KR102152900B1 - Method and apparatus for processing data whth ambiguous syllabl - Google Patents

Abstract

A data processing method of a data processing apparatus according to an embodiment comprises the steps of: receiving recognition target data including ambiguous syllables which can be recognized as numbers or characters; selecting a plurality of pieces of conversion candidate data from the recognition target data, by using an electronic dictionary in which input data including one or more ambiguous syllables and output data corresponding thereto are included as conversion rules; selecting at least one of a plurality of pieces of conversion candidate data as conversion target data; and replacing the selected conversion target data among the recognition target data with the output data of the electronic dictionary and outputting the replaced data. According to the present invention, the method can maximize the readability of address information.

Description

METHOD AND APPARATUS FOR PROCESSING DATA WHTH AMBIGUOUS SYLLABL}

The present invention relates to a data processing method and apparatus for recognizing data including ambiguous syllables.

In general, at the time of writing, numbers are written in Arabic numerals in consideration of the reader's readability, and foreign words are written in characters such as Roman or Chinese characters. In particular, the reader's readability is considered more in the case of an ambiguous syllable that can be recognized as a number or a letter.

However, in the case of speech recognition, since the input method is speech and the reader's readability is not taken into account in the text conversion process performed automatically by the machine, an error occurs when the object of speech recognition contains an ambiguous syllable with the same pronunciation. Can occur. For example, when the pronunciation is “Three thousand days”, “3001 dong” may be a correct conversion result or “Three thousand one dong” may be a correct conversion result. Also, if the pronunciation is "Oisaiyo", it can be converted into "52429yo", "5242goyo", "524iyo", "Oisaiyo", etc., but only one of these may be the correct conversion result with high readability. have.

Meanwhile, conversion techniques used in speech recognition technology include dictionary-based conversion methods, conversion methods using regular expressions, conversion methods based on probability or rules after analysis by natural language processing, and sequence to sequence conversion using neural networks. There are methods and so on.

In the dictionary-based conversion method, when a word or sentence registered in a dictionary is found among input sentences, the word or sentence designated in the dictionary is replaced. The structure is simple and can be designed to enable high-speed conversion. However, the dictionary-based conversion method is usually used for 1:1 conversion into a prescribed word or sentence for a given input, and cannot be applied when a conversion result must be selected according to the context or the input type is variable.

The conversion method using a regular expression is a pattern conversion according to conditions. For example, if a number type comes after "apartment", the number is converted to Arabic numerals. It is labor-intensive because it is slow and requires each person to write rules. Accuracy can increase depending on the level of effort, but the speed is slower.

The conversion method by natural language processing is to judge the ambiguous part through language analysis, check all candidates for which the ambiguous part can be changed, and then select and replace the most influential one. Here, the selection of one promising candidate is made by rules, probabilities, and patterns, and a process of grafting with related data is required for conversion optimized for a target use. The conversion method by natural language processing is affected by speed, accuracy, and optimization of usage depending on what is the underlying natural language processing module. Since the natural language processing module cannot be developed in a short time, a previously developed module is used. In this case, it is not easy to achieve speed, accuracy, and application optimization.

The sequence-to-sequence conversion method using a neural network is a method of converting a string into another string using a neural network. The method of implementing this consists of a process of generating a specific output for a specific input using a transform model generated by learning a large-scale input-output sentence pair. Since the process of generating the conversion result is closer to creation rather than conversion, a conversion result that is separate from the input may occur. In addition, because there is little room for humans to fine-tune the learning and conversion process, it is not easy to control or correct wrong results, and the speed is also slow.

Korean Patent Registration No. 10-1964514 (registered on March 26, 2019)

According to an embodiment, there is provided a data processing method and apparatus for mechanically converting data to be recognized including an ambiguous syllable that may be recognized as a number or a character to have high readability.

The problem to be solved of the present invention is not limited to those mentioned above, and another problem to be solved that is not mentioned will be clearly understood by those of ordinary skill in the art from the following description.

