JPH06309358A - Natural language processor - Google Patents

Abstract

PURPOSE:To improve the efficiency of correcting operation when the correcting operation is performed by using temporarily printed matter of a process result. CONSTITUTION:An original sentence in natural language is inputted from an input device 1 and then analyzed by an analyzing device 1. A printing processor 4 encodes the information found by the analyzing device 1 and prints it out in bar code on a printer. An editing operator who makes corrections perform postediting operation by using the printed form and then when there is an unknown word in the original sentence or its translation, the bar code on the form corresponding to the word is scanned by a handy scanner to easily and instantaneously refer to information on a machine. Further, the result of the correcting operation can easily be outputted.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a natural language processing device, and more particularly, a machine translation device having a printing function for printing various information obtained during processing along with an original sentence or a translated sentence by a barcode. Related to a natural language processor.

[0002]

2. Description of the Related Art In recent years, many machine translation devices using electronic computers have been developed. Usually, when translating using such a machine translation device, first ambiguous or difficult parts in the original sentence are changed to expressions that are easy to machine (pre-editing). Next, the pre-edited sentence is translated by the machine translation device. Furthermore, the steps of correcting errors in the translated text output by the machine translation device and preparing a more natural and easy-to-understand sentence (post-editing) are taken. Since the pre-editing and the post-editing are manually performed, these two tasks are often a bottleneck in improving the efficiency of translation using a machine translation device. For this reason, the machine translation device usually has a function of supporting pre-editing work and post-editing work.

Although it is possible to carry out all the work online by utilizing such a support function, it is not uncommon for the work to be done by making corrections on the printed matter printed on the paper. The reason for this is that once the work is printed on paper, it is possible for multiple workers to divide the work without monopolizing an expensive machine translation device. There is no problem even for a person, and there is a high degree of freedom in work such as taking notes and drawing drawings on paper rather than on a display device.

As a known document describing the prior art according to the present invention, for example, there is a "language reading device" in Japanese Patent Laid-Open No. 62-294278. In order to enable the learner to instantly and easily listen to his or her desired sentence reading voice, the bar code printed in this publication is in a one-to-one correspondence with the first word of each printed sentence. And address information for accessing the first word of each sentence of the read voice information in the compact disc.

[0005]

As described above, in the conventional natural language processing apparatus, when pre-editing or post-editing the original sentence on the printed matter is performed, the sentence (original sentence, translated sentence) on the printed matter is If you want to know the information (meaning in the dictionary, dependency, etc.), you have to operate the translation device to obtain them. For example, when the original sentence includes an unknown word, the worker cannot determine whether the translated sentence is an appropriate sentence without knowing the meaning of the word. Even if an electronic dictionary is available, it takes time and effort to search for the word through a keyboard or the like. Although techniques such as character recognition can be used for inputting words, there are problems that the device is complicated, the cost is high, and the word recognition cannot be sufficiently accurate. Further, after the correction on the printed matter is completed, there is a problem that it takes time and effort to re-input the work result into the translation device.

The present invention has been made in view of the above circumstances, and when a language processing device such as a machine translation device is used for correction work using a printed material obtained as a result of processing, the working efficiency is improved. It is an object of the present invention to provide a natural language processing device as described above.

[0007]

In order to achieve the above object, the present invention comprises (1) an input means for inputting an original sentence of a natural language, a storage means for storing a natural language dictionary, and the input means. An analyzing unit for analyzing the input original sentence and a printing unit for encoding the information obtained by the analyzing unit and printing it by a bar code are provided, and further, (2) an identifier of a word included in the original sentence is provided. Printing with a bar code; further, (3) printing with the bar code attached to the original text; and (4) coding with position information in the original text and printing with a bar code, Is (5)
Converting means for converting the information obtained by the analyzing means into a translated sentence of a second language; and encoding the information obtained by the conversion processing of the converting means and printing it by a bar code; and (6) In (5) above, a printing means for printing the identifier of the word included in the translated text along with the translated text by a bar code is provided, or (7) an input means for inputting an original sentence in natural language, and a natural language In a natural language processing apparatus including a storage unit for storing a dictionary and an analysis unit for analyzing an input original sentence, when the analysis unit detects an ambiguity, a candidate which constitutes the ambiguity is encoded to form a barcode. And (8) a barcode reading means for reading the ambiguity candidate printed by the barcode in (7), and (8) selecting by the barcode reading means. And a correction means for correcting the ambiguity removed sentence by using the read ambiguity candidate, and further, (9) in (7), the ambiguity printed by the barcode is provided. Bar code reading means for reading the candidates, and selected by the bar code reading means,
A translation means for performing translation using the read ambiguity candidate is provided, or (10) an input means for inputting a natural language original sentence, a storage means for storing a natural language dictionary, and the input means. In a natural language processing device comprising an analyzing means for analyzing an original sentence input by the above and a converting means for converting the information obtained by the analyzing means into a translated sentence of a second language, a plurality of translated words of the original sentence word exist. If you do
A print means for printing the translated word and a bar code of a word identification code corresponding to the translated word, and (11) in (10), the bar code reading means is provided. It is characterized in that translation is performed using the translated word selected and read by the code reading means.

