined from the part of speech of the word and its arrangement that the “ppen” part is inverted.

Next, the analysis result of the English sentence (8) is extracted from the translation database 104 (step S22). Then, the respective syntax analysis results are compared as described above (step S23).

By the way, this type of inversion clause is represented by "auxiliary verb + subject + past participle", and the usual unmarked phrase is represented by "if + subject + auxiliary verb + past participle". I have. From this, the former is prefixed with “i
By inserting "f" and exchanging the word order between the subject and the auxiliary verb, it can be seen that the latter sentence pattern is obtained.

Utilizing this fact, the English sentence (8) is described as “So
me people claimthat the n
ation's economy worldnever
r have been in this message,
hadtheir monopolyes been
abolished. (Step S2
4).

Finally, at the position where the English sentence (8) was located, the newly obtained "Some people claim that the
t the nation's economywow
ld never have been in thi
smess, had the monopoly
s beenablished. Is displayed (step S25), and the process ends.

In the following, some functions that are preferably added are described. First, it is preferable to add a function of canceling the style once copied.

For example, a morphological analysis unit of the translation engine 103 (ie, a function of performing a morphological analysis of a target sentence)
Are "marked", "unmarked (unmark)
ed) ", and when a sentence of" marked "is specified and a request for deletion is issued, a process of converting the sentence to" unmarked "may be performed.

FIG. 14 shows an example of the procedure of the cancellation process. Here, as a specific example, “Not a sing
le myskedid I find in y
our composition. "
Consider deleting the "marked" syntax information of this sentence. Since this sentence is an inverted sentence and the inverted is "marked", deleting it is equivalent to converting it to a sentence in normal word order.

Upon receiving the request for deletion, the translation management unit 102 extracts the syntax analysis result of this sentence from the translation database 104 (step S31), and from this syntax analysis result, the syntax information ("marked") Here, inversion is extracted and presented (step S32).

Subsequently, when the confirmation of the syntax information to be deleted is obtained, for example, "Not a single mission"
e did I find in your compo
site. "To" id not find "
asingle missake in your c
ompposition. (Step S3
3) This is displayed (step S34), and the process ends.

Next, when copying the style, if it is not clear which syntactic feature of the source sentence to focus on, or if it has a plurality of syntactic features, the syntactic feature is presented. It is preferable to add a function that allows the user to select a desired one.

In this case, "Sinc it was"
Mr. Johnson whos killed
in a traffic accident, weh
ave to find another persono
n treatplace him. Is an English sentence (example sentence 1), and the reflection destination of this syntactic feature is “3.
members of the assembly
that heremained committed
to carrying outthe ××× 's
rescue package, but he sa
idadditional strategies w
ee neededas life gets to
ugher. This process will be described.

For example, first, syntax information candidates are displayed in a menu as shown in FIG. Here, the above English sentence (Example sentence 1)
Is (i) “emphasis syntax of it is ... that”, (i
i) It has at least two features of "compound sentence including adverb clause", and these are determined at the time of parsing. More specifically, regarding (i), “it is + emphasis element +
It has knowledge that it consists of "the rest of the sentence elements." Also, in the English sentence (example sentence 1), the original subject is an emphasis element, so if it is selected as the syntax information reflected by (i), the English sentence "it is ..." in FIG.
"It was the the emphasized syntax of that"
President who tol the the me
mbers of the assemblyat
he removed committed to
carrying out the ××× 's re
squarepack, but the Said a
dditionalaltiesties were
needed as life getstoughhe
r. Is obtained.

On the other hand, regarding (ii), since (3) in FIG. 5 is a multiple sentence, reflecting this feature in (3) corresponds to conversion from a multiple sentence to a compound sentence. Therefore,
When (ii) is selected as the reflected syntax information, for example, "The Presidency told the
members of the tearsembly th
at the removed committedt
o carrying out the ××× 's
rescuepackage, although he
Said additional strategye
s weird needed as life get
stoger. Is obtained.

Of course, a plurality of features listed as candidates may be selectable. In this example, "it is ... th
If both the “at emphasis syntax of at” and the “compound sentence including adverbial clause” are selected, as a result, for example, “It was the the
President who tol the the me
mbers of the assemblyat
he removed committed to
carrying out the ××× 's re
squarepack, although hes
aid additional strategies
Were needed as life getst
oher. Is obtained.

In FIG. 15, the candidates are shown by the type of syntax. However, as shown in FIG. 16, the output result when each candidate is selected, that is, the corrected English sentence is displayed instead, and a method of selecting it is used. You may take it. Alternatively, as shown in FIG. 17, the type of the syntax and the correction result may be described together.

