JP2023149188A - Correction support method, correction support program, and information processing apparatus - Google Patents

Abstract

To present a correction candidate on the basis of contents of editing performed in the past by a worker.SOLUTION: An information processing apparatus 100 accepts a data table including a plurality of records. The information processing apparatus 100 specifies a plurality of pieces of correction candidate data being candidates to be corrected, out of a plurality of pieces of data included in the plurality of records on the basis of an analysis result of the accepted data table. The information processing apparatus 100 selects one piece of correction candidate data from the plurality of pieces of specified correction candidate data on the basis of records subjected to data editing in the past out of the plurality of records. The information processing apparatus 100 outputs the selected piece of correction candidate data.SELECTED DRAWING: Figure 2

Description

The present invention relates to a modification support method and the like.

Terminology used in various fields may not be standardized. For example, "name" and "name" have the same meaning but are written differently, so they are treated as completely different data on the system, making it impossible to accurately link the data.

By creating and using controlled vocabulary data that defines synonyms, hypernyms, and hyponyms, ambiguity in terms can be absorbed and problems such as those described above can be resolved. Controlled vocabulary data is a dictionary that summarizes the semantic relationships between multiple terms, and is created manually to prevent omissions in searches caused by ambiguity, homographs, and synonyms of terms.

FIG. 15 is a diagram illustrating an example of the data structure of controlled vocabulary data. As an example, the data structure of controlled vocabulary data will be explained in a tabular format. As shown in FIG. 15, the controlled vocabulary data 10 includes a term name column 10a, a representative word column 10b, a language column 10c, a representative word URI column 10d, and a hypernym URI column 10e.

The term name column 10a is set with term names from a glossary used in a specific field. Representative words are set in the representative word string 10b. A representative word is a name (heading) that represents multiple types of term names. For example, in the example shown in FIG. 15, "3D printer" is set as the representative word for the term "3D printer" and "3D printer."

The language column 10c is set with the language used by the operator when inputting the controlled vocabulary data. For example, "ja" indicates Japanese, and "en" indicates English. In the example shown in FIG. 15, the language in the first and second lines of the controlled vocabulary data 10 is "ja", so the operator can write the data in the first and second lines of the controlled vocabulary data 10 in Japanese. Indicates input. Since the language in the third line of the controlled vocabulary data 10 is "en," this indicates that the operator has input the data in the third line of the controlled vocabulary data 10 in English.

The URI (Uniform Resource Identifier) of the representative word is set in the representative word URI column 10d. The URI of a broader term is set in the URI column 10e of a broader term.

In the following description, a group of terms whose values (terms) in the term name column 10a and values (representative terms) in the representative word string have a synonymous relationship will be referred to as a "synonymous subgroup." In the example shown in FIG. 15, the information (records) in the first and second lines belong to the synonymous subgroup 10-sub1. The information on the third line belongs to the synonymous subgroup 10-sub2. It is assumed that each synonymous subgroup has a unique representative word, URI of the representative word, and URI of a broader term.

For example, synonymous subgroup 10-sub1 includes the unique representative word "3D printer," the unique representative word URI "http://myVocab/1," and the unique hypernym URI "http://myVocab/24." " is set.

Further, each synonym group having a different language is allowed to have a URI of the same representative word. They have the same representative word URI. Synonymous subgroups having the same representative word URI are collectively referred to as a synonymous group. In the example shown in FIG. 15, the information (records) in the first to third lines belong to the synonymous group 10-1. It is assumed that a synonym group has a unique URI of a broader term. For example, a unique hypernym URI "http://myVocab/24" is set for the synonym group 10-1.

Next, "inconsistency" regarding synonymous subgroups and synonymous groups will be defined. FIGS. 16 and 17 are diagrams for explaining mismatch. As an example, mismatches A, B, C, and D will be described in FIG. 16, and mismatches E and F will be described in FIG. 17.

FIG. 16 will be explained. “Inconsistency A” will be explained using the controlled vocabulary data 11a. Inconsistency A is when two or more types of representative words exist within the same synonymous subgroup. For example, in the controlled vocabulary data 11a, two types of representative words "3D printer" and "3D printer" are set in the same synonymous subgroup 11a-sub1, which corresponds to inconsistency A.

“Inconsistency B” will be explained using the controlled vocabulary data 11b. Inconsistency B is when there are two types of URIs for representative words in the same synonymous subgroup. For example, in the controlled vocabulary data 11b, two types of representative word URIs "http://myVocab/1" and "http://myVocab/2" are set for the same synonymous subgroup 11b-sub1, This corresponds to inconsistency B.

“Inconsistency C” will be explained using the controlled vocabulary data 11c. Inconsistency C is one in which there are two or more types of URI of a hypernym set in a certain synonymous subgroup. For example, in the controlled vocabulary data 11c, two types of hypernym URIs "http://myVocab/25" and "http://myVocab/24" are set for the same synonymous subgroup 11c-sub1, This corresponds to Inconsistency C.

“Inconsistency D” will be explained using the controlled vocabulary data 11d. Inconsistency D is one in which the URI of the representative word is the same between different synonymous subgroups that use the same language. For example, in the controlled vocabulary data 11d, the languages of the synonymous subgroups 11c-sub1 and 11c-sub2 are "ja", which is the same language, and the URIs of the representative words are "http://myVocab/1", which are the same, This corresponds to inconsistency D. If the language of either the synonymous subgroup 11c-sub1 or the synonymous subgroup 11c-sub2 is "en", the mismatch D does not apply.