[0008]

[Function] Since the printing function of the processing device is provided with a function of printing various information in the form of a bar code in addition to the character strings of the original sentence and the translated sentence at the time of printing, the editorial worker who makes corrections can use the above-mentioned language. If post-editing work is performed using the paper printed by the processor, if there is a word that you do not know in the original sentence or translated sentence, operate the barcode on the paper corresponding to that word with the handy barcode scanner. This makes it possible to easily and instantly refer to the information on the machine. Further, the result of the correction work can be easily input.

[0009]

Embodiments will be described below with reference to the drawings. FIG. 1 is a block diagram for explaining an embodiment of a natural language processing device according to the present invention, in which 1 is an analysis device and 2 is
Is an input device, 3 is a printer, 4 is a print processing device, 5 is a control device, 6 is a grammar rule, 7 is a dictionary, and 8 is a conversion device. The input device 2 is a keyboard, a mouse or the like for inputting an original sentence or an instruction to the device. The analysis device 1 is for analyzing the original text input by the input device 2. The grammar rules 6 and the dictionary 7 are stored in the storage device. The analysis device 1 performs analysis according to the dictionary 7 and the grammar rule 6. The conversion device 8 uses the information obtained by the analysis device 1 to perform conversion into the second language according to the dictionary 7 and the grammar rule 6. The print processing device 4 is the analysis device 1
The information sent from the converter 8 is printed on the printer 3 in a format according to the instruction from the controller 5. If no translation is performed, the conversion device 8 can be omitted.

FIG. 2 is a block diagram of the print processing apparatus in FIG. 1, in which 9 is a bar code generating section, 10 is a printer command combining section, 11 is a print processing control section, and 12 is a print format adjusting section. . The barcode generator 9 converts the data into a barcode. The printer command synthesis unit 10 synthesizes a control command for controlling the printer. The print processing controller 11 controls the bird view generator 9 and the printer command synthesizer 10 according to an instruction from the controller 5.

FIG. 3 is an operation flowchart showing the operation from the input of the original text to the printing. St in the figure
ep2 is an analysis process, and step3 is a conversion process, which are performed by the analysis device 1 and the conversion device 8, respectively. When translation is not performed, step 3 is omitted and the result of the analysis process is directly subject to the printing process.

FIG. 4 is a flow chart showing the operation of the printing process. Hereinafter, each step will be described in order. The printing process is executed under the control of the control device 5, but if the control device 5 instructs to print with a barcode in step 5, the barcode generation part 9 analyzes the barcode in step 6. The information sent from the device 1 or the conversion device 8 is encoded to create a bar code image. In step 9, the original sentence and the translated sentence and this bar code are printed. In step 7, the print format adjustment unit determines the print layout. For example,
If the text is printed in the vertical direction, it is better to print the bar code vertically as well, and the adjustment of the printing color of the bar code portion is also performed here. When printing a sentence and a bar code at the same time, the width between the two is also adjusted here. st
In ep8, the printer command synthesis unit 10 causes the printer 3 to conform to the format determined by the print format adjustment unit 12.
Synthesizes a sequence of instructions to pass to. The printer 3 prints according to this command sequence.

The printed matter as shown in FIGS. 5, 6 and 7 can be obtained by the above-described structure and operation. In FIG. 5, 1
3 is the print of the original English sentence "He saw the outcrop.", 14 is the print of its Japanese translation "He saw the outcrop.", 15 is the word of each word contained in the sentence. This is the barcode of the identifier. FIG. 6 shows that the analysis device 1 detects that it is ambiguous whether "Yesterday" is "buy" or "read" in the original "Taro read the book bought yesterday". According to the sent information, the bar code corresponding to each dependency is printed in a format like this kind of selection branch.

