PROGRAM, INFORMATION PROCESSING APPARATUS, AND METHOD

Abstract

To contribute to the prompt determination of a company's own product number corresponding to a product number of another company. [Solution] In a system that manages the product numbers of materials used in transactions between a first user and a second user, the product number managed by the first user and the product number managed by the second user are not necessarily the same even for the same material, and a memory unit is configured to store a history that associates the product number of the material at the first user with the product number of the material at the second user, and executes the steps of accepting input of the product number of the first user, determining the display mode of candidate product numbers at the second user for the material corresponding to the input product number of the first user using the input product number of the first user and the history, and displaying the extracted product number candidates to the second user.

Description

[Technical field]

[0001] The present disclosure relates to a program, an information processing device, and a method. [Background technology]

[0002] There is a technology that includes a master file that stores item information consisting of multiple supplier item codes corresponding to one orderer item code, and the multiple selectable shipping instruction items are at least multiple supplier item codes corresponding to the one orderer item code, so that a company can define multiple item codes that are convenient for its own company in relation to other companies' item codes, and can quickly issue shipping instructions using its own item codes (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2003-187114 A Summary of the Invention [Problem to be solved by the invention]

[0004] The above technology associates and stores the part numbers of other companies with the company's part numbers in advance, and searches for the company's part numbers using the other company's part numbers. However, when there are multiple material candidates for a company's part number, it is not possible to determine the appropriate material by taking into account the price, quality, etc. Therefore, the company's part numbers must be determined manually to correspond to the other company's part numbers, which is an extremely time-consuming task. Furthermore, an object of the present disclosure is to provide technology that enables each business operator to maintain the master data necessary for order placement and receipt operations without having to be aware of updating their own company's master data, and that can contribute to determining their own company's part numbers that correspond to other companies' part numbers.

[0005] Therefore, the present disclosure provides a technique that can contribute to quickly determining a company's own product number that corresponds to a company's product number. [Means for solving the problem]

[0006] The program of the present disclosure is a program for operating a computer having a processor and a memory unit in a system that manages product numbers of materials used in transactions between a first user and a second user, wherein even for the same material, the product number managed by the first user and the product number managed by the second user are not necessarily the same product number, and the memory unit is configured to store a history in which the second user has associated the product numbers of the material at the first user with the product numbers of the material at the second user, and executes the steps of accepting input of the product number of the first user, using the input product number of the first user and the history to extract candidate product numbers at the second user for the material corresponding to the input product number of the first user, and displaying the extracted candidate product numbers to the second user. Effect of the Invention

[0007] The present disclosure can contribute to the rapid determination of a company's own part number that corresponds to a company's other company's part number. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing the overall configuration of a system 1. [Diagram 2] FIG. 2 is a diagram showing a functional configuration of a server 10. [Diagram 3] FIG. 2 is a diagram illustrating a data structure of a database. [Figure 4] FIG. 2 is a diagram illustrating a data structure of a database. [Diagram 5] FIG. 2 is a diagram illustrating a data structure of a database. [Figure 6] FIG. 2 is a diagram illustrating a data structure of a database. [Figure 7] FIG. 2 is a diagram illustrating a data structure of a database. [Figure 8] FIG. 2 is a diagram showing the functional configuration of a user terminal 20. [Figure 9] FIG. 2 is a diagram showing an example of a processing flow in the system 1. [Figure 10] FIG. 2 is a diagram showing an example of a processing flow in the system 1. [Figure 11] FIG. 2 is a diagram showing an example of a processing flow in the system 1. [Figure 12] FIG. 2 is a diagram showing an example of a processing flow in the system 1. [Figure 13] 13 is an example of a screen displayed on a user terminal 20. [Figure 14] 13 is an example of a screen displayed on a user terminal 20. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. In the following description, the same components are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed description thereof will not be repeated.