A data processing method of a data processing apparatus according to a first aspect includes receiving data to be recognized including an ambiguous syllable that may be recognized as a number or a character, and includes at least one of the middle syllables. Selecting a plurality of conversion candidate data pieces from the recognition target data using an electronic dictionary including input data and output data corresponding thereto as conversion rules, and converting at least one of the plurality of conversion candidate data pieces And selecting data as data, and replacing and outputting the selected conversion target data among the recognition target data with output data of the electronic dictionary.

The data processing device according to the second aspect includes an input unit for receiving data to be recognized including an ambiguous syllable that may be recognized as a number or a character, and processing the recognition target data input through the input unit. A processing unit and an output unit for outputting a processing result by the processing unit, wherein the processing unit uses an electronic dictionary including input data including one or more intermediate syllables and output data corresponding thereto as conversion rules, Selecting a plurality of conversion candidate data pieces from the recognition target data, selecting at least one of the plurality of conversion candidate data pieces as conversion target data, and outputting the selected conversion target data among the recognition target data from the electronic dictionary It is replaced with data and provided to the output unit.

According to the third aspect, a computer-readable recording medium storing a computer program, when the computer program is executed by a processor, includes recognition target data including ambiguous syllables that may be recognized as numbers or characters. Receiving input, and selecting a plurality of pieces of conversion candidate data from the recognition target data using an electronic dictionary including input data including one or more intermediate syllables and output data corresponding thereto as a conversion rule And, selecting one or more of the plurality of conversion candidate data fragments as conversion target data, and replacing and outputting the selected conversion target data among the recognition target data with output data of the electronic dictionary. And an instruction for causing the processor to perform.

The computer program stored in the computer-readable recording medium according to the fourth aspect, when executed by a processor, receives data to be recognized including an ambiguous syllable that may be recognized as a number or a character; and , Selecting a plurality of pieces of conversion candidate data from the recognition target data using an electronic dictionary in which input data including one or more intermediate syllables and output data corresponding thereto are included as a conversion rule, and the plurality of The processor performs a method comprising selecting one or more of the pieces of conversion candidate data as conversion target data, and replacing and outputting the selected conversion target data among the recognition target data with output data of the electronic dictionary. Contains commands to do.

According to an embodiment, recognition target data including an ambiguous syllable that may be recognized as a number or a character may be mechanically converted and output to have high readability. For example, when the data to be recognized is address information, the reader's readability of the address information provided as a result of the recognition can be maximized by converting the highly readable part that must be recognized as a character and the part that is highly readable only when it is recognized as a number. I can.

In addition, when white space is included in the data to be recognized, white space is removed during data conversion, etc., to ensure flexible and fast processing speed, while restoring and providing necessary white space in the final result, so that readability due to changes in white space does not occur. Avoid.

1 is a block diagram of a data processing apparatus according to an exemplary embodiment.
2 is a flowchart illustrating a data processing method according to an exemplary embodiment.
3 is a flowchart illustrating a process of selecting data to be converted from among a data processing method according to an exemplary embodiment.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described later together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms, and only these embodiments make the disclosure of the present invention complete, and are common knowledge in the technical field to which the present invention pertains. It is provided to completely inform the scope of the invention to those who have, and the invention is only defined by the scope of the claims.

The terms used in the present specification will be briefly described, and the present invention will be described in detail.

The terms used in the present invention have been selected from general terms that are currently widely used while considering functions in the present invention, but this may vary depending on the intention or precedent of a technician working in the field, the emergence of new technologies, and the like. In addition, in certain cases, there are terms arbitrarily selected by the applicant, and in this case, the meaning of the terms will be described in detail in the description of the corresponding invention. Therefore, the terms used in the present invention should be defined based on the meaning of the term and the overall contents of the present invention, not a simple name of the term.

When it is said that a part in the specification'includes' a certain element, this means that other elements may be further included instead of excluding other elements unless specifically stated to the contrary.

In addition, the term'unit' used in the specification means software or hardware components such as FPGA or ASIC, and'unit' performs certain roles. However,'part' is not limited to software or hardware. The'unit' may be configured to be in an addressable storage medium, or may be configured to reproduce one or more processors. Thus, as an example,'unit' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, procedures, Includes subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, database, data structures, tables, arrays and variables. The functions provided within the components and'units' may be combined into a smaller number of components and'units' or may be further divided into additional components and'units'.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art may easily implement the present invention. In the drawings, parts not related to the description will be omitted in order to clearly describe the present invention.