FIG. 7 shows that the original words "Taro rented a room in front of the station" have translated words "borrowed", "borrow", "rent", "us".
Since there are three "e", this translated word and the word identifier of the translated word are printed as a set. The correction operator who obtained the printed material as shown in FIG. 5 also determines the meaning of the word "outcrop" in the translated text. Suppose that he did not know the meaning of the word "outcrop" in the original sentence corresponding to the same word. In such a case, the post-editing worker cannot determine whether this sentence is a proper translation. Since it is possible to input the word identifier of "outcrop" by using the bar code reading device 16 as shown, the electronic dictionary is searched by using this word identifier as a key, and the dictionary description of "outcrop" is displayed on the display device 17. 8 can be displayed, which is shown in FIG.
Further, in FIG. 7, the translation device selects borrow as the translated word of “borrowed”, but this translated word is “rent”.
If the worker determines that the
Put a mark in the rent column with a writing instrument, and when all the sentences have been corrected, scan the barcode along the "rent" to input the translation selection into the machine translation device. You can The translation device retranslates using the selected translation word. FIG. 9 is a diagram showing the situation.

[0015]

As is apparent from the above description, the present invention has the following effects. That is, if there is a word that the user does not know in the original sentence or the translated sentence, the information on the machine can be referred to easily and instantly by operating the barcode on the paper corresponding to the word with the handy barcode scanner. Also, the result of the correction work can be easily input.

[Brief description of drawings]

FIG. 1 is a configuration diagram for explaining an embodiment of a natural language processing device according to the present invention.

FIG. 2 is a configuration diagram of a print processing apparatus in FIG.

FIG. 3 is a flowchart showing an operation from inputting an original sentence to printing in the present invention.

FIG. 4 is a flowchart showing an operation of a printing process according to the present invention.

FIG. 5 is a diagram showing a print example in which a word identifier is added to an original sentence and a translated sentence in a printed matter obtained by the present invention.

FIG. 6 is a diagram showing a print example in which the ambiguity of analysis is printed by a barcode on a printed matter obtained by the present invention.

FIG. 7 is a diagram showing a print example in which a translation word candidate is printed by a bar code on a printed matter obtained by the present invention.

FIG. 8 is a diagram showing how to search an electronic dictionary using a barcode according to the present invention.

FIG. 9 is a diagram showing how a translated word is selected and re-translated using a barcode according to the present invention.

[Explanation of symbols]

1 ... Analysis device, 2 ... Input device, 3 ... Printer, 4 ... Print processing device, 5 ... Control device, 6 ... Grammar rule, 7 ... Dictionary, 8
... conversion device, 9 ... bar code generation unit, 10 ... printer command synthesis unit, 11 ... print processing control unit, 12 ... print format adjustment unit.

Claims (11)

[Claims] 1. Input means for inputting an original sentence in a natural language,
A storage unit for storing a natural language dictionary; an analysis unit for analyzing the original sentence input by the input unit; and a printing unit for encoding the information obtained by the analysis unit and printing it by a bar code. Characteristic natural language processor. 2. The natural language processing apparatus according to claim 1, wherein the identifier of the word included in the original sentence is printed by a bar code. 3. The natural language processing apparatus according to claim 1, wherein the barcode is printed together with the original text. 4. The natural language processing apparatus according to claim 1, wherein the position information in the original sentence is encoded and printed as a bar code. 5. A conversion means for converting the information obtained by the analysis means into a translated sentence in a second language, and the information obtained by the conversion processing of the conversion means is encoded and printed as a bar code. The natural language processing device according to claim 1. 6. The natural language processing apparatus according to claim 5, further comprising: a printing unit that prints the identifier of the word included in the translated text with the barcode attached to the translated text. 7. Input means for inputting an original sentence in natural language,
In a natural language processing apparatus including a storage unit that stores a natural language dictionary and an analysis unit that analyzes an input original sentence, if the analysis unit detects an ambiguity, a candidate that constitutes the ambiguity is encoded. A natural language processing device comprising a printing means for printing with a bar code. 8. A bar code reading means for reading the ambiguity candidate printed by the bar code, and a ambiguity removed sentence using the ambiguity candidate selected and read by the bar code reading means. 8. The natural language processing apparatus according to claim 7, further comprising a correction unit for correcting. 9. A bar code reading unit for reading a ambiguity candidate printed by the bar code, and a translation unit for translating using the ambiguity candidate selected by the bar code reading unit and read. The natural language processing device according to claim 7, wherein 10. An input unit for inputting a natural language original sentence, a storage unit for storing a natural language dictionary, an analyzing unit for analyzing the original sentence input by the input unit, and information obtained by the analyzing unit. In a natural language processing device having a conversion means for converting a translated sentence of a second language, when there are a plurality of translated words of the word of the original sentence, those translated words, and a bar code of a word identification code corresponding to the translated word A natural language processing device comprising a printing means for printing as a set. 11. The natural language processing apparatus according to claim 10, further comprising the bar code reading means, wherein the translation is performed using the translated word selected and read by the bar code reading means.