[0010] <Summary of this Disclosure> The present disclosure contributes to the rapid determination of the part number of the second user, which is the same company, corresponding to the part number of the first user, which is another company, in a transaction in which the part number managed by the first user and the part number managed by the second user are not necessarily the same even for the same material. The technology of the present disclosure makes it possible to easily convert part numbers that differ between companies, such as the ordering side and the receiving side, in order to support the completion of an incomplete BOM (Bill of Materials). In addition, it is assumed that the first user and the second user are companies, but unless otherwise specified, the users acting as the first user and the second user are employees of the first user and the second user.

[0011] In the conventional technology, when there are multiple candidate materials for a competitor's product number in such a transaction situation, it is necessary to determine the appropriate material by taking into consideration the price, quality of the material, etc., which is extremely time-consuming. In addition, in industries such as the apparel industry where the same part number is not used continuously even for the same material, it is not possible to search the database to find the company's part number. For this reason, in such industries, there is a problem that it is particularly time-consuming to search for the company's part number that corresponds to a new company's part number that has been changed by another company. In addition, in such industries, it is difficult to search for the appropriate company's part number even if the company's part number is different for each apparel brand with which the other company does business. This is because the name of the other party's part number is different for each apparel brand with which the other company does business. This is because, as a problem unique to the apparel industry, there is no industry-wide standard or name for a single material. The person in charge of processing the order data searches the company's core system for the company's part number based on the other party's part number, but it is difficult to accurately search for the company's part number when the part number differs or there are multiple parts for the same material depending on the other party. As a result, the person in charge at the company does not know which material to use, so it is necessary to inquire of the other party or make an estimate from past transaction data.

[0012] The technology disclosed herein allows users to maintain the master data required for order placement and receipt operations through their daily work of selecting product numbers, without the need for each business operator to be aware of constantly updating the master data, and contributes to determining the company's own product number that corresponds to the product number of another company. In the following explanation, the other party will be referred to as the first user and the company will be referred to as the second user.

[0013] <1. Embodiment> <1.1. Overall configuration of System 1> System 1 is a system for extracting candidates for product numbers of a second user according to the present disclosure. 1, the system 1 includes a server 10, a user terminal 20, and a network 80. The server 10 and the user terminal 20 are communicatively connected via the network 80.

[0014] The user terminal 20 is a terminal operated by a first user and a second user.

[0015] The server 10 is a device that performs a function of extracting candidates for a second user's part number from a first user's part number. The server 10 is realized by a mobile terminal such as a smartphone or tablet compatible with a mobile communication system, a stationary PC (Personal Computer), a laptop PC, or the like.

[0016] As shown in FIG. 1, the server 10 includes a communication IF 12, an input / output IF 13, a memory 15, a storage 16, and a processor 19.

[0017] The communication IF 12 is an interface for inputting and outputting signals so that the server 10 can communicate with an external device. The input / output IF 13 functions as an interface with an input device for receiving an input operation from a user and an output device for presenting information to a user. The input / output IF 13 is for temporarily storing a program and data processed by the program, and is a volatile memory such as a dynamic random access memory (DRAM). The memory 15 is a storage device for saving data, and is a flash memory or a hard disk drive (HDD). The storage 16 is hardware for executing an instruction set described in a program, and is composed of an arithmetic unit, a register, peripheral circuits, etc. The processor 19 is hardware for executing an instruction set described in a program stored in the memory 15, and is composed of an arithmetic unit, a register, peripheral circuits, etc.

[0018] The user terminal 20 is a terminal device operated by a first user and a second user. The user terminal 20 is realized by a mobile terminal such as a smartphone or tablet compatible with a mobile communication system, a stationary PC (Personal Computer), a laptop PC, etc. The user terminal 20 used by the first user and the user terminal used by the second user do not need to be the same terminal.

[0019] As shown in FIG. 1, the user terminal 20 includes a communication IF 22, an input / output IF 23, a memory 25, a storage 26, and a processor 29.