1 is a block diagram of a data processing apparatus according to an exemplary embodiment.

Referring to FIG. 1, the data processing apparatus 100 may include an input unit 110, a processing unit 120, and an output unit 130, and may further include a storage unit 140. The processing unit 120 may include a data piece selection unit 121, a conversion target selection unit 122, a data conversion unit 123, and a blank removal/restore unit 124.

The input unit 110 receives data to be recognized and provides the data to the processing unit 120. Here, the recognition target data inputted through the input unit 110 may include an ambiguous syllable that may be recognized as a number or a character.

The processing unit 120 processes the recognition target data received through the input unit 110.

The data piece selection unit 121 of the processing unit 120 uses an electronic dictionary in which input data including one or more intermediate syllables and output data corresponding thereto are included as conversion rules, and a plurality of conversion candidate data from the recognition target data Select a piece. The electronic dictionary used by the data piece selection unit 121 may be stored in advance in the storage unit 140 or in an internal memory (not shown) of the processing unit 120 in advance.

The conversion target selection unit 122 of the processing unit 120 selects one or more of a plurality of conversion candidate data fragments as conversion target data. The data fragment selection unit 121, the conversion target selection unit 122, the data conversion unit 123, and the blank removal/restore unit 124 constituting the processing unit 120 are computing computing means such as a microprocessor, etc. It may include.

The data conversion unit 123 of the processing unit 120 replaces the conversion target data selected by the conversion target selection unit 122 among the recognition target data with output data from the electronic dictionary and provides the data to the output unit 130.

The white space removal/restore unit 124 of the processing unit 120 controls the white space of the recognition target data so that the recognition target data from which the white space has been removed is input into the electronic dictionary, and removes the white space that was removed before being provided to the output unit 130. Restore and provide. Here, the blank removal/restore unit 124 may perform blank recovery on a portion of the recognition target data that has not been replaced with the output data of the electronic dictionary.

The output unit 130 outputs a result of processing by the processing unit 120. For example, the output unit 130 may include an output interface, and may output the converted data provided from the processing unit 120 to another electronic device connected to the output interface under the control of the processing unit 120. . Alternatively, the output unit 130 may include a network card, and may transmit the converted data provided from the processing unit 120 through a network under the control of the processing unit 120. Alternatively, the output unit 130 may include a display device capable of displaying a result of processing by the processing unit 120 on a screen.

The storage unit 140 may store an electronic dictionary including input data including one or more intermediate syllables and output data corresponding thereto as a conversion rule, and may store a result of processing by the processing unit 120. For example, the storage unit 140 includes magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and floptical disks. It may be a computer-readable recording medium such as a hardware device specially configured to store and execute program instructions such as magneto-optical media or flash memory.

FIG. 2 is a flowchart illustrating a data processing method according to an exemplary embodiment, and FIG. 3 is a flowchart illustrating a process of selecting data to be converted among data processing methods according to an exemplary embodiment.

Hereinafter, a data processing method performed by the data processing apparatus 100 according to an embodiment of the present invention will be described in detail with reference to FIGS. 1 to 3.