Moving on to the description of FIG. 17. “Inconsistency E” will be explained using the controlled vocabulary data 11e. Inconsistency E is one in which there are two or more types of URIs for a broader term within the same synonymous group. For example, in the controlled vocabulary data 11e, two types of hypernym URIs "http://myVocab/24" and "http://myVocab/25" are set in the synonym group 11e-1, and an inconsistent E Applies to.

“Inconsistency F” will be explained using the controlled vocabulary data 11f. Inconsistency F is that the hierarchical relationship between the URI of a representative word and the URI of a broader term circulates between different synonymous groups. The controlled vocabulary data 11f includes synonymous groups 11f-1, 11f-2, and 11f-3. For example, in the synonym group 11f-3, the URI of the representative word is "http://myVocab/24" and the URI of the broader term is "http://myVcoab/2." In the synonym group 11f-2, the URI of the representative word is "http://myVocab/2" and the URI of the broader term is "http://myVcoab/1." In the synonym group 11f-1, the URI of the representative word is "http://myVocab/1" and the URI of the broader term is "http://myVcoab/24." That is, in the synonymous groups 11f-1, 11f-2, and 11f-3, the hierarchical relationship between the hypernym and the URI is circular, and this corresponds to mismatch F.

In order to correct the above-mentioned inconsistency, the amount of effort required to manually correct the value of each cell of the controlled vocabulary data is referred to as "correction cost." FIG. 18 is a diagram for explaining modification costs.

In the controlled vocabulary data 11a, there are two or more types of representative words in the same synonymous subgroup 11a-sub1, resulting in an inconsistency A. When the operator corrects the representative word "3D printer" in the controlled vocabulary data 11a to "3D printer," the inconsistency A is resolved and the data becomes controlled vocabulary data 12a. In this case, the modification cost is "1".

In the controlled vocabulary data 11e, there are two types of URIs of hypernyms in the same synonym group 11e-1, resulting in a mismatch E. Inconsistency E is resolved by the worker correcting the URI "http://myVocab/24" of the hypernym in the controlled vocabulary data 11e to "http://myVocab/25" (correcting in two places). , the controlled vocabulary data 12e. In this case, the modification cost is "2".

JP2020-52690A

As described with reference to FIG. 18, when manually correcting controlled vocabulary data, input errors or oversights may occur. Furthermore, correction work takes time, and the quality of the controlled vocabulary data may deteriorate.

Therefore, when controlled vocabulary data contains inconsistencies, it is required to present correction candidates. Further, it is preferable that correction candidates be presented based on the contents of edits made in the past by the operator. For example, when a worker is editing controlled vocabulary data in Japanese, if the worker can make corrections to correction candidates in Japanese, the burden on the worker when making corrections can be reduced.

In one aspect, an object of the present invention is to provide a modification support method, a modification support program, and an information processing device that can present modification candidates based on editing contents that have been edited by an operator in the past.

In the first proposal, the computer performs the following processing. A computer accepts a data table containing multiple records. Based on the analysis result of the received data table, the computer identifies a plurality of correction candidate data that are correction target candidates from among the plurality of data included in the plurality of records. The computer selects any correction candidate data from the specified plurality of correction candidate data based on records in which data has been edited in the past among the plurality of records. The computer outputs the selected correction candidate data.

It is possible to present correction candidates based on the contents of edits made by the operator in the past.

FIG. 1 is a diagram for explaining reference technology. FIG. 2 is a functional block diagram showing the configuration of the information processing apparatus according to this embodiment. FIG. 3 is a diagram showing an example of the data structure of controlled vocabulary data. FIG. 4 is a diagram for explaining the processing of the specifying section. FIG. 5 is a diagram showing an example of controlled vocabulary data when corrected using correction candidate A. FIG. 6 is a diagram showing an example of controlled vocabulary data when corrected using correction candidate B. FIG. 7 is a diagram showing an example of controlled vocabulary data when corrected using correction candidate C. FIG. 8 is a diagram showing an example of controlled vocabulary data when modified using modification candidate D. FIG. 9 is a diagram showing correction candidates X1 to X8. FIG. 10 is a diagram for explaining the first selection process. FIG. 11 is a diagram for explaining the second selection process. FIG. 12 is a diagram illustrating an example of a display screen generated by the display control unit. FIG. 13 is a flowchart showing the processing procedure of the information processing apparatus according to this embodiment. FIG. 14 is a diagram illustrating an example of the hardware configuration of a computer that implements the same functions as the information processing device of the embodiment. FIG. 15 is a diagram illustrating an example of the data structure of controlled vocabulary data. FIG. 16 is a diagram (1) for explaining mismatch. FIG. 17 is a diagram (2) for explaining mismatch. FIG. 18 is a diagram for explaining modification costs.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Examples of a modification support method, a modification support program, and an information processing apparatus disclosed in the present application will be described in detail below with reference to the drawings. Note that the present invention is not limited to this example.

Before explaining the information processing apparatus according to the present embodiment, a reference technique for presenting correction candidates when controlled vocabulary data includes an inconsistency will be explained. Such reference technology is not prior art. FIG. 1 is a diagram for explaining reference technology. The reference technique detects inconsistencies in controlled vocabulary data and extracts correction candidates for the inconsistencies. In the reference technique, the modification cost of each modification candidate is calculated, and the modification candidate with the minimum modification cost is displayed preferentially.

For example, in the controlled vocabulary data 13, the records in the first to third lines are synonymous groups. In this synonymous group, there are two types of hypernym URIs "http://myVocab/24" and "http://myVocab/25", so there is an inconsistency.

Correction candidates for inconsistencies in the controlled vocabulary data 13 are the following correction candidates (1) and (2).

Correction candidate (1) changes the hypernym URI “http://myVocab/24” to “http://myVocab/25” for the record whose term name is “3D printer” (record on the third line). It is something to be corrected. The modification cost of modification candidate (1) is "1".

Correction candidate (2) changes the URI of the broader term "http://myVocab/25" to "http://myVocab/25" for the records whose term names are "3D printer" and "3D printer" (records in the 1st and 2nd lines). ://myVocab/24”. The modification cost of modification candidate (2) is "2".

In the reference technique, since the cost of modifying the modification candidate (1) is lower than the cost of modifying the modification candidate (2), the modification candidate (1) is presented with priority over the modification candidate (2).

Next, problems with the reference technology will be explained. Each term has language information, and a situation may arise in which an operator temporarily edits only the information for a term in a specific language. However, the reference technology simply presents correction candidates in order of the lowest correction cost, without taking into account the perspective of which language the worker is editing the terms, and therefore Cannot present correction candidates in order.

For example, as explained in FIG. 1, in the reference technology, when presenting correction candidates for the controlled vocabulary data 13, the correction cost of correction candidate (1) is less than the correction cost of correction candidate (2). , the modification candidate (1) is presented preferentially over the modification candidate (2).

Here, if the operator who actually makes corrections to the controlled vocabulary data 13 edits terms for the language "ja", the highest priority correction candidate based on the operator's editing content is the language "ja". ” is a modification candidate (2) that modifies the records in the first and second lines. On the other hand, in the reference technique, modification candidate (1) is presented preferentially over modification candidate (2), and modification candidates cannot be sequentially presented based on the contents edited by the operator.

Next, an information processing apparatus according to this embodiment will be explained. The information processing apparatus according to the present embodiment presents correction candidates based on the operator's edited content, based on information about which language the operator is editing terms.

FIG. 2 is a functional block diagram showing the configuration of the information processing apparatus according to this embodiment. As shown in FIG. 2, the information processing device 100 includes a communication section 110, an input section 120, a display section 130, a storage section 140, and a control section 150.

The communication unit 110 receives various data from an external device via a network. Communication unit 110 is an example of a communication device. For example, the communication unit 110 may receive controlled vocabulary data 141, which will be described later, from an external device.

The input unit 120 is an input device that inputs various information to the control unit 150 of the information processing device 100. The input unit 120 corresponds to a keyboard, a mouse, a touch panel, etc. The operator may input data regarding the controlled vocabulary data 141 by operating the input unit 120.

The display unit 130 is a display device that displays information output from the control unit 150. For example, the display unit 130 displays controlled vocabulary data 141, information on correction candidates, and the like.

The storage unit 140 has controlled vocabulary data 141. The storage unit 140 corresponds to a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as an HDD (Hard Disk Drive).

The controlled vocabulary data 141 is dictionary data that summarizes the semantic relationships between a plurality of terms in order to prevent omissions in searches caused by ambiguity, homographs, and synonyms of terms.

FIG. 3 is a diagram showing an example of the data structure of controlled vocabulary data. As shown in FIG. 3, the controlled vocabulary data 141 includes a term name column 10a, a representative word column 10b, a language column 10c, a representative word URI column 10d, and a hypernym URI column 10e. The explanation regarding the term name string 10a, representative word string 10b, language string 10c, representative word URI string 10d, and hypernym URI string 10e is the same as that given in FIG. 15.

The control unit 150 includes a reception unit 151, a specification unit 152, a selection unit 153, and a display control unit 154. The control unit 150 is realized by hardwired logic such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), and an FPGA (Field Programmable Gate Array).

The reception unit 151 receives controlled vocabulary data 141 from the input unit 120 or the like. The reception unit 151 stores the controlled vocabulary data 141 in the storage unit 140. The reception unit 151 may receive the controlled vocabulary data 141 from an external device via the communication unit 110.

The specifying unit 152 specifies a plurality of correction candidates based on the controlled vocabulary data 141. Each process of the specifying unit 152 will be explained below. FIG. 4 is a diagram for explaining the processing of the specifying section.

FIG. 4 will be explained. First, the specifying unit 152 estimates the language in which the operator is editing based on the controlled vocabulary data 141. For example, the specifying unit 152 scans the values of the language string 10c and estimates the language with the highest frequency as the language being edited by the operator. In the example shown in FIG. 4, since the most frequently used language is "ja", the identification unit 152 estimates that the language being edited by the operator is "ja". In the following description, the language estimated by the specifying unit 152 and edited by the operator will be referred to as an "editing language."

Next, the specifying unit 152 classifies the records of the controlled vocabulary data 141 into a plurality of synonymous subgroups based on the term names in the term name column 10a and the representative words in the representative word column 10b of the controlled vocabulary data 141. do.

For example, the specifying unit 152 detects a connected component between the term name "3D printer" and the representative word "3D printer" in the second line of the controlled vocabulary data 141. As a result, the specifying unit 152 sets the records in the first to third lines in which "3D printer" and "3D printer" are set as term names or representative words to the same synonymous subgroup 50a.

The identification unit 152 identifies each record as different because the term names and representative words included in the records in the 4th to 6th lines of the controlled vocabulary data 141 are not common to the term names or representative words of other records. Set to synonymous subgroup. For example, the specifying unit 152 sets the record in the fourth line of the controlled vocabulary data 141 to the synonymous subgroup 50b. The specifying unit 152 sets the record in the fifth line of the controlled vocabulary data 141 to the synonymous subgroup 50c. The specifying unit 152 sets the record in the sixth line of the controlled vocabulary data 141 to the synonymous subgroup 50d.

Subsequently, the identification unit 152 identifies modification candidates and modification costs for each synonymous subgroup. For example, the specifying unit 152 specifies a correction candidate (a) and a correction candidate (b) based on the records of the synonymous subgroup 50a.

Candidate modification (a) is a modification that unifies the representative words of the synonymous subgroup 50a to "3D printer." The modification cost of modification candidate (a) is "2".

Correction candidate (b) is a correction that unifies the representative words of the synonymous subgroup 50a to "3D printer." The modification cost of modification candidate (b) is "1".

Subsequently, the identification unit 152 identifies modification candidates and modification costs between different synonymous subgroups. For example, when comparing the synonymous subgroups 50a, 50b, 50c, and 50d, the identification unit 152 identifies a modification candidate (c) and a modification candidate (d) between the synonymous subgroup 50a and the synonymous subgroup 50c. do.

Correction candidate (c) is a correction that unifies the URI of the representative word of the synonymous subgroup 50a to a value other than "http://myVocab/1". The modification cost of modification candidate (c) is "3".

Correction candidate (d) is a correction that unifies the URI of the representative word of the synonymous subgroup 50c to a value other than "http://myVocab/1". The modification cost of modification candidate (d) is "1".

From the above, the modification candidates within each synonymous subgroup and between the synonymous subgroups identified by the specifying unit 152 are the following modification candidate A, modification candidate B, modification candidate C, and modification candidate D.

Modification candidate A is a modification that performs modification candidate (a) and modification candidate (c). The modification cost of modification candidate A is "5", which is the sum of the modification cost "2" of modification candidate (a) and the modification cost "3" of modification candidate (c).

Modification candidate B is a modification that performs modification candidate (a) and modification candidate (d). The modification cost of modification candidate B is "3", which is the sum of the modification cost "2" of modification candidate (a) and the modification cost "1" of modification candidate (d).

Modification candidate C is a modification that performs modification candidate (b) and modification candidate (c). The modification cost of modification candidate C is "4", which is the sum of the modification cost "1" of modification candidate (b) and the modification cost "3" of modification candidate (c).

Modification candidate D is a modification that performs modification candidate (b) and modification candidate (d). The modification cost of the modification candidate D is "2", which is the sum of the modification cost "1" of the modification candidate (b) and the modification cost "1" of the modification candidate (d).

In the following explanation, the controlled vocabulary data will be explained when each of the correction candidates A, B, C, and D is corrected.

FIG. 5 is a diagram showing an example of controlled vocabulary data when corrected using correction candidate A. Controlled vocabulary data 141A shown in FIG. 5 is the result of correcting the controlled vocabulary data 141 shown in FIG. 4 using correction candidate A.

The specifying unit 152 classifies synonymous subgroups having the same representative word URI into a plurality of synonymous groups 51a, 51b, and 51c based on the URI of the representative word in the controlled vocabulary data 141A.

Subsequently, the identification unit 152 identifies modification candidates and modification costs for each synonymous group. For example, the identifying unit 152 identifies the modification candidate (a1) and the modification candidate (b1) based on the records of the synonymous group 51b.

The modification candidate (a1) is a modification that unifies the URI of the hypernym of the synonym group 51b to "http://myVocab/24". The modification cost of the modification candidate (a1) is "1". The modification candidate (a1) is a modification candidate that includes a record of the editing language "ja" as a modification target.

The modification candidate (b1) is a modification that unifies the URI of the hypernym of the synonym group 51b to "http://myVocab/10". The modification cost of the modification candidate (b1) is "1".

Subsequently, the identification unit 152 identifies modification candidates and modification costs between different synonymous groups. For example, the identifying unit 152 specifies the modification candidate (c1) because if the modification candidate (b1) is modified, a mismatch will occur between the synonymous group 51b and the synonymous group 51c.

The modification candidate (c1) is a modification that unifies the URI of the hypernym of the synonym group 51c to a value other than "http://myVocab/1". The modification cost of the modification candidate (c1) is "1".

As described above, when the specifying unit 152 makes a correction using the correction candidate A, the correction candidates within each synonymous group and between synonymous groups are the following correction candidates A-1 and A-2.

Modification candidate A-1 is a modification that performs modification candidate (a1). The modification cost of modification candidate A-1 is "1". Correction candidate A-1 is a correction candidate that includes a record of the editing language "ja" as a correction target.

Modification candidate A-2 is a modification that performs modification candidate (b1) and modification candidate (c1). The modification cost of modification candidate A-2 is "2", which is the sum of the modification cost "1" of modification candidate (b1) and the modification cost "1" of modification candidate (c1).

FIG. 6 is a diagram showing an example of controlled vocabulary data when corrected using correction candidate B. Controlled vocabulary data 141B shown in FIG. 6 is the result of correcting the controlled vocabulary data 141 shown in FIG. 4 using correction candidate B.

The specifying unit 152 classifies synonymous subgroups having the same representative word URI into a plurality of synonymous groups 52a, 52b, and 52c based on the URI of the representative word in the controlled vocabulary data 141B.

Subsequently, the identification unit 152 identifies modification candidates and modification costs for each synonymous group. For example, the identifying unit 152 identifies a modification candidate (a2) and a modification candidate (b2) based on the records of the synonymous group 52a.

The modification candidate (a2) is a modification that unifies the URI of the hypernym of the synonym group 52a to "http://myVocab/25". The modification cost of the modification candidate (a2) is "1".

The modification candidate (b2) is a modification that unifies the URI of the hypernym of the synonym group 52a to "http://myVocab/24". The modification cost of the modification candidate (b2) is "3". The modification candidate (b2) is a modification candidate that includes a record of the editing language "ja" as a modification target.

Subsequently, the identification unit 152 identifies modification candidates and modification costs between different synonymous groups. Note that since there is no inconsistency between the synonymous groups, the identifying unit 152 skips this process.

As described above, when the specifying unit 152 makes a correction using the correction candidate B, the correction candidates within each synonymous group and between synonymous groups are the following correction candidate B-1 and correction candidate B-2.

Modification candidate B-1 is a modification that performs modification candidate (a2). The modification cost of modification candidate B-1 is "1".

Modification candidate B-2 is a modification that performs modification candidate (b2). The modification cost of modification candidate B-2 is "3". Correction candidate B-2 is a correction candidate that includes a record of the editing language "ja" as a correction target.

FIG. 7 is a diagram showing an example of controlled vocabulary data when corrected using correction candidate C. Controlled vocabulary data 141C shown in FIG. 7 is the result of correcting the controlled vocabulary data 141 shown in FIG. 4 using correction candidate C.

The specifying unit 152 classifies synonymous subgroups having the same representative word URI into a plurality of synonymous groups 53a, 53b, and 53c based on the URI of the representative word in the controlled vocabulary data 141C.

Subsequently, the identification unit 152 identifies modification candidates and modification costs for each synonymous group. For example, the identifying unit 152 identifies a modification candidate (a3) and a modification candidate (b3) based on the records of the synonymous group 53b.

The modification candidate (a3) is a modification that unifies the URI of the hypernym of the synonym group 53b to "http://myVocab/24". The modification cost of the modification candidate (a3) is "1". The modification candidate (a3) is a modification candidate that includes a record of the editing language "ja" as a modification target.

The modification candidate (b3) is a modification that unifies the URI of the hypernym of the synonym group 53b to "http://myVocab/10". The modification cost of the modification candidate (b3) is "1".

Subsequently, the identification unit 152 identifies modification candidates and modification costs between different synonymous groups. For example, the specifying unit 152 specifies the correction candidate (c3) because if the above correction candidate (b3) is corrected, a mismatch will occur between the synonymous group 53b and the synonymous group 53c.

The modification candidate (c3) is a modification that unifies the URI of the hypernym of the synonym group 53c to a value other than "http://myVocab/1". The modification cost of the modification candidate (c3) is "1".

As described above, when the specifying unit 152 makes a correction using the correction candidate C, the correction candidates within each synonymous group and between synonymous groups are the following correction candidates C-1 and C-2.

Correction candidate C-1 is a correction that performs correction candidate (a3). The modification cost of modification candidate C-1 is "1". Correction candidate C-1 is a correction candidate that includes a record of the editing language "ja" as a correction target.

Modification candidate C-2 is a modification that performs modification candidate (b3) and modification candidate (c3). The modification cost of modification candidate C-2 is "2", which is the sum of the modification cost "1" of modification candidate (bC) and the modification cost "1" of modification candidate (c3).

FIG. 8 is a diagram showing an example of controlled vocabulary data when modified using modification candidate D. Controlled vocabulary data 141B shown in FIG. 8 is the result of correcting the controlled vocabulary data 141 shown in FIG. 4 using correction candidate D.

The specifying unit 152 classifies synonymous subgroups having the same representative word URI into a plurality of synonymous groups 54a, 54b, and 54c based on the URI of the representative word in the controlled vocabulary data 141D.

Subsequently, the identification unit 152 identifies modification candidates and modification costs for each synonymous group. For example, the identifying unit 152 identifies a modification candidate (a4) and a modification candidate (b4) based on the records of the synonymous group 53a.

The modification candidate (a4) is a modification that unifies the URI of the hypernym of the synonym group 54a to "http://myVocab/25". The modification cost of the modification candidate (a4) is "1".

The modification candidate (b4) is a modification that unifies the URI of the hypernym of the synonym group 54a to "http://myVocab/24". The modification cost of the modification candidate (b4) is "3". The modification candidate (b4) is a modification candidate that includes a record of the editing language "ja" as a modification target.

Subsequently, the identification unit 152 identifies modification candidates and modification costs between different synonymous groups. Note that since there is no inconsistency between the synonymous groups, the identifying unit 152 skips this process.

As described above, when the specifying unit 152 makes a correction using the correction candidate D, the correction candidates within each synonymous group and between synonymous groups are the following correction candidates D-1 and D-2.

Modification candidate D-1 is a modification that performs modification candidate (a4). The modification cost of modification candidate D-1 is "1".

Modification candidate D-2 is a modification that performs modification candidate (b4). The modification cost of modification candidate D-2 is "3". Correction candidate D-2 is a correction candidate that includes a record of the editing language "ja" as a correction target.

The specifying unit 152 specifies all correction candidates and correction costs by executing the processes described in FIGS. 4 to 8. For example, all correction candidates include correction candidates X1, X2, X3, X4, X5, X6, X7, and X8, which will be described next. FIG. 9 is a diagram showing correction candidates X1 to X8.

As shown in FIG. 9, modification candidate X1 is a modification that performs modification candidate A and modification candidate A-1. The modification cost of modification candidate A is "5", the modification cost of modification candidate A-1 is "1", and the modification cost of modification candidate X1 is "6". Correction candidate A-1 is a correction candidate that includes a record of the editing language "ja" as a correction target.

Modification candidate X2 is a modification that performs modification candidate A and modification candidate A-2. The modification cost of modification candidate A is "5", the modification cost of modification candidate A-2 is "2", and the modification cost of modification candidate X2 is "7".