[0020] The communication IF 22 is an interface for inputting and outputting signals so that the user terminal 20 can communicate with an external device. The input / output IF 23 functions as an interface with an input device for receiving an input operation from a user and an output device for presenting information to a user. The input / output IF 23 is for temporarily storing a program and data processed by the program, and is a volatile memory such as a dynamic random access memory (DRAM). The memory 25 is a storage device for saving data, and is a flash memory or a hard disk drive (HDD). The storage 26 is hardware for executing an instruction set described in a program, and is composed of an arithmetic unit, a register, peripheral circuits, etc. The processor 29 is hardware for executing an instruction set described in a program stored in the memory 25, and is composed of an arithmetic unit, a register, peripheral circuits, etc.

[0021] <1.2. Functional configuration of server 10> 2 is a diagram showing a functional configuration of the server 10. As shown in FIG. 2, the server 10 fulfills the functions of a communication unit 101, a storage unit 102, and a control unit 103.

[0022] The communication unit 101 performs processing for communicating with an external device.

[0023] The storage unit 102 stores data and programs to be used.

[0024] Specifically, the storage unit 102 stores a first DB (database) 121, a second DB 122, a third DB 123, a fourth DB 124, a fifth DB 125, and the like.

[0025] The following describes the data structures of the first DB 121 to the fifth DB 125 stored in the server 10 shown in Figures 3 to 7. Figures 3 to 7 are diagrams showing the data structures of each database.

[0026] The first DB 121 is a database that holds a history of the second user associating the part numbers of materials in the first user with the part numbers of materials in the second user. As shown in Fig. 3, the first DB 121 includes an item "ID", an item "master ID", an item "company part number", an item "first user part number", an item "first user surface name", and an item "date and time".

[0027] The item "ID" is information for identifying a record.

[0028] The item "master ID" is the name of the material. Specifically, the item "material" stores the name, material, etc. of the material.

[0029] The item "company product number" is the product number of the material in question for the second user. There is only one company product number, but one company product number may be linked to multiple product numbers of the first user.

[0030] The item "first user product number" is the product number of the first user for the material.

[0031] The item "First user surface name" is the surface name of the specification for the first user.

[0032] The item "date and time" is the date and time when the record was registered.

[0033] The second DB 122 is a database that holds a first feature vector obtained by converting the history stored in the first DB 121 into a first feature vector.

[0034] As shown in FIG. 4, the second DB 122 includes an item "ID", an item "first feature amount vector", and an item "date and time".

[0035] The item “ID” is information for identifying a record. Specifically, the item “ID” may be used as version information of the first feature vector.

[0036] The item “first feature amount vector” is data of the first feature amount vector converted by the conversion unit 134.

[0037] The item "date and time" is the date and time when the first feature amount vector was stored. Specifically, the item "date and time" is the date and time when the conversion unit 134 converted the history into the first feature amount vector.

[0038] The third DB 123 is a database that holds a trained model. As shown in Fig. 5, the third DB 123 includes an item "model ID", an item "model", and an item "training date and time".

[0039] The item "model ID" is information for identifying a model.

[0040] The item "Model" is a trained model. Details will be described later.

[0041] The item “learning date and time” is the date and time when the trained model was trained.

[0042] The fourth DB 124 is a database that holds information about materials of the second user. The fourth DB 124 corresponds to master data for the second user. Therefore, the second user may hold the fourth DB 124.

[0043] As shown in FIG. 6, the fourth DB 124 includes an item "ID", an item "company product number", an item "company product name", an item "material information", and an item "date and time".

[0044] The item "ID" is information for identifying a record.

[0045] The item "company product number" is the product number of the material in question for the second user. There is only one company product number, but one company product number may be linked to multiple product numbers of the first user.

[0046] The item "own product name" is the product name of the material indicated by the own product number in the second user.

[0047] The item "material information" is information related to the material of the record. This information is, for example, the unit of the material, the quantity, the supplier company name, etc.

[0048] The item "date and time" is the date and time when the record was registered.

[0049] The fifth DB 125 is a database that holds information regarding materials that have been confirmed between the second user and the first user.

[0050] As shown in FIG. 7, the fifth DB 125 includes an item "ID", an item "first user", an item "specification ID", an item "first user product number", an item "first user surface name", and an item "date and time".

[0051] The item "ID" is information for identifying a record.

