JPS63236163A - Character processor - Google Patents

Abstract

PURPOSE:To improve the converting efficiency of characters without adding a field-based dictionary, a technical term dictionary, etc., by providing plural memory means for each field to store the learning information for the priority of conversion candidates of a dictionary and a means which uses those learning information of each memory means with switching. CONSTITUTION:A KANA (Japanese syllabary)/KANJI (Chinese character) conversion independent word dictionary 23 stores the KANJI (description) against the Japanese rendering of the independent words for KANA/KANJI conversion respectively in the desired number of words. The data on the Japanese rendering to be converted are received and the KANJI corresponding to the rendering is retrieved and sent to a candidate editing control part 20. A learning information control part 24 switches the learning information stored in the part 24 to a learning information storing area 25 of the corresponding field by a learning information switching instruction given from an input control part 13. The area 25 includes the memories storing the learning information on each field. These learning information are stored individually or all together into an external memory 4 and also read out of the memory 4.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a character processing device having at least a character conversion function in word processors, office computers, personal computers, data processors, workstations, and the like.

In general, in the various character processing devices mentioned above,
Many devices have a character conversion function that converts characters or character strings input from an input device into other characters or character strings using a dictionary and displays the converted characters.

For example, a word processor that has a kana-kanji conversion function has a table called an independent word dictionary that is used when converting kana to kanji, and uses this table as input reading (kana information).
The method used is to search for the kanji and convert it into kanji.

At that time, if there are multiple kanji information for the same reading,
It is common practice to have the user select a kanji as a conversion candidate and then determine the kanji in -i manner.

In the case of Japanese, many words have multiple kanji for the same reading, so when processing these, grammatical information and other category (field) information are used to select and display one of the candidates. Measures are being taken.

However, this is not an absolute method, so a learning process is performed as a complementary method, and if the displayed candidate is changed and selected, the selected candidate is displayed preferentially in the next conversion. has been taken.

In addition, Japanese documents have various forms depending on the profession and field of the creator, and commonly available independent word dictionaries are not sufficient.

In order to compensate for this insufficiency, field-specific dictionaries (dictionaries for each specialized field), specialized language dictionaries, etc. are added using external storage devices or external expansion memory, and searches for these dictionaries are added to the dictionary search order. Measures have also been taken to substantially improve conversion efficiency by adding

In order to simplify the internal processing logic, these field-specific dictionaries and specialized word dictionaries are added in a form that also includes data in the independent word dictionary, making it possible to completely search in field-specific dictionaries, etc. (extended dictionaries). Switch and do not access the internal dictionary,
There is an extended dictionary that stores only words that are not in the independent word dictionary and accesses both the independent word dictionary and the extended dictionary.

However, since the former also stores independent word dictionary information, its capacity is large, and when it is supplied by an external storage device, it requires access time each time conversion is performed.

In the latter case, the logic for accessing two or more dictionaries is required, and the complexity of the switching and learning process (such as learning words in a field-specific dictionary and processing when the field-specific dictionary is removed) is required. However, there was a drawback that the operation of the processing part was complicated.

In addition, the dictionary data usually stores readings and candidate parts of speech, but these are compressed and are expanded by an expansion algorithm at the time of conversion and then displayed as candidates. If the data is divided, it is expected that the compression efficiency will be lower than if it were combined into one.

Moreover, in the latter case 1 the data in the independent word dictionary is usually in standard form, ie in its original form, and it is not possible to change the priority depending on the field.

In the former case, even the information included in the independent words is rearranged for each field.

There are various methods for learning processing, such as one that performs it frequently and one that performs it in priority order.
This will be explained using a general and easy-to-understand example of learning to use the latest technology.

Latest usage learning means selected candidates among homophones,
It displays the first lump candidate from the time of the next conversion, and is characterized by its simple logic and quick effect.

The following three methods are generally used to perform this learning process.

The first method is to add learning information to dictionary information and update the dictionary information each time a selection is made.

The second method is to update learning information provided in 1:1 correspondence with dictionary information each time a selection is made.

The third method is to sequentially store selected candidates in the learning information area.

The first method requires updating dictionary information, so it is necessary to prepare a dictionary in a writable memory. If a large capacity dictionary can be stored, processing is simple, but if the dictionary cannot be stored, overlay etc. complicated methods must be used.

The second method is that the dictionary itself is read-on memory (RO
M), but the learning information corresponding to the 1:1 ratio is also relatively large.

The third method is that the dictionary itself is read-on memory (RO
As for the learning information, the memory size can be determined relatively freely by limiting the amount of information to be stored, so it can be realized even with a small memory size.

Usually, the third method is used in portable word processors, and the first or second method is often used in large-scale word processors.

Among these conventional learning methods, the second. In the third method, learning information and dictionary information are separated, so it is technically possible to have multiple pieces of learning information for one dictionary information.
Each individual (user) learning information could be stored in an external storage device. However, this method loads learning information from external storage into memory and uses it when the device is started up, and it was not possible to switch during document creation.

This invention was made in view of the above points, and it is possible to improve conversion efficiency in the same way as adding field-specific dictionaries without adding field-specific dictionaries or specialized language dictionaries. The purpose is to make it measurable.

In order to achieve the above object, the present invention includes an input device;
In a character processing device that includes a dictionary for converting characters or character strings input from the input device into other characters or character strings, and a display device for displaying the converted characters or character strings, A plurality of storage means for learning information in which the priorities of dictionary conversion candidates are learned are provided for each field, and means for switching and using the learning information in each storage means is provided.

Hereinafter, a detailed explanation will be given based on one embodiment of the present invention.

FIG. 2 is an external view of a character processing device, such as a personal word processor, embodying the present invention.

This personal word processor includes an input device 1 using a keyboard for inputting character information, format information, and control information necessary for creating and editing a document, a display device 2 using a liquid crystal display device (LCD) for displaying character information, etc., and a document created by the personal word processor. a printer 3 for printing etc., and an external storage device 4 using a floppy disk device (FDD) for storing document information.
It is equipped with

FIG. 3 is a block diagram showing an example of the system control section of this personal word processor.

This system control unit 6 is a C
PU10. ROM11. An input control section (keyboard interface) 13 that controls the transfer of key information between a microcomputer system consisting of a RAM 12 and an input device (keyboard) 1, and a display control section (display controller) that controls a display device (LCD) 2. ) 14, a printer interface 15 that controls information transfer to the printer 3, and an external storage device (FDC) IE 3 that controls an external storage device (FDD) 4.

FIG. 1 is a functional block diagram showing the functions of this system control unit according to the present invention, and parts corresponding to those in FIG. 3 are given the same reference numerals.

Then, the control code determination section 172 characters [collection control section 18.
Reading editing control section 19. Candidate editing control unit 20, format storage area 212, character storage area 22. Kana-Kanji Conversion Independent Word Dictionary 23. Learning information control unit 24. Each function of the plurality of field-specific learning information storage areas 25 is controlled by CPU10. R-
This is done by a microcomputer system consisting of OMI L and RAMI.

Next, the functions of each of these parts will be explained.

The input device 1 is, needless to be explained, a keyboard or the like (including a touch panel, mouse, etc. if used).
Many character keys such as kana keys, alphabet keys, and
In addition to the numeric keys "~9", cursor movement keys, and function keys for selecting various functions, it is equipped with an execution key, a conversion key, and a learning information switching key.

The input control section 13 selects various types of key information input from the keyboard 1 and transfers it to the control code determination section 17, the learning information control section 24, and the external storage control section 1.

The control code determination unit 17 transfers the character information input from the input control unit 13 directly to the character editing control unit 18 for editing processing, transfers it as reading data to the reading editing control unit 1, or displays the character information while displaying candidates. Whether or not to transfer it to the candidate editing control unit 20 as an instruction input is determined based on the sequence of key data so far, and is transmitted to each of them.

The character editing control unit 18 determines the position from the cursor position based on the character code input from the control code determination unit 17, format information, and character information that has already been input, and changes the position to the character information that has already been input. After inserting the newly input character code, perform m collection of screen information again,
The image is displayed on the display device 2 via the display control section 14.

On the 1st, the reading editing control unit transfers character data that is the reading of kanji such as hiragana from the control code determination unit 17.
This character data is displayed in reverse video to indicate that it is the reading data to be converted.