First, an electronic dictionary including input data including one or more intermediate syllables and output data corresponding thereto as a conversion rule must be created, and the generated electronic dictionary is stored in the storage unit 140 in advance or the processing unit 120 It must be stored in the internal memory (not shown) in advance. Here, the electronic dictionary may be generated to exhibit a unique function according to the purpose or purpose of use of the data processing device 100. Such an electronic dictionary may include a plurality of data pairs including input data and output data acquired through machine learning as a transformation rule. For example, if the data processing device 100 is used for geographic address conversion, address data that can be represented in various forms may be used as output data, and pronunciation information for such output data may be used as corresponding input data. have. For example, as the output data of the electronic dictionary, "Samcheon 1-dong, Wansan-gu", "33 Seongjisan-ro, Samcheon 1-dong", "Samcheon 1-dong 573-2", "Samcheon 1-dong Community Center", "Samcheon 1-dong Geoma Park" , “3001 Building, “Apartment 3001”, etc., and pronunciation information of each output data may be used as corresponding input data In addition, output data of a conversion rule to be registered in the electronic dictionary is first determined. Then, a plurality of input data can be generated based on the determined output data. For example, if the output data of the conversion rule to be registered in the electronic dictionary is “Samcheon 1-dong 1004”, the possible input data corresponding thereto is “Three thousand It is possible to create "Ildong Angel", "Samcheonildong Ilgong Corporation", etc. Or, if the output data of the conversion rule to be registered in the electronic dictionary is "010-1234-5242", the corresponding input data is "Gongilgong Ilii Samsa It is possible to create "Oisai", "Gongil-gong one-two-three, again five-two-three", and "Gongil-gong 1,200-three and fifty-two hundred forty-two", etc. For example, "input data → output data" of the conversion rule registered in the electronic dictionary. The pair may include "Samcheonil-dong Angel → Samcheon1-dong 1004", "Samcheonil-dong Industrial Construction → Samcheon1-dong 1004", "Gongil-dong Ilyi Samsa and Oisai → 010-1234-5242." When comparing the case where the input data of the conversion rule to be registered in advance contains blanks with the case where blanks are not included, the input data has a relatively larger number in terms of the form of change, and this may be disadvantageous in terms of the amount of computation. , When registering the input data of the conversion rule in the electronic dictionary, it can be registered after removing spaces.

Recognition target data is input to the input unit 110 of the data processing device 100 while the electronic dictionary generated as described above is stored in the storage unit 140 or stored in the internal memory (not shown) of the processing unit 120 The recognition target data may include an ambiguous syllable that may be recognized as a number or a character. For example, "The address is Samcheonil-dong Angel Street, Wansan-gu, Samcheonil Children's Song", "My number, Gongilgong, and I'm a three-year-old Oisai" may be input as recognition target data (S210).

Then, as described above, when the blank is removed from the input data of the conversion rule registered in the previously generated electronic dictionary, the blank removal/restore unit 124 of the processing unit 120 performs the recognition target data input through step S210. Also, the blank may be removed and then provided to the data piece selection unit 121 of the processing unit 120 (S220).

Next, the data piece selection unit 121 uses an electronic dictionary in which input data including one or more intermediate syllables and output data corresponding thereto are included as conversion rules from the recognition target data provided through steps S210 and S220. Thus, a plurality of transform candidate data fragments are selected (S230). As described above, the process of selecting a data fragment for conversion may be viewed as a process of finding search results for all character string fragments that can be created from the data to be recognized. For example, if "Isa-goo Oh" is included in the data to be recognized, "Isa-goo Oh", "Isa-goo Oh", "Isa-goo Oh", "Isa-goo Lee", "Isa-goo Lee" ", "four," "four," and "gu" may all be selected as the transformation candidate data pieces. At this time, it is possible to place restrictions on the boundaries of the transformation candidate data fragments. For example, if there is no starting point of a fragment in the middle of a word, "four", "four phrase", and "phrase" cannot be transform candidate data fragments. In addition, restrictions such as that the end of a piece cannot be in the middle of a word or the beginning or end of a piece cannot be blank. For example, time complexity can be reduced by using a trie data structure and an algorithm for fragment retrieval.

Then, the conversion target selection unit 122 of the processing unit 120 selects one or more of the plurality of conversion candidate data fragments selected in step S230 as the conversion target data (S240).

The process of selecting a transform candidate data piece in step S240 is shown in more detail in the flowchart of FIG. 3. Prior to describing step S250, a process of selecting a transform candidate data fragment will be described in more detail with reference to FIG. 3.

First, the data piece selection unit 121 selects a conversion candidate data piece from recognition target data based on input data of a conversion rule registered in an electronic dictionary. For example, if the data to be recognized is "The address is Samcheonil-dong angel address, Wansan-gu Samcheonil-dong", "[0] address is Samcheonil-dong", "[4] Samcheonil-dong", "[4] Samcheonil-dong angel address. ", "[9]Angel's address", "[16] Samcheonil-dong, Wansan-gu", and "[20] Samcheonil-dong", etc. may be selected as the candidate data fragments. That is, the selected pieces of conversion candidate data are data registered as input data in the electronic dictionary. Here, the value of "[]" is location information on the recognition target data.