[0052] The item "First User" is the name or title of the first user.

[0053] The item "specification ID" is a specification used with the first user. The specification is, for example, a sewing specification, a materials list, a purchase order, or other document. The data of the specification (not shown) is stored in the storage unit 102 in association with this "specification ID".

[0054] The item "first user product number" is the product number of the first user for the material.

[0055] The item "First user surface name" is the surface name of the specification for the first user.

[0056] The item "date and time" is the date and time when the record was registered.

[0057] The control unit 103 exerts functions shown as various modules by the processor of the server 10 performing processes according to the programs.

[0058] The control unit 103 includes a reception control unit 131, a transmission control unit 132, an input unit 133, a conversion unit 134, an extraction unit 135, a detection unit 136, a display unit 137, a learning unit 138, and the like.

[0059] The reception control unit 131 controls the processing in which the server 10 transmits and receives signals from external devices in accordance with a communication protocol.

[0060] The transmission control unit 132 controls the process in which the server 10 transmits a signal to an external device in accordance with a communication protocol.

[0061] The input unit 133 accepts input of the product number of the first user from the user terminal 20 of the second user.

[0062] Specifically, input unit 133 accepts the input of the first user's part number by receiving a document such as a first specification sheet on which the first user's part number is written from user terminal 20. Input unit 133 accepts the input of the first user's part number contained in, for example, electronic data of the document, data of the document read by OCR, or the like.

[0063] The conversion unit 134 converts the product number data into a feature vector.

[0064] Specifically, the conversion unit 134 converts the product number of the first user that has been input into a second feature vector. The feature vector is, for example, a feature vector used by Word2Vec, N-Gram, etc. By converting the product number into a feature vector, the conversion unit 134 makes it easier to calculate the similarity (distance) between vectors, thereby making it easier to extract product number candidates.

[0065] Furthermore, the conversion unit 134 converts the history of the second user associating the part numbers of the materials of the first user and the part numbers of the materials of the second user, which are stored in the first DB 121, into a first feature vector. At this time, the conversion unit 134 uses the "company part number" and the "first user part number" from the data stored in the first DB 121. The conversion unit 134 can convert the entire first DB 121 into a feature vector as a single corpus. The conversion unit 134 stores the first feature vector converted in advance in the second DB 122.

[0066] The extraction unit 135 uses the input product number of the first user and the history to determine the display mode of candidates for the product number of the second user for the material corresponding to the product number of the first user.

[0067] Specifically, the extraction unit 135 uses the input product number of the first user and the history to extract candidates for the product number of the second user for the material corresponding to the product number of the first user. More specifically, the extraction unit 135 extracts candidates for the product number of the second user by searching the first feature vector using the second feature vector. In the search, the extraction unit 135 searches the first feature vector for a feature vector whose distance from the second feature vector is equal to or less than a predetermined threshold. The distance between the vectors can be calculated using, for example, the Levenshtein distance, the cosine distance, or the like.

[0068] The extraction unit 135 extracts the product number of the second user, which is the company, by re-converting the feature amount vector of the search result. The extracted company's product number is set as a candidate for the product number.

[0069] Furthermore, the extraction unit 135 extracts the number of times that the candidate part number has been selected for the part number of the first user. Next, the extraction unit 135 uses the number of times, etc. to calculate a prediction score indicating the accuracy of the possibility that the candidate part number is the part number of the second user. The extraction unit 135 may calculate the prediction score, for example, so that the prediction score increases in proportion to the number of times. The extraction unit 135 uses the prediction score to determine the display mode of the part number candidates. More specifically, the extraction unit 135 determines the display mode such as displaying the part number candidates in descending order of prediction score, highlighting part number candidates with high prediction scores, and so on.

[0070] Furthermore, the extraction unit 135 may extract candidates for the product number using a trained model that outputs candidates for the product number of the second user when the second feature vector is input. The trained model will be described later.

[0071] Specifically, the extraction unit 135 acquires the latest trained model from the third DB 123. Next, the extraction unit 135 inputs the second feature vector to the trained model to extract the product number of the second user.