The candidate editing control unit 20 extracts the reading data input when the conversion key instructing conversion is pressed, and converts it into the Kana-Kanji conversion independent word dictionary 2 using the Kana-Kanji conversion algorithm.
3, perform a connection check, and if the connection is OK, select a candidate, then retrieve the learning information of the corresponding candidate from the learning information control unit 24, rearrange the priority order,
After removing items to be deleted, this data is output to the display control unit 14 in order to display candidates in order from the top.

The format storage area 21 stores format data such as lines and format information, and is used by the character editing control unit 1.
Sends format data for display on the day.

The character storage area 22 reads and stores character data and confirmed candidate character data from the character editing control section 18 and candidate editing control section 20, and also sends it to the character editing control section 18.

The Kana-Kanji conversion independent word dictionary 23 is, as shown in Figure 4,
Kanji (notation) for each independent word reading for kana-kanji conversion
When it receives the data of the pronunciation to be converted, it searches for the kanji corresponding to the pronunciation and sends it to the candidate editing control section 20.

The learning information control unit 24 switches the learning information switch in the learning information control unit to the learning information storage area 25 of the corresponding field in response to the learning information switching instruction from the input control unit 13.

The learning information storage area 25 consists of a memory area for storing learning information by field as shown in FIG. Device.

4.

Next, the operation of this embodiment will be specifically explained with reference to the drawings from FIG. 4 onwards.

FIG. 4 is an example of a kana-kanji conversion independent word dictionary.As shown in the figure, the contents of the dictionary in this embodiment are composed of reading data, notation data, and part-of-speech data. This data is stored according to a predetermined priority order for the same reading.

FIGS. 5(a) and 5(b) are examples of field-specific learning information, both of which consist of priority flag and deletion flag data (on, off) corresponding to the dictionary shown in FIG. Priority flag is on
The deletion flag is a flag that gives priority to displaying the notation ``on'' and does not display the notation ``on''.

For example, if you enter "Aiu" in Figure 6 (a) and then enter "Kanshin" from the cursor C position, this "
Since "Kanshin" is reading data, it is displayed in reverse (indicated by RV) as shown in FIG.

Next, when you press the conversion key on input device 1 to convert this into kanji, if the learning information in Figure 5 (a) is selected at this time, The priority flag corresponding to the first candidate expression “impressed” is roffJ,
The deletion flag is ronJ, the priority flag corresponding to the second candidate expression "interest" is ronJ, and the deletion flag is roffJ, so "impressed" is deleted,
The second candidate "interest" is displayed as shown in FIG. 3(c).

After that, when the conversion key is pressed again, since there is no other notation for "kanshin" that has not been deleted, the notation returns to the same reading as in (b), as shown in FIG. 6(d).

If the learning information in Figure 5(b) is selected, the priority flags corresponding to the expressions "impressed" and "interested" are turned on and off.
The ff data is the same as the learning information in (a), but the deletion flags are all set to roffJ, so Figure 6 (
When the conversion key is pressed in the display state b), "interest" is first displayed as shown in FIG. 7(c), and when the conversion key is further pressed, "impressed" is displayed as shown in FIG. 7.

In addition, although this example describes the case where the field-specific learning information corresponds one-to-one with the contents of the Kana-Kanji Conversion Independent Word Dictionary, the field-specific learning information may be stored using other methods6. FIG. 8 is an example of a display screen for switching learning information by one field. In this example, names such as general, electrical, two mechanical, etc. are displayed, and a necessary field is selected manually from among these by keystrokes.

FIG. 9 is a flowchart of learning information switching processing.

Normally, learning information in general fields is applied unless specified otherwise, but if you want to switch learning information, first press the learning information switching key on the input device @1. As a result, a learning information selection message is displayed as shown in FIG. At this time, the currently selected field "General"
is displayed in reverse video.

Next, when the keys necessary to select another field are pressed manually, it is determined whether the key input in the first step is a cursor key, an execution key, or a cancel key. Then, use the cursor keys to move the highlighted display position of the field name. If the execution key is pressed, the learning information for the field highlighted is switched to be applied, the display screen returns to the editing screen, and this process ends.