In this case, a deletion function may be provided. Here, "It was the Presid"
ent who toldthe members o
f the assembly that herem
ainmitted committed to carryy
ng outthe ××× 's rescue pa
cage, alternative hesaid add
itional strategies weene
eded as life gets tough
r. Is a sentence to be edited, for example, first, FIG.
A candidate for deletion is displayed as shown.

Here, if only (i) "it is ... synthesis emphasized syntax" is selected, "the Presi
dent told the members of
he assembly that the remai
necessitated to carrying
out the ××× 'rescue packa
ge, although he Saidadditi
Online Strategies Were Need
ed aslife gets tougher. "
(Ii) If only "compounds containing adverbial clauses" is selected,
"It was the President who
told the members of the a
ssmbly thatthe retained co
millited to carryingout th
e ××× 's rescue package, bu
t hesaid additional strat
egies were needed as life
gets tougher. And (iii) "it i
If both the "s ... that emphasis syntax" and "the compound sentence including the adverbial clause" are selected, "The President t
old the members of the ass
embly that the retainedcom
committed to carrying out th
e ××× 'rescue package, butt
he Saidadditional strate
gies weird need aslife g
ets tougher. ] Respectively.

Note that the candidate syntax information (attribute) is not limited to those exemplified above, and may include, for example, passive voice, independent participle syntax, sentence types (question sentence, additional question sentence,
It is possible to use various things such as rhetorical questions, imperative sentences, exclamation sentences, prayer sentences).

The syntax information (attribute) is copied /
Deletion can be performed collectively. For example, when three pieces of syntax information (attributes) of passive voice, negation, and question are to be operated, on a screen as shown in FIG.
By checking the information to be reflected and unchecking the information that is not reflected, 2 × 2 × 2 = 8 types of rewriting are possible.

That is, "Your mother c
books your meals. "
ur meals are cooked by yo
urmother. ”,“ Your mother d
oesn't cook your meals. "
"Are your meals cooked by
yourother? "," Your meals "
aren't cooked by yourourmot
her. ”,“ Doesn't your mother ”
r cook your meals? ”,“ Are y
our meals cooked by yourm
other? ","Aren't your meal
s cooked by your mother? Is possible as a candidate.

By the way, such rewriting is quite similar to the rewriting often seen in practice problems assigned in school English. Therefore, the syntax that the parser particularly emphasizes in school grammar, for example, “Heis not so
much as ascholar as a jo
urnalist. "Not so as seen in"
The "much-as ..."("rather than ...") syntax is "-rather tan ... (He is a
journalistthanthan as
color. ) "," More ... than- (He
is more a journalisthan
a scholar. ) ”,“ Less-than…
(Heis less a scholarthan)
a journalist. )), It can be used as a drill for English grammar practice exercises, and can be used for educational purposes such as CAI. Can also be applied.

In the above, an example in which the style is copied has been described. However, it is also possible to copy features in another language or language processing. In the above description, the unit of processing is described as one sentence, but other units such as clauses and phrases can also be used.

Here, as another application, an example in which the present invention is used in a word game will be described. In this example, attention is paid not to syntax information but to pronunciation information. If this is used, for example, “The cow is of
thebovine ilk; one end is
moo, the othermilk. (Poet Og
The sentence "den Nash's poem)
It can be seen from the following operation that the part of “and“ milk ”rhymes.

First, the phonetic symbols of each word selected by the dictionary lookup at the time of syntactic analysis are stored as data as shown in FIG. Next, comparing these, ilk
[Ilk] and milk [milk] end with [ilk]
, Which means that they are rhyme. Since each of these words is located at the end of the phrase, the phonetic attribute of the sentence is the rhyme at the end of the word at the end of the phrase.

Therefore, the pronunciation information is, for example, “A
new bell hung; the birds t
twittered. "(Example sentence 2). As shown in FIG. 20, in this English sentence (example sentence 2), the rhyme at the end of the word at the end of the phrase is not obtained.

Thus, reflecting the above pronunciation information in the above English sentence (example sentence 2) is based on “hung” and “twit”.
This means that the ending of the word that comes in the position of “tered” rhymes.

Therefore, 1) fixing “hung”,
If you change the word at "twittered",
2) Fix "twittered" and set "hung"
Consider two cases where the word at the position is changed.

For example, in the case of 1), this English sentence (example sentence 2)
If there is a Japanese original sentence of, a translation database is referred to, and another verb candidate of the tweeted original word is searched from, for example, a table as shown in FIG. 21. Search for In this case, “chirp”, “chirrup”, “sin
"g", "trill", and "warble", and the past tense and its pronunciation are stored as shown in FIG. Of these, it can be seen that only the past tense “sang” of “sing” rhymes with “hung”, so “A new as a candidate for rewriting the English sentence (example sentence 2)”
bell hung; the birds san
g. Is generated. In the case of 2), the same operation is performed, and when a plurality of candidates are finally formed, the user is caused to select the most preferable one.

On the other hand, if the English sentence (example sentence 2) is not a Japanese English translation, a similar word search is performed from an English thesaurus without using the Japanese-English dictionary, and the word is substituted into a desired position. I just need.

Although various processes (functions) have been described above, the respective functions can be implemented by appropriately combining them (that is, it is optional to provide each function or not to provide each function).

In the present embodiment, the post-editing process is performed after translating a document in the first language into a document in the second language. Necessary information may be input, and editing processing may be performed on the second language document input from outside. In this case, an embodiment without a translation function is also possible.

Each of the above functions can be realized as software. When implemented as software, the program may be stored in a ROM, but may be implemented by installing the program on a general-purpose computer as follows. That is, first, the program is stored in a computer-readable recording medium (for example, a floppy disk or a CR-
The program is read using a disk drive device corresponding to the recording medium, stored in the RAM, and executed. Alternatively, the program is once installed in a hard disk device or the like, and is stored in the RAM from the hard disk device or the like at the time of execution and executed. When the recording medium storing the program is an IC card, the program can be read using an IC card reader. Alternatively, a program can be received from a predetermined interface device via a network. The present invention is not limited to the above-described embodiment, and can be implemented with various modifications within the technical scope.

[0070]

According to the present invention, a desired sentence or the like can be rewritten to a desired style or tone without bothering with detailed operations.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration example of a document editing device according to an embodiment of the present invention.

FIG. 2 is a flowchart illustrating an example of a translation processing procedure;

FIG. 3 is a diagram showing a specific example of a document in a first language;

FIG. 4 is a flowchart showing an example of a post-editing processing procedure;

FIG. 5 is a diagram showing a specific example of a second language document obtained by translating a first language document;

FIG. 6 shows an example of a display screen.

FIG. 7 is a diagram showing an example of a menu screen of an edit command.

FIG. 8 is a diagram showing an example of a tree structure of a copy source sentence;

FIG. 9 illustrates an example of a tree structure of a copy destination sentence.

FIG. 10 is a diagram showing an example of a syntax dictionary.

FIG. 11 is a diagram illustrating an example of a syntax conversion rule.

FIG. 12 is a diagram illustrating an example of a tree structure of a sentence after conversion.

FIG. 13 is a flowchart showing another example of the post-editing processing procedure.

FIG. 14 is a flowchart showing still another example of the post-editing processing procedure.

FIG. 15 is a diagram showing an example of a syntax information candidate menu screen.

FIG. 16 is a diagram showing an example of a conversion candidate menu screen.

FIG. 17 is a diagram showing an example of a menu screen for syntax information candidates and conversion candidates.

FIG. 18 is a diagram illustrating another example of a syntax information candidate menu screen.

FIG. 19 is a diagram showing an example of a table storing words and their phonetic symbols;

FIG. 20 is a diagram illustrating a conversion process based on pronunciation information.

FIG. 21 is a diagram showing an example of a table storing words and their phonetic symbols;

[Explanation of symbols]

 101 input unit 102 translation management unit 103 translation engine 104 translation database 105 output unit

Claims (6)

[Claims] A first language document is converted to a second language document by natural language processing.
Means for translating into a document in a language, and a predetermined attribute in natural language processing for one sentence or a part thereof constituting the document in the second language, another sentence or a part thereof constituting the document in the second language And a copying means for copying the document. 2. The apparatus according to claim 1, further comprising means for restoring all or a part of the predetermined attribute to the other sentence or a part thereof whose predetermined attribute has been changed by the copying means. Item 2. The document editing device according to Item 1. 3. The method according to claim 1, wherein the copying unit performs a change process for changing an attribute of the other sentence or a part thereof.
3. The apparatus according to claim 1, further comprising means for designating which attribute in the one sentence or a part thereof is focused on, wherein the copying means performs the change processing based on the designation. Document editing device as described. 4. A means for designating a first character string to be a copy source of a predetermined feature; a means for designating a second character string to be a copy destination of a predetermined feature; And copying means for converting the second character string so that the second character string has predetermined characteristics of the designated first character string. Document editing device. 5. A document in a first language is converted into a second document by natural language processing.
Translated into a document in a language, and a predetermined attribute of one sentence or a part thereof constituting the second language document in natural language processing is copied to another sentence or a part thereof constituting the second language document A document editing method. 6. A computer which causes a computer to translate a document in a first language into a document in a second language by natural language processing, and performs a predetermined processing on one sentence or a part thereof constituting the document in the second language in natural language processing. A computer-readable recording medium storing a program for copying the attribute of the second language into another sentence constituting the second language document or a part thereof.