Subsequently, the conversion target selection unit 122 selects any one of the plurality of conversion candidate data pieces as conversion target data. Here, the longest one of the plurality of conversion candidate data fragments may be first selected as the conversion target data. For example, "[4] three thousand-il-dong angel address" may be selected as the conversion target data (S310).

In addition, the conversion target selection unit 122 determines a conversion candidate data fragment in which at least a portion of the conversion target data of step S310 overlaps positionally on the recognition target data among the plurality of conversion candidate data pieces. For example, the "[0] address may be identified as a piece of conversion candidate data in which "[4]", "[4]", "[9]" is overlapped (S320).

Subsequently, the conversion target selection unit 122 excludes the overlapping conversion candidate data piece identified in step S320 from among the plurality of conversion candidate data pieces. For example, "[0] address is Samcheonil-dong", "[4] Samcheonil-dong", and "[9] Angela address" have been removed, and conversion candidate data fragments include "[16] Samcheonil-dong" and "[20]3cheon. Only "all" can remain (S330).

In addition, if the conversion candidate data piece remains after excluding the overlapping conversion candidate data piece through step S330 among the plurality of conversion candidate data pieces, the conversion target selection unit 122 selects any one of the plurality of conversion candidate data pieces. The data selection is performed again from step S310. That is, by step S330, it repeats until there are no pieces of conversion candidate data in step S230 (S340).

Referring back to FIG. 2, the data conversion unit 123 of the processing unit 120 replaces the conversion target data selected in step S240 among the recognition target data with corresponding output data of the conversion rule registered in the electronic dictionary and outputs the data. For example, when the recognition target data is "The address is Samcheonil-dong Angel Address, Wansan-gu Samcheonil Children's Song", and in step S240, "[4] Samcheonil-dong Angel Address" is selected as the conversion target data, among the recognition target data, " [4] Samcheonil-dong angel address" can be replaced with "Samcheon 1-dong 1004 address". Alternatively, in the case where the recognition target data is "My number, Gongilgong Iisamsa and Oisai", and "[7] Gongilgong Igongilgong Iisamsa and Oisai" is selected as the conversion target data, "[7] of the recognition target data Gongilgong Ilyisamsa to Oisai" may be replaced with "010-1234-5242" (S250).

In addition, when a blank in the recognition target data is removed in step S220, the blank removal/restore 124 may restore the blank removed in step S220 with respect to the result of step S250. For example, restore the blank removed in step S220 for the portion of the recognition target data that was not replaced in step S250, that is, the portion that was not changed when comparing the replacement result data of step S250 and the original recognition target data input in step S210. can do. This blank restoration process will be described again below (S260).

In addition, the converted data by the processing unit 120 as described above may be output through the output unit 130 or may be stored in the storage unit 140. For example, the output unit 130 outputs the converted data through steps S250 and S260 to another electronic device connected to the output interface according to the control by the processing unit 120, transmits it through a network, or It can be displayed on the screen (S270).

Hereinafter, the process of removing the space in step S220 and the process of restoring space in step S260 will be described in more detail.

The blank removal/restore unit 124 must store blank removal information according to step S220 in order to restore blank space removed in step S220 in step S260.

For example, if the recognition target data input through step S210 is "My number and Gong Il Gong is between three companies and cucumbers", blank removal information may be stored as shown in Table 1 below.

My time arc Yo zero Work zero Work this three four on Five this four this phrase Yo XO OX XX XO OX XX XO OX XO OX XX OO OX XO OX XX XX XX

For example, as in the example of Table 1, blank removal information may be divided into "XO", "XX", "OX", and "OO" and stored. "X" means there are no leading or trailing spaces, and "O" means that there are leading or trailing spaces. That is, "XO" means that there is space only at the back, "XX" means that there is no space at the front and the back, "OX" means that there is space only at the front, and "OO" means that there are spaces at the front and back.

In addition, the blank removal/restore unit 124 needs to check the part replaced by step S250 before restoring the blank in step S260.