If the cancel key is pressed, the display screen returns to the editing screen and the process ends without switching the learning information.

FIG. 10 is a flowchart of the kana-kanji conversion process. The kana-kanji conversion process will be explained according to this flowchart.

When reading data is input from the input device 1 shown in FIG. 1, the control code determining section 17 transmits this reading data to the reading editing control section 19. The reading editing control unit 1 inverts this reading data. When the conversion instruction key is pressed for this inverted reading, the process shown in FIG. 10 is started.

First, the reading data is imported, then a dictionary search is performed using the longest match method, and a connection test is performed on the retrieved kanji based on its context, particles, etc.

Based on the result of the connection test, it is determined whether the connection is possible or not, and if the connection cannot be established, it is treated as an error, and if the connection is possible, the candidates are rearranged according to the priority flag of the learning information, and candidates whose deletion flag is on are deleted. . First, the first candidate is displayed and the system waits for key input.

If you have the key power, you can determine whether the key to select the displayed candidate is a key (key other than the conversion key) or not (conversion key).
If not selected (NO), the next candidate is set and displayed. If it is selected (execution key, etc.), the priority flag of the learning information is updated, this candidate character is inserted into the character storage area 22 of FIG. 1, the display screen is restored, and the process ends.

In this way, the priority flag of the field-specific learning information is learned and updated as it is used.

In order to simplify the explanation, the dictionary in Figure 4 is explained using general independent words.
By storing words, etc., and preventing them from being output by turning on the deletion flag for those words in the learning information for that field in advance for words that interfere with conversion depending on the field, efficiency can be improved depending on the field. You can perform good kana-kanji conversion. Moreover, since a common dictionary is used for all fields, the compression effect is large and the dictionary is memory efficient.

Furthermore, when the candidate character (kanji) being converted in Figure 10 is displayed, pressing the delete key will set the deletion flag for that word in the field-specific learning information currently in use.
It may be possible to set the number to n.

In this way, by implementing the present invention, it is possible to obtain the same effect as adding a field-specific dictionary or a technical language dictionary, etc., without adding them, and the conversion can be done with a simple operation. Highly efficient and low-cost kana-kanji conversion processing becomes possible. Furthermore, if users are allowed to create learning information in addition to field-specific dictionaries, it is easy to create user-specific dictionaries.

As described above, according to the present invention, it is possible to provide a character processing device that is easy to operate, has high character conversion efficiency, and is inexpensive.

[Brief explanation of the drawing]

FIG. 1 is a functional block diagram showing one embodiment of the present invention. FIG. 2 is a perspective view showing an example of the external appearance, FIG. 3 is a block diagram showing an example of the system control section, and FIG. FIG. 1 is an explanatory diagram showing a part of the Kana-Kanji conversion independent word dictionary 23. FIGS. 5(a) and 5(b) are explanatory diagrams showing examples of field-specific learning information stored in the learning information area 25 of FIG. , FIG. 6 and FIG. 7 are explanatory diagrams showing examples of display during conversion processing, FIG. 8 is an explanatory diagram showing selection display of learning information by field, and FIG.
The figure is a flow diagram of learning information switching processing. FIG. 10 is a flowchart of the kana-kanji conversion process. 1... Input device (keyboard) 2... Display device (liquid crystal display device) 3... Printer 4... External storage device (floppy disk device) 6...
...System control section 13...Input control section (keyboard interface)
14...Display control unit (display controller) 1
Day...External storage control unit (FDC) 17...Control code determination unit 18...Character editing control unit 1st...Reading editing control unit 20...Candidate editing control unit 21...Formats 1~Storage Area 22...Character storage area 23...Kana-Kanji conversion independent word dictionary 24...Learning information control section 25...Learning information storage area Figure 2 Figure 3 Figure 4 Figure 5 (Q) (b )

Claims (1)

[Claims] 1 Character processing equipped with an input device, a dictionary for converting characters or character strings input from the input device into other characters or character strings, and a display device for displaying the converted characters or character strings The apparatus is characterized in that a plurality of learning information storage means for each field are provided in which the priorities of conversion candidates of the dictionary are learned in advance, and a means is provided for switching and using the learning information in each of the storage means. Character processing device.