JPH0290261A - Kana/kanji conversion system - Google Patents

Abstract

PURPOSE:To reduce misconversion correcting work by applying the same KANJI (Chinese character) conversion processing to a KANJI character having the same KANA (Japanese syllabary) reading in a specific conversion range when an operator executes the KANJI conversion processing based upon correcting operation from a 'KANJI' in a 'KANA'-KANJI mixed sentence within a specific conversion range to another KANJI. CONSTITUTION:Each KANA reading inputted from a keyboard on an I/O device 1 is successively stored in an input buffer 3 by an I/O control device 2. When a KANA-KANJI conversion executing instruction is inputted from the keyboard on the I/O device 1 following the input of the KANA reading, the KANA-KANJI conversion executing instruction is transferred from an I/O control device 2 to a system control device 5 and the device 5 commands a KANA-KANJI conversion control device 6 to execute KANA-KANJI conversion. When the operator executes KANJI conversion processing based upon correcting operation from a KANJI in a KANA-KANJI mixed sentence within the specific conversion range to another KANJI, the same KANJI conversion processing is applied to another KANJI having the same KANA reading in the specific conversion range. Consequently, operator's misconversion correcting work can be reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a kana-kanji conversion system, and particularly to a kana-kanji conversion system that effectively corrects erroneous conversions in kana-kanji conversion.

[Conventional technology]

Conventional kana-kanji conversion methods that convert input kana readings into kanji include a sequential kana-kanji conversion method that inputs the kana readings for kanji and characters and converts them one character at a time. There are two methods: a batch type kana-kanji conversion method in which a series of kana readings of Japanese sentences are input and the input kana readings are converted all at once. In the former method of kana-kanji conversion, a desired kanji is selected for each - character, so basically no erroneous conversion occurs. However, in the latter case, the first kanji displayed is a conversion candidate, and not all kanji are necessarily converted correctly.Therefore, in this -discarded kana-kanji conversion method, An operation to correct the erroneous conversion must be performed later. This operation becomes extremely inefficient as the amount of kana-kanji conversion increases, and the work required to correct the erroneous conversions increases.

On the other hand, as a method to reduce the work of correcting erroneous conversions in kana-kanji conversion, for example, as described in Japanese Patent Laid-Open No. 60-39226, the contents of corrections made by the operator are memorized.
There is a proposal to perform kana-kanji conversion by referring to the stored correction details when determining the next conversion candidate, thereby reducing the frequency of erroneous conversions.

According to this method, for a specific kana reading, the desired kanji is shown first as a conversion candidate from the second time onwards.
It is possible to reduce the work of correcting erroneous conversions by the operator.

In addition, in recent years, phrase conversion methods have been developed that increase the conversion efficiency of kana-kanji conversion by being aware of phrases.
In this phrase conversion method, in addition to the problem of erroneous conversion as described above, another erroneous conversion occurs due to incorrect delimitation of phrases. Correcting this bunsetsu error is complicated due to the correction of the above-mentioned erroneous conversion.

To deal with this, for example, as described in Japanese Patent Application Laid-open No. 11964/1983, the next conversion candidates are displayed for each phrase and selected, so that the operator does not have to be conscious of how to separate the phrases. There is a proposal to obtain the desired conversion result with a single operation. Since this method automatically recognizes phrases, the operator's correction work is greatly reduced.

[Problem to be solved by the invention]

However, the aim of these conventional techniques is to reduce the frequency of erroneous conversions occurring during the first conversion execution in the former case, and to reduce the correction work for a single erroneous conversion in the latter case. That is. Therefore, in the conventional technology, no consideration is given to reducing the work of correcting the same erroneous conversion within a specific conversion range, and even if multiple identical erroneous conversions exist within a specific conversion range, The problem was that it had to be corrected one by one.

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

An object of the present invention is to enable a kana-kanji conversion method to obtain desired results with fewer corrections for multiple identical erroneous conversions within a specific conversion range.

The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

[Means to solve the problem]

In order to achieve the above object, 1 the present invention provides a kana-kanji conversion method that inputs kana-read Japanese and converts it into a kana-kanji mixed sentence in a kana-kanji conversion dictionary. When an operator performs a kanji conversion process by correcting a kanji into a different character, the same kanji conversion process must be performed on other kanji with the same kana reading within a specific conversion range. It is characterized by

[Effect]

According to the means, in the kana-kanji conversion method, when the operator performs kanji conversion processing by correcting a kanji into a different character for a kanji in a sentence containing kana-kanji within a specific conversion range, the specific conversion is performed. The same kanji conversion process is performed for other kanji with the same kana reading within the range. In other words, the correction content for the operator's erroneous conversion performed within a specific conversion range is memorized, and other kanji conversion processes are performed based on the correction content. After searching for the presence of the same erroneous conversion, if there is the same erroneous conversion, the same correction process is performed based on the stored correction content, and this is repeated within a specific conversion range.

In this way, within a specific conversion range, the correction work for an erroneous conversion performed by the operator is also performed for other identical erroneous conversions. Therefore, the work of correcting erroneous conversions performed by the operator can be reduced.

〔Example〕

Hereinafter, one embodiment of the present invention will be specifically described using the drawings.

FIG. 1 is a system configuration diagram of a Japanese document processing device according to an embodiment of the present invention. In FIG. 1, a single line indicates the affiliation of storage areas, and two lines indicate the control relationship between devices. Further, a dashed-dotted arrow indicates an instruction using a pointer.

Kana reading is input from the keyboard of the input/output device 1, and the input kana reading is sequentially stored in the input buffer 3 by the input/output control device 2. When a kana-kanji conversion execution instruction is input from the keyboard of the input/output device 1 following the kana reading input, the kana-kanji conversion execution instruction is passed from the input/output control device 2 to the system control device 5. The system control device 5 instructs the kana-kanji conversion control device 6 to execute the kana-kanji conversion, and at the same time transfers the kana reading in the input buffer 3 to the kana reading buffer 10. The kana-kanji conversion control device 6 creates a single-word pointer 11 and a phrase pointer 12 using the single-kanji conversion device 7 and the bunsetsu delimiter selection device 8, and creates conversion candidate character strings in the kanji buffer 9. When all the kana readings have been converted, the kana-kanji conversion control device 6 notifies the system control device 5 of the completion of the conversion. The system control bag W5 transfers the converted character string in the kanji buffer 9 from the kana-kanji conversion control device 6 to the output buffer 4, and issues a display command to the input/output control device 2.

The input/output control device 2 displays the contents of the output buffer 4 on the display device of the input/output device 1, and waits for the next input from the operator.

The operator looks at the displayed content and, if there is any erroneous conversion, performs correction work for the erroneous conversion. Now, in the example shown in FIG. 1, if an operation is performed to correct the character "atsu" pointed to by Pl to "heat", in this case, the kana-kanji conversion control device 6 converts the uncorrected kanji into the uncorrected kanji. The corrected kanji are transferred to the kanji buffer 15, the corrected kanji are transferred to the corrected kanji buffer 16, and the kana reading is transferred to the corrected kana reading buffer 14, and the automatic correction control device 13 is activated.6 Figure 2 shows the flow of the automatic correction process. FIG. Referring to FIG. 2, the automatic correction process performed by the operator's correction operation will be described.

This process is started when the operator performs a correction operation. First, in step 31, the content of the pointer P2 indicating the operator's correction part is stored in the character string search pointer (P) 17. Next, in step 32, the kana reading indicated by P and the corrected kana reading in the corrected kana reading buffer 14 are stored. Compare the kana reading and if they are different, proceed to step 35.
A process is performed to advance P to the next one, and if the kana readings are equal, then in step 33, the kanji indicated by P is compared with the kanji in the pre-correction kanji buffer 15. In this comparison, if the kanji are different.

Proceeding to step 35, processing for advancing P to the next step is performed.

Furthermore, if the kanji are the same, it can be considered that the content is the same as the content that the operator has determined to be erroneously converted, so in step 34,
The kanji character indicated by P is replaced with the kanji character in the post-correction kanji buffer 16, and in the next step 35, processing is performed to advance P to the next step. In the process of advancing P to the next step in step 35, the content of P3 next to P2 in the word pointer 11 is extracted and newly stored in the character string search pointer 17. Read kana and determine whether there are kanji. In this case, since P indicates the contents of the kana reading buffer 10, the process returns to step 32 for determination. This process is repeated until there is no kana reading pointed to by P in the kana reading buffer 10, and the automatic correction process is completed.

After correcting all "thickness" to "heat" in this way, the automatic correction control device 13 notifies the kana-kanji conversion control device 6 of the end of the process, and then notifies the system control device 5 of the end of the correction work. be done. The system control device 5 transfers the corrected character string from the kanji buffer 9 to the output buffer 4 and issues a display command to the input/output control device 2. The input/output control device 2 transfers the contents of the output buffer to the input/output device 1. display on the display device and wait for the next input from the operator. Here, if the operator instructs another correction, the same process as described above is performed. Also, if you perform the conversion completion operation here,
It is assumed that the conversion currently in the kana reading buffer 1o has been completed, and the next kana reading input and other processing continue.

In this way, according to this embodiment, within a specific conversion range,
Since it is not necessary to repeat the same erroneous conversion correction work, it is possible to improve the efficiency of Japanese input by the operator.

In addition, in the process of searching for the same erroneous conversion location, as the initial value of the two-character string search pointer 17, a pointer indicating the next position after the position corrected by the operator (in this case, P2
), but by using the value of the first pointer of the word pointer 11, it is possible to automatically correct the same erroneous conversion that precedes the position corrected by the operator, and further efficiency can be achieved.

The error conversion correction process of this embodiment has been described above.

It can also be applied to the case of sequential automatic conversion method. This sequential automatic conversion method is a method in which kana reading is converted into kanji each time a character is manually read. Even in this case, buffers and pointers as shown in reference numbers 9 to 12 in Figure 1 can be created, so if the operator corrects the erroneous conversion at an appropriate point, the same process as described above can be performed. process,
Same erroneous conversions can be automatically corrected.

Further, in the above-described embodiment, the operator's operation for correcting misconversion was limited to homophones, but the present invention can also be applied to corrections for changing the position of phrase breaks.

FIG. 3 is a system configuration diagram illustrating the buffers and pointers of the main parts when correcting the change in the position of a bunsetsu break. That is, the buffers and pointers controlled by the automatic correction control device 13 in FIG. 3 are configured as shown in 18-22.

Here, 18 is a kanji buffer before correction, 19 is a kana reading buffer before correction, 20 is a correction front-adjustment pointer, and 21 is a pre-correction clause pointer. Further, 22 is a post-correction phrase pointer, which holds the corrected content by directly pointing to the word pointer 11. In this way, if the information before correction is stored including the phrases, 1. For example, the next pointer of a group of phrases that have been corrected by the operator in the phrase pointer 12 can be stored in the phrase search pointer (P) 23. , P, searches for a portion that matches the content stored in each of the buffers 18 to 21 in FIG.

In this way, it is possible to automatically correct the change in phrase delimiter, and the burden on the operator to create the sentence can be reduced.

Even in this case, by setting the initial value of the clause search pointer 23 to the first pointer in the pre-correction clause pointer 21, it is possible to automatically correct the same word conversion before the part corrected by the repeater.

In all of the embodiments described above, more accurate automatic correction can be performed by further taking into account information before and after the correction location. For example, if the word corrected by the operator is a particle, the part of speech next to the word found in the search for the same misconversion is checked, and if it is a particle, automatic correction is performed, and if it is different, 1 By adopting such a method of proceeding to the next search, malfunctions of automatic correction can be prevented.

Conversely, instead of searching using a pointer, there are methods of directly searching only kana readings and methods of searching only kanji. This makes it possible to simplify the process.

In addition, the pointers and buffers 9 to 12 shown in Figure 1 are
It is obvious that it does not matter whether it is in the main memory or in the magnetic memory. Therefore, the scope for automatically correcting the same erroneous conversion can be within a single conversion unit or across multiple conversion units, and it is also possible to perform automatic correction for the entire page or document. . This is because if the amount of data held is large, it can be stored in the magnetic storage device. Specifically, for example, the conversion range may be set to be the range of one screen of the display device of the input/output device.

The present invention has been specifically explained above based on examples, but
It goes without saying that the present invention is not limited to the embodiments described above, and can be modified in various ways without departing from the spirit thereof.

〔Effect of the invention〕

As described above, according to the present invention, even if an erroneous conversion occurs due to kana-kanji conversion, it can be easily corrected, and the efficiency of Japanese input can be improved.

[Brief explanation of drawings]

FIG. 1 is a system configuration diagram of a Japanese document processing device according to an embodiment of the present invention. FIG. 2 is a flowchart showing the flow of automatic correction processing. FIG. 3 is a correction for changing the position of bunsetsu breaks. FIG. 2 is a system configuration diagram illustrating buffers and pointers of main parts when performing the above. In the figure, 9... Kanji buffer, 10... Kana reading buffer, 11... Word pointer, 12... Clause pointer, 13... Automatic correction control device, 14... Correction kana reading buffer, 15...Kanji buffer before correction, 16...
- Kanji buffer after correction.

Claims (1)

[Claims] 1. In the kana-kanji conversion method, inputting kana-reading Japanese and converting it into sentences containing kana-kanji using a kana-kanji conversion dictionary,
If an operator performs a kanji conversion process by correcting a kanji into a different character for a kanji in a sentence containing kana-kanji within a specific conversion range, the conversion process for other kanji with the same kana reading within the specific conversion range This is a kana-kanji conversion method that performs the same kanji conversion process for both characters.