For example, when the blank removal/restore unit 124 checks the part replaced by step S250, by using the Levenstein's Algorithm, etc., the recognition target data from which the blank is removed and the converted data to which the replacement is applied are compared. , It is possible to check the deformed part for each location on the recognition target data. For example, if the recognition target data from which the blanks are removed is "My number required and public day is three companies," and the converted data to which the replacement is applied is "My number required 010-1234-5242", as shown in Table 2 below. You can check the replacement part.

My time arc Yo 0 One 0 - One 2 3 4 - 5 2 4 2 phrase Yo = = = = R R R R R R R R R R R R R = =

As in the example of Table 1, parts that have not been replaced can be identified by "=", and parts that have been replaced can be identified by "R".

Then, the blank removal/restore unit 124 combines Tables 1 and 2 as in the example of Table 3 below in step S260 and restores the blank at a position marked "0" among the parts separated by "=", " My number 010-1234-5242 can be output as the converted data.

My time arc Yo 0 One 0 - One 2 3 4 - 5 2 4 2 phrase Yo = = = = = = XO OX XX XO XX XX → My number is 010-1234-5242

Meanwhile, each step included in the data processing method according to the above-described exemplary embodiment may be implemented in a computer-readable recording medium that records a computer program including instructions for performing these steps.

In addition, each step included in the data processing method according to the above-described exemplary embodiment may be implemented in the form of a computer program stored in a computer-readable recording medium, programmed to include instructions for performing these steps.

Combinations of each step of each flowchart attached to the present invention may be performed by computer program instructions. Since these computer program instructions can be mounted on the processor of a general purpose computer, special purpose computer or other programmable data processing equipment, the instructions executed by the processor of the computer or other programmable data processing equipment are the functions described in each step of the flowchart. Will create a means of doing things. These computer program instructions can also be stored on a computer-usable or computer-readable recording medium that can be directed to a computer or other programmable data processing equipment to implement a function in a specific manner, so that the computer-readable or computer-readable recording medium. It is also possible to produce an article of manufacture that includes instruction means for performing the functions described in each step of the flow chart with instructions stored on the recording medium. Computer program instructions can also be mounted on a computer or other programmable data processing equipment, so a series of operating steps are performed on a computer or other programmable data processing equipment to create a computer-executable process to create a computer or other programmable data processing equipment. It is also possible for instructions to perform processing equipment to provide steps for executing the functions described in each step of the flowchart.

In addition, each step may represent a module, segment, or part of code that contains one or more executable instructions for executing the specified logical function(s). Further, it should be noted that in some alternative embodiments, the functions mentioned in the steps may occur out of order. For example, two steps shown in succession may in fact be performed substantially simultaneously, or the steps may sometimes be performed in the reverse order depending on the corresponding function.

The above description is merely illustrative of the technical idea of the present invention, and those of ordinary skill in the art to which the present invention pertains will be able to make various modifications and variations without departing from the essential quality of the present invention. Accordingly, the embodiments disclosed in the present invention are not intended to limit the technical idea of the present invention, but to explain the technical idea, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be interpreted by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

100: data processing device 110: input unit
120: processing unit 121: data piece selection unit
122: conversion target selection unit 123: data conversion unit
124: blank removal/restore unit 130: output unit
140: storage unit

Claims (9)

delete As a data processing method of a data processing device,
Receiving recognition target data including an ambiguous syllable that may be recognized as a number or a letter; and
Selecting a plurality of pieces of conversion candidate data from the recognition target data using an electronic dictionary in which input data including one or more intermediate syllables and output data corresponding thereto are included as a conversion rule; and
Selecting one or more of the plurality of conversion candidate data fragments as conversion target data,
Replacing the selected conversion target data among the recognition target data with output data of the electronic dictionary and outputting the data,
The step of selecting as the conversion target data,
Selecting any one of the plurality of conversion candidate data fragments as the conversion target data,
Grasping a transformation candidate data fragment in which at least a part of the transformation target data is positioned on the recognition target data among the plurality of transformation candidate data fragments;
Including the step of excluding the overlapping conversion candidate data pieces from among the plurality of conversion candidate data pieces
Data processing method. The method of claim 2,
Selecting the longest of the plurality of conversion candidate data fragments as the conversion target data
Data processing method. The method of claim 2,
If the transformation candidate data fragment remains after excluding the overlapping transformation candidate data fragment among the plurality of transformation candidate data fragments, re-performing from the step of selecting any one of the plurality of transformation candidate data fragments as the transformation target data
Data processing method. As a data processing method of a data processing device,
Receiving recognition target data including an ambiguous syllable that may be recognized as a number or a letter; and
Selecting a plurality of pieces of conversion candidate data from the recognition target data using an electronic dictionary in which input data including one or more intermediate syllables and output data corresponding thereto are included as a conversion rule; and
Selecting one or more of the plurality of conversion candidate data fragments as conversion target data,
Replacing the selected conversion target data among the recognition target data with output data of the electronic dictionary and outputting the data,
The input data of the electronic dictionary does not contain spaces,
In the selecting of the plurality of conversion candidate data pieces, blanks in the recognition target data are removed and input into the electronic dictionary to select the plurality of conversion candidate data pieces,
The outputting includes restoring the removed blank with respect to the recognition target data.
Data processing method. The method of claim 5,
The step of restoring the removed blank is performed on a portion of the recognition target data that has not been replaced with the output data of the electronic dictionary.
Data processing method. An input unit for receiving data to be recognized including an ambiguous syllable that may be recognized as a number or a character;
A processing unit that processes the recognition target data input through the input unit,
And an output unit for outputting a result of processing by the processing unit,
The processing unit,
Selecting a plurality of pieces of conversion candidate data from the recognition target data by using an electronic dictionary in which input data including one or more intermediate syllables and output data corresponding thereto are included as conversion rules, and the plurality of conversion candidate data At least one of the pieces is selected as conversion target data, and the selected conversion target data among the recognition target data is replaced with output data of the electronic dictionary and provided to the output unit,
When selecting the data to be converted,
Selecting any one of the plurality of conversion candidate data pieces as the conversion target data, and selecting a conversion candidate data piece in which at least a part of the conversion target data is positionally overlapped on the recognition target data among the plurality of conversion candidate data pieces Grasping and excluding the overlapping conversion candidate data pieces among the plurality of conversion candidate data pieces
Data processing unit. As a computer-readable recording medium storing a computer program,
The computer program, when executed by a processor,
Receiving recognition target data including an ambiguous syllable that may be recognized as a number or a letter; and
Selecting a plurality of pieces of conversion candidate data from the recognition target data using an electronic dictionary in which input data including one or more intermediate syllables and output data corresponding thereto are included as a conversion rule; and
Selecting one or more of the plurality of conversion candidate data fragments as conversion target data,
Replacing the selected conversion target data among the recognition target data with output data of the electronic dictionary and outputting the data,
The step of selecting as the conversion target data,
Selecting any one of the plurality of conversion candidate data fragments as the conversion target data,
Grasping a transformation candidate data fragment in which at least a part of the transformation target data is positioned on the recognition target data among the plurality of transformation candidate data fragments;
A computer-readable recording medium comprising instructions for causing the processor to perform a method including excluding the overlapping conversion candidate data pieces from among the plurality of conversion candidate data pieces. As a computer program stored in a computer-readable recording medium,
The computer program, when executed by a processor,
Receiving recognition target data including an ambiguous syllable that may be recognized as a number or a letter; and
Selecting a plurality of pieces of conversion candidate data from the recognition target data using an electronic dictionary in which input data including one or more intermediate syllables and output data corresponding thereto are included as a conversion rule; and
Selecting one or more of the plurality of conversion candidate data fragments as conversion target data,
Replacing the selected conversion target data among the recognition target data with output data of the electronic dictionary and outputting the data,
The step of selecting as the conversion target data,
Selecting any one of the plurality of conversion candidate data fragments as the conversion target data,
Grasping a transformation candidate data fragment in which at least a part of the transformation target data is positioned on the recognition target data among the plurality of transformation candidate data fragments;
A computer program comprising instructions for causing the processor to perform a method comprising excluding the overlapping conversion candidate data pieces from among the plurality of conversion candidate data pieces.

Images