[0072] The detection unit 136 extracts the difference between the first specification in which the part number of the first user is described and the second specification, which is a past specification corresponding to the extracted part number candidate.

[0073] Specifically, the detection unit 136 extracts the difference by comparing data for each item between the first specification and the second specification.

[0074] The display unit 137 displays the extracted product number candidates to the second user.

[0075] Specifically, the display unit 137 displays the extracted product number candidates on the user terminal 20 of the second user so that the candidates can be selected. At this time, the display unit 137 may display the number of times the product number candidates have appeared and the prediction score together with the product number candidates. The display unit 137 may also be configured to acquire the company's product name corresponding to the product number candidate from the fourth DB 124 and display the company's product name together with the product number candidate. Therefore, the display unit 137 may also display the first specifications and highlight the difference from the second specifications for each candidate.

[0076] Furthermore, when display unit 137 receives a selection from the second user, it may input a candidate for the selected part number into the first specification. Furthermore, display unit 137 associates the input part number of the first user with the selected part number and stores them in first DB 121. In this way, the more the user inputs, the more data associated with part numbers is stored in first DB 121. Note that first DB 121 may not initially have sufficient data. For this reason, associated performance data may be stored in first DB 121 in advance.

[0077] The learning unit 138 uses the history to learn the model so that when the second feature amount vector is input, candidates for the product number of the second user are output.

[0078] Specifically, the learning unit 138 learns a learned model using a history in which the second user associates the part numbers of materials of the first user with the part numbers of materials of the second user.

[0079] The model may be, for example, a model that expresses a feature vector of a target as a vector, such as Word2Vec, etc. As for the learning method, any learning method may be used depending on the model to be adopted.

[0080] When extraction unit 135 has performed extraction to a certain extent, learning unit 138 learns the model. This enables the second user to extract product numbers specialized for the first user. Learning unit 138 stores the learned model in third DB 123 as a learned model.

[0081] <1.3. Functional configuration of user terminal 20> 8 is a diagram showing the functional configuration of the user terminal 20. As shown in FIG. 8, the user terminal 20 fulfills the functions of a communication unit 201, a storage unit 202, and a control unit 203.

[0082] The communication unit 201 performs processing for communicating with an external device.

[0083] The storage unit 202 stores data and programs to be used.

[0084] The control unit 203 includes a reception control unit 231, a transmission control unit 232, an input unit 233, a display unit 234, and the like.

[0085] The reception control unit 231 controls the processing in which the user terminal 20 transmits and receives signals from external devices in accordance with a communication protocol.

[0086] The transmission control unit 232 controls the process in which the user terminal 20 transmits a signal to an external device in accordance with a communication protocol.

[0087] The input unit 233 receives input of a predetermined operation from the second user.

[0088] Specifically, input unit 233 accepts an input of a first specification from the user, and also accepts a selection of the displayed part number candidates.

[0089] The display unit 234 displays the product number candidates received from the server 10.

[0090] <1.4. System 1 processing flow> Next, the process flow of the system 1 will be described.

[0091] 9 and 10 are diagrams showing an example of a processing flow in the system 1. The server 10 presents candidates for part numbers of the second user, which are the company's part numbers of materials corresponding to the part number of the first user, based on the part number of the first user, through the processing shown in FIG. In step S711, the user terminal 20 acquires a purchase order including a first product number from the first user. Specifically, the user terminal 20 of the second user acquires the purchase order as electronic data from the user terminal 20 of the first user. If the purchase order is not electronic data, the second user digitizes the purchase order using OCR or the like, and the user terminal 20 acquires the purchase order.

[0092] In step S713, the user terminal 20 transmits the purchase order to the server 10.

[0093] In step S715, the server 10 receives the purchase order from the user terminal 20.

[0094] In step S717, the server 10 extracts the product number of the first user from the purchase order.

[0095] In step S719, the server 10 converts the product number of the first user into a second feature vector. Specifically, the conversion unit 134 converts the product number of the first user into the second feature vector.