Modification candidate X3 is a modification that performs modification candidate B and modification candidate B-1. The modification cost of modification candidate B is "3", the modification cost of modification candidate B-1 is "1", and the modification cost of modification candidate X3 is "4".

Modification candidate X4 is a modification that performs modification candidate B and modification candidate B-2. The modification cost of modification candidate B is "3", the modification cost of modification candidate B-2 is "3", and the modification cost of modification candidate X4 is "6". Correction candidate B-2 is a correction candidate that includes a record of the editing language "ja" as a correction target.

Modification candidate X5 is a modification that performs modification candidate C and modification candidate C-1. The modification cost of modification candidate C is "3", the modification cost of modification candidate C-1 is "1", and the modification cost of modification candidate X5 is "4". Correction candidate C-1 is a correction candidate that includes a record of the editing language "ja" as a correction target.

Modification candidate X6 is a modification that performs modification candidate C and modification candidate C-2. The modification cost of modification candidate C is "3", the modification cost of modification candidate C-2 is "2", and the modification cost of modification candidate X6 is "5".

Modification candidate X7 is a modification that performs modification candidate D and modification candidate D-1. The modification cost of modification candidate D is "2", the modification cost of modification candidate D-1 is "1", and the modification cost of modification candidate X7 is "3".

Modification candidate X8 is a modification that performs modification candidate D and modification candidate D-2. The modification cost of modification candidate D is "2", the modification cost of modification candidate D-2 is "3", and the modification cost of modification candidate X8 is "5". Correction candidate D-2 is a correction candidate that includes a record of the editing language "ja" as a correction target.

The specifying unit 152 outputs information about the correction candidates X1 to X8 described in FIG. 9 to the selecting unit 153. In the following description, the information on the correction candidates X1 to X8 will be collectively referred to as "correction candidate information." It is assumed that the modification candidate information includes, for each modification candidate, the modification content, modification cost, and information as to whether or not the modification target includes a record of the editing language "ja". .

The selection unit 153 selects a modification candidate to be displayed from among the modification candidates X1 to X8 based on the modification candidate information. For example, the selection unit 153 executes either the first selection process or the second selection process. It is assumed that which selection process is to be executed is set in advance.

First, the first selection process will be explained. FIG. 10 is a diagram for explaining the first selection process. The selection unit 153 refers to the correction candidate information and selects among the correction candidates X1 to X8, a correction candidate group G1 having correction candidates that target records with the editing language "ja" and records with the editing language "ja". is classified into a group G2 of correction candidates that does not have any correction candidates to be corrected.

As shown in FIG. 10, group G1 includes correction candidates X1, X4, X5, and X8. The selection unit 153 sorts the modification candidates X1, X4, X5, and X8 of the group G1 in descending order of modification cost. As for correction candidates with the same correction cost, either one may be selected first. For example, as a result of the sorting, the modification candidates X1, X4, X5, and X8 of the group G1 are arranged in order from the top to the modification candidates X5, X8, X1, and X4. The selection unit 153 preferentially selects correction candidates with low correction costs. The selection unit 153 may select N correction candidates from the beginning. N is a preset natural number.

Group G2 includes correction candidates X2, X3, X6, and X7. The selection unit 153 sorts the modification candidates X2, X3, X6, and X7 of the group G2 in descending order of modification cost. As for correction candidates with the same correction cost, either one may be selected first. For example, as a result of the sorting, the modification candidates X2, X3, X6, and X7 of the group G2 are arranged in the order of modification candidates X7, X3, X6, and X2 from the top. The selection unit 153 preferentially selects correction candidates with low correction costs. The selection unit 153 may select N correction candidates from the beginning.

The selection unit 153 outputs information on the selected correction candidates to the display control unit 154. The information on the selected modification candidates includes the modification details, modification cost, and sorting order.

Next, the second selection process will be explained. FIG. 11 is a diagram for explaining the second selection process. The selection unit 153 refers to the modification candidate information and modifies the modification cost of a modification candidate among the modification candidates X1 to X8 that has a modification candidate that targets a record with the editing language "ja", using a predetermined weight. In this embodiment, the predetermined weight is "0.6". For example, it is assumed that the predetermined weight is set in advance to be greater than 0 and less than 1.

As explained with reference to FIG. 9, among the modification candidates X1 to X8, the modification candidates X1, X4, X5, and X8 have modification candidates that target records in the editing language "ja". Therefore, as shown in FIG. 11, the selection unit 153 uses the value "3.6" obtained by multiplying the modification cost "6" of the modification candidate X1 by the weight "0.6" as the new modification cost of the modification candidate X1. Set to . The selection unit 153 sets the value "3.6" obtained by multiplying the modification cost "6" of the modification candidate X4 by the weight "0.6" as the new modification cost of the modification candidate X4.

The selection unit 153 sets the value "2.4" obtained by multiplying the modification cost "4" of the modification candidate X5 by the weight "0.6" as the new modification cost of the modification candidate X5. The selection unit 153 sets the value "3" obtained by multiplying the modification cost "5" of the modification candidate X8 by the weight "0.6" as the new modification cost of the modification candidate X8.

The selection unit 153 sorts the modification candidates X1 to X8 in descending order of modification cost, taking into account the modification cost after modification. For example, as a result of sorting, the modification candidates X1 to X8 are arranged in the following order: modification candidates X5, X7, X8, X1, X4, X3, X3, X6, X, 2. The selection unit 153 preferentially selects correction candidates with low correction costs. The selection unit 153 may select N correction candidates from the beginning.

The selection unit 153 outputs information on the selected correction candidates to the display control unit 154. The information on the selected modification candidates includes the modification details, modification cost, and sorting order.

Returning to the explanation of FIG. 2. The display control unit 154 generates a display screen based on the information on the correction candidates selected by the selection unit 153, and outputs the generated display screen to the display unit 130 for display. For example, the display control unit 154 arranges on the display screen text information that associates the modification content of the modification candidate with the modification cost. The display control unit 154 sets the correction candidates so that the one with the lowest order (the order when sorted by the selection unit 153) is placed at the top of the display screen.

FIG. 12 is a diagram illustrating an example of a display screen generated by the display control unit. In the example shown in FIG. 12, a modification candidate (1) and a modification candidate (2) are set on the display screen 60, and the modification content and modification cost of the modification candidate (1) and modification candidate (2) are set. There is.

Next, an example of a processing procedure of the information processing apparatus according to this embodiment will be described. FIG. 13 is a flowchart showing the processing procedure of the information processing apparatus according to this embodiment. As shown in FIG. 13, the reception unit 151 of the information processing device 100 receives controlled vocabulary data 141 (step S101). The identification unit 152 of the information processing device 100 estimates the language in which the operator is editing based on the controlled vocabulary data 141 (step S102).

The specifying unit 152 specifies correction candidates and correction costs in the synonymous subgroup based on the controlled vocabulary data 141 (step S103). The specifying unit 152 specifies correction candidates and correction costs between different synonymous subgroups based on the controlled vocabulary data 141 (step S104).

The specifying unit 152 executes corrections to the correction candidates within the synonymous subgroup and the correction candidates between the synonymous subgroups to the controlled vocabulary data 141 (step S105). The specifying unit 152 specifies correction candidates and correction costs within the synonymous group for the corrected controlled vocabulary data 141 (step S106).

The specifying unit 152 specifies correction candidates and correction costs between different synonymous groups for the corrected controlled vocabulary data 141 (step S107). The specifying unit 152 generates correction candidate information (step S108).

The selection unit 153 of the information processing device 100 selects correction candidates to be displayed based on the language edited by the creator and the correction candidate information (step S109). The display control unit 154 of the information processing device 100 generates a display screen based on the selected correction candidates (step S110). The display control unit 154 outputs the display screen to the display unit 130 (step S111).

Next, the effects of the information processing device 100 according to this embodiment will be explained. The information processing device 100 specifies a plurality of correction candidates based on the controlled vocabulary data 141, and performs a process of selecting and displaying a correction candidate according to the content edited by the worker from among the plurality of correction candidates. As a result, correction candidates can be presented based on the contents edited by the operator, and the burden on the operator can be reduced.

Based on the languages set in the language string of the controlled vocabulary data 141, the information processing device 100 selects a language type whose appearance frequency is higher than other language types as the language type being edited by the operator. presume. This allows the operator to specify the type of language being edited.

The information processing apparatus 100 calculates a modification cost for each modification candidate, and selects a modification candidate with a lower modification cost with priority. Here, if the record to be modified is a record corresponding to the language type edited by the operator, the information processing device 100 adds a weight ( Correct by multiplying by 0<weight value<1). This makes it easier for the operator to preferentially select correction candidates related to the record corresponding to the language type being edited.

Note that the information processing device 100 used the appearance frequency of the language set in the language string 10c of the controlled vocabulary data 141 when identifying the type of language that the operator is editing, but this is not limited to this. It is not something that will be done. The identification unit 152 of the information processing device 100 scans the editing time set for each record of the controlled vocabulary data 141, and identifies the language type set in the record with the latest editing time as the language being edited by the operator. It may also be specified as a type.

Next, an example of the hardware configuration of a computer that implements the same functions as the information processing apparatus 100 shown in the above embodiment will be described. FIG. 14 is a diagram illustrating an example of the hardware configuration of a computer that implements the same functions as the information processing device of the embodiment.

As shown in FIG. 14, the computer 200 includes a CPU 201 that executes various calculation processes, an input device 202 that receives data input from a user, and a display 203. The computer 200 also includes a communication device 204 and an interface device 205 that exchange data with an external device or the like via a wired or wireless network. The computer 200 also includes a RAM 206 that temporarily stores various information and a hard disk device 207. Each device 201-207 is then connected to a bus 208.

The hard disk device 207 has a reception program 207a, a specific program 207b, a selection program 207c, and a display control program 207d. Further, the CPU 201 reads each program 207a to 207d and expands it into the RAM 206.

The reception program 207a functions as a reception process 206a. The specific program 207b functions as a specific process 206b. The selection program 207c functions as a selection process 206c. The display control program 207d functions as a display control process 206d.

The processing of the reception process 206a corresponds to the processing of the reception unit 151. The processing of the identification process 206b corresponds to the processing of the identification unit 152. The processing of the selection process 206c corresponds to the processing of the selection unit 153. The processing of the display control process 206d corresponds to the processing of the display control unit 154.

Note that each of the programs 207a to 207d does not necessarily need to be stored in the hard disk device 207 from the beginning. For example, each program is stored in a "portable physical medium" such as a flexible disk (FD), CD-ROM, DVD, magneto-optical disk, or IC card that is inserted into the computer 200. Then, the computer 200 may read and execute each program 207a to 207d.

Regarding the embodiments including each of the above examples, the following additional notes are further disclosed.

(Additional note 1) Accepts a data table containing multiple records,
Based on the analysis result of the received data table, identifying a plurality of correction candidate data that are correction target candidates among the plurality of data included in the plurality of records;
Selecting any correction candidate data from the identified plurality of correction candidate data based on records in which data has been edited in the past among the plurality of records;
outputting the selected correction candidate data;
A correction support method characterized in that processing is executed by a computer.

(Additional Note 2) The language types used when editing the data are set in the plurality of records, and the identifying process prioritizes the language types set in the plurality of records. Supplementary Note 1, wherein the process of further specifying and selecting a language type selects any correction candidate data from the identified plurality of correction candidate data based on the prioritized language type. Correction support method described in.

(Additional Note 3) The identifying process further specifies a modification cost based on the number of modifications to the data table indicated in the modification candidate data, and the selecting process includes modifications to modify records related to the preferred language type. The modification support method according to appendix 2, characterized in that modification is performed to reduce the modification cost to the candidate data, and priority is given to selecting modification candidate data with a small modification cost.

(Additional note 4) The identifying process is to identify a language type whose appearance frequency is higher than other language types among the language types set in the plurality of records as the priority language type. The correction support method according to appendix 2, characterized by:

(Appendix 5) Accepts a data table containing multiple records,
Based on the analysis result of the received data table, identifying a plurality of correction candidate data that are correction target candidates among the plurality of data included in the plurality of records;
Selecting any correction candidate data from the identified plurality of correction candidate data based on records in which data has been edited in the past among the plurality of records;
outputting the selected correction candidate data;
A modification support program that causes a computer to perform processing.

(Additional note 6) The language types used when editing the data are set in the plurality of records, and the identifying process prioritizes the language types set in the plurality of records. Supplementary note 5, wherein the process of further specifying and selecting a language type selects any correction candidate data from the identified plurality of correction candidate data based on the prioritized language type. Modification support program as described in .

(Additional Note 7) The identifying process further specifies a modification cost based on the number of modifications to the data table indicated in the modification candidate data, and the selecting process includes modifications to modify records related to the preferred language type. The correction support program according to appendix 6, characterized in that the correction support program performs correction to reduce the correction cost to the candidate data, and selects correction candidate data with a lower correction cost with priority.

(Additional note 8) The identifying process is to identify, among the language types set in the plurality of records, a language type whose appearance frequency is higher than other language types as the priority language type. The correction support program described in appendix 6, characterized by:

(Appendix 9) Accepts a data table containing multiple records,
Based on the analysis result of the received data table, identifying a plurality of correction candidate data that are correction target candidates among the plurality of data included in the plurality of records;
Selecting any correction candidate data from the identified plurality of correction candidate data based on records in which data has been edited in the past among the plurality of records;
outputting the selected correction candidate data;
An information processing device that has a control unit that executes processing.

(Additional Note 10) The plurality of records are set with language types used when editing the data, and the control unit selects a language that is prioritized among the language types set in the plurality of records. The information processing device according to appendix 9, further comprising: further specifying a type of the preferred language, and selecting one of the specified correction candidate data from among the specified plurality of correction candidate data.

(Supplementary Note 11) The control unit further specifies a modification cost based on the number of modifications to the data table indicated in the modification candidate data, and the selecting process includes modification candidates for modifying records related to the prioritized language type. The information processing apparatus according to appendix 9, wherein correction is performed to reduce the correction cost to the data, and correction candidate data with a small correction cost is selected with priority.

(Additional Note 12) The control unit may specify, among the language types set in the plurality of records, a language type whose appearance frequency is higher than other language types as the priority language type. The information processing device according to feature Supplementary Note 10.

100 Information processing device 110 Communication unit 120 Input unit 130 Display unit 140 Storage unit 141 Controlled vocabulary data 150 Control unit 151 Reception unit 152 Specification unit 153 Selection unit 154 Display control unit

Claims (6)

Accepts a data table containing multiple records,
Based on the analysis result of the received data table, identifying a plurality of correction candidate data that are correction target candidates among the plurality of data included in the plurality of records;
Selecting any correction candidate data from the identified plurality of correction candidate data based on records in which data has been edited in the past among the plurality of records;
outputting the selected correction candidate data;
A correction support method characterized in that processing is executed by a computer. Language types used when editing the data are set in the plurality of records, and the identifying process selects a preferred language type among the language types set in the plurality of records. 2. The process of further specifying and selecting selects any correction candidate data from the identified plurality of correction candidate data based on the type of the prioritized language. Correction support method. The identifying process further specifies a modification cost based on the number of modifications to the data table indicated in the modification candidate data, and the selecting process includes a modification to the modification candidate data that modifies a record related to the preferred language type. 3. The modification support method according to claim 2, wherein modification is performed to reduce the cost, and priority is given to selecting modification candidate data with a small modification cost. The identifying process is characterized in that, among the language types set in the plurality of records, a language type whose appearance frequency is higher than other language types is specified as the priority language type. The correction support method according to claim 2. Accepts a data table containing multiple records,
Based on the analysis result of the received data table, identifying a plurality of correction candidate data that are correction target candidates among the plurality of data included in the plurality of records;
Selecting any correction candidate data from the identified plurality of correction candidate data based on records in which data has been edited in the past among the plurality of records;
outputting the selected correction candidate data;
A modification support program that causes a computer to perform processing. Accepts a data table containing multiple records,
Based on the analysis result of the received data table, identifying a plurality of correction candidate data that are correction target candidates among the plurality of data included in the plurality of records;
Selecting any correction candidate data from the identified plurality of correction candidate data based on records in which data has been edited in the past among the plurality of records;
outputting the selected correction candidate data;
An information processing device that has a control unit that executes processing.

Images