[0096] In step S721, the server 10 acquires the first feature amount vector from the second DB 122. Specifically, the extraction unit 135 acquires from the second DB 122 the first feature amount vector that has been converted in advance by the conversion unit .

[0097] In step S723, the server 10 uses the first feature vector and the second feature vector to extract candidates for the product number of the second user. Specifically, the extraction unit 135 searches the first feature vector using the second feature vector to extract candidates for the product number of the second user. A plurality of candidates may be extracted.

[0098] In step S725, server 10 acquires the number of conversions and purchase orders corresponding to the candidate part numbers. Specifically, extraction unit 135 counts the number of records of the part number of the first user by referring to first DB 121 for the input part number of the first user. Extraction unit 135 also acquires a first specification corresponding to the part number of the first user from fifth DB 125. When there are multiple first specifications, extraction unit 135 can be configured to acquire the latest one.

[0099] In step S727, the server 10 calculates the score of the candidate part number from the number of conversions. Specifically, the extraction unit 135 uses the number of conversions to calculate a prediction score indicating the accuracy of the candidate part number. When there are multiple candidates for the part number, the extraction unit 135 calculates a prediction score for each of the candidates for the part number.

[0100] In step S729, the server 10 detects the difference between the purchase order and the corresponding purchase order.

[0101] In step S731, the server 10 displays on the user terminal 20 the product number candidates, the number of times, and the predicted score.

[0102] In step S733, the user terminal 20 displays the product number candidates, the number of times, and the predicted score.

[0103] Fig. 13 is an example of a screen displayed on the user terminal 20. As shown in Fig. 13, a purchase order is displayed on a screen 1000. Fig. 13 also includes a product number candidate screen 1010. The product number candidate screen 1010 displays product number candidates, the number of times, and the prediction score.

[0104] In step S735, the user terminal 20 selects or inputs the correct product number.

[0105] In step S737, the server 10 inputs the selected part number into the purchase order, and displays the difference from the first specifications.

[0106] In step S739, the user terminal 20 inputs the selected part number into the order sheet, and displays the difference from the first specifications.

[0107] Fig. 14 is an example of a screen displayed on the user terminal 20. As shown in Fig. 14, a purchase order is displayed on a screen 1100. Fig. 14 also includes a difference highlighting icon 1030 and a company part number column 1050. The difference highlighting icon is displayed so as to highlight the difference from the past specifications. A selected part number is input into the company part number column 1050.

[0108] In step S741, the server 10 inputs to the first DB 121 the input product number of the first user and the product number of the selected second user.

[0109] Fig. 11 is a diagram showing an example of a processing flow in the system 1. The server 10 learns a model by the processing shown in Fig. 11 so as to output product number candidates for the second user when the second feature vector is input. In step S811, the server 10 acquires a history from the first DB 121 in which results of extraction performed to a certain extent are stored.

[0110] By performing extraction to a certain extent, a history that automatically associates the part number of the first user with the part number of the second user is accumulated in the first DB 121. Therefore, a correspondence history specialized for the relationship between the first user and the second user is stored. The learning unit 138 uses this history to learn the learned model.

[0111] In step S813, the server 10 converts the history into a first feature amount vector. Specifically, the conversion unit 134 converts the history acquired in step S811 into a first feature amount vector.

[0112] In step S815, the server 10 uses the first feature vector to train a model. Specifically, the learning unit 138 trains the model so that, when the second feature vector is input, the product number of the second user is extracted.

[0113] In step S817, the server 10 stores the first model in the third DB 123 as a trained model.

[0114] Fig. 12 is a diagram showing an example of a processing flow in the system 1. Through the processing shown in Fig. 12, the server 10 presents candidates for part numbers of the second user, which are the company's part numbers of materials corresponding to the part number of the first user. In step S921, the server 10 obtains the trained model from the third DB123.

[0115] In step S923, the server 10 inputs the second feature vector into the trained model to extract candidates for the product number of the second user. In step S831, the server 10 displays on the user terminal 20 the product number candidates, the number of orders, and the difference between past orders.

[0116] <1.5.Summary> As described above, the technology of the present disclosure is a system for managing part numbers of materials used in transactions between a first user and a second user, where the part numbers managed by the first user and the second user are not necessarily the same even for the same material. The technology of the present disclosure is configured such that a storage unit stores a history in which the second user associates part numbers of materials in the first user with part numbers of materials in the second user. The technology of the present disclosure accepts an input of a part number of the first user, and uses the input part number of the first user and the history to extract part number candidates of the second user for materials corresponding to the input part number of the first user. The technology of the present disclosure displays the extracted part number candidates to the second user. As a result, the technology of the present disclosure can contribute to the rapid determination of the company's part number corresponding to the other company's part number. Furthermore, by using the technology of the present disclosure, the second user can easily turn an incomplete BOM into a complete BOM without having to be aware of the brand of the trading partner. Furthermore, the master data required for order placement and receipt can be maintained through the daily work of users selecting product numbers without the need for each user to be aware of constantly updating the master data.

[0117] In addition, the technology disclosed herein matches the part numbers of the first user with the part numbers of the second user, converts the track record and the input part number of the first user into a feature vector, and searches for the part number of the second user. As a result, even if there are part numbers that do not match in the history, it is possible to extract candidates for the part number of the second user by searching using similarity based on the distance between vectors in the search space. In addition, the technology disclosed herein extracts candidates for the second user's part number using a trained model that outputs candidates for the second user's part number when a second feature vector is input, and the second feature vector obtained by converting the input part number of the first user. In the technology disclosed herein, the trained model is a model that has been trained using history to output candidates for the second user's part number when a second feature vector is input. As a result, the model undergoes reinforcement learning as actual correspondence records are accumulated, improving the accuracy of extracting candidate part numbers. As a result, the technology disclosed herein can contribute to the rapid determination of a company's part number that corresponds to a company's part number.

[0118] <2. Modifications> Although several embodiments of the present disclosure have been described above, these embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and modifications are within the scope of the invention and its equivalents as described in the claims, as well as the scope and spirit of the invention.

[0119] In the above disclosure, a purchase order has been described as an example, but the present disclosure is not limited thereto. The purchase order is merely an example for assuming a case in which the part number of the second user corresponding to the part number of the first user is unknown at the stage when the second user receives an order. The technology disclosed herein can be applied to any document in which the part number of the second user corresponding to the part number of the first user is unknown.

[0120] In addition, in the above disclosure, the number of times is used to calculate the prediction score, but the present invention is not limited to this. Various information can be used for the prediction score. For example, the prediction score can be calculated using a first feature vector, a second feature vector, etc., to calculate a score based on past history. This makes it easier for the second user who sees the prediction score to specify the product number to be used.

[0121] In addition, the technology disclosed above is configured to automatically extract the first user's part number from the specifications, but is not limited to this. For example, it may be configured to accept input of the first user's part number, the first user's face name, etc. as search words from the second user and extract part number candidates. In this case, it is sufficient to add an item "search word" to the first DB 121 and store the associated results.

[0122] In the above disclosure, a process for extracting part number candidates immediately after a specification is input has been described, but the present invention is not limited to this. Prior to the above process, the extraction unit 135 may be configured to search whether a first user part number matching the search word is stored in the first DB 121. If a first user part number matching the search word exists, the search result may be displayed and the process may end. Also, if a first user part number matching the search word does not exist, the above process may be executed.

[0123] In the above disclosure, the distance between feature vectors is used to extract part number candidates, but the present invention is not limited to this. For example, the extraction unit 135 may extract part number candidates using the probability of simultaneous existence of each part number in one specification. In this case, the extraction unit 135 extracts part number candidates that are likely to exist simultaneously with multiple input part numbers. At this time, the conversion unit 134 may treat the part number as one word, the probability of simultaneous existence of the word in one specification as the first feature vector, and multiple part numbers as the second feature vector. In this way, the technology of this modified example can extract part numbers that are likely to appear next (are likely to be missing) from the input part number group.

[0124] <Additional Notes> The matters described in the above embodiments will be supplemented below.

[0125] <Appendix 1> A program for operating a computer having a processor and a memory unit in a system that manages product numbers of materials used in transactions between a first user and a second user, wherein even for the same material, the product number managed by the first user and the product number managed by the second user are not necessarily the same, and the memory unit is configured to store a history that associates the product numbers of the material at the first user with the product numbers of the material at the second user, the program executing the steps of accepting input of a product number of the first user, determining a display mode of candidate product numbers of the second user for the material corresponding to the input product number of the first user using the input product number of the first user and the history, and displaying the candidate product numbers to the second user in the determined display mode.

[0126] <Appendix 2> The program described in <Appendix 1>, in the determining step, calculates a prediction score indicating the correctness of the part number for the second user using the part number of the first user and the history, and determines a display mode of the candidate part number according to the prediction score.

[0127] <Appendix 3> The program described in <Appendix 2>, wherein in the displaying step, the product number candidates are displayed to the second user together with the prediction score.

[0128] <Appendix 4> The program described in <Appendix 2> executes a step of extracting the number of times the second user's product number was selected for the first user's product number based on the history, and in the display step, displays the number of times the product number candidates were selected together with the product number candidates to the second user.

[0129] <Appendix 5> The program described in <Appendix 4>, wherein in the determining step, a part number of the second user having a high frequency is extracted as a candidate for the part number of the second user for the material.

[0130] <Appendix 6> The program described in <Appendix 1>, which executes a step of converting the history into a first feature vector and a step of converting the input product number of the first user into a second feature vector, and in the determining step, extracts product number candidates for the second user by searching the first feature vector with the second feature vector.

[0131] <Appendix 7> The program described in <Appendix 6>, wherein in the determining step, candidates for the second user's product number are extracted using a trained model that outputs candidates for the second user's product number when the second feature vector is input, and the second feature vector obtained by converting the input product number of the first user, and the trained model is a model trained using the history to output candidates for the second user's product number when the second feature vector is input.

[0132] <Appendix 8> The program described in <Appendix 1> includes a step of accepting a selection of the candidate part number from the second user, and determining the candidate part number for which the selection has been accepted as the part number corresponding to the part number input by the first user.

[0133] <Appendix 9> The program described in <Appendix 8>, which executes a step of inputting the determined product number into a document used by the second user.

[0134] <Appendix 10> The program described in <Appendix 1> includes a step of accepting input of a first specification containing the first user's product number from the first user, a step of extracting a second specification which is a past specification corresponding to the candidate product number extracted in the determining step, a step of extracting a difference between the first specification and the second specification, and a step of displaying the second specification containing the candidate product number to the second user so that the difference from the first specification can be seen.

[0135] <Appendix 11> In a system that manages the product numbers of materials used in transactions between a first user and a second user, the information processing device has a processor and a memory unit, and even for the same material, the product number managed by the first user and the product number managed by the second user are not necessarily the same, and the memory unit is configured to store a history that associates the product number of the material at the first user with the product number of the material at the second user, and the information processing device executes the steps of: accepting input of a product number of the first user; using the input product number of the first user and the history, determining a display mode of candidate product numbers of the second user for the material corresponding to the input product number of the first user; and displaying the extracted product number candidates to the second user.

[0136] <Appendix 12> A method for operating a computer having a processor and a memory unit in a system that manages product numbers of materials used in transactions between a first user and a second user, wherein even for the same material, the product number managed by the first user and the product number managed by the second user are not necessarily the same, and the memory unit is configured to store a history that associates the product numbers of the material at the first user with the product numbers of the material at the second user, and the method executes the steps of: accepting input of a product number of the first user; determining a display mode of candidate product numbers of the second user for the material corresponding to the input product number of the first user using the input product number of the first user and the history; and displaying the extracted product number candidates to the second user. [Explanation of symbols]

[0137] 1 system, 10 server, 12 communication IF, 13 input / output IF, 15 memory, 16 storage, 19 processor, 20 user terminal, 22 communication IF, 23 input / output IF, 25 memory, 26 storage, 29 processor, 80 network, 101 communication unit, 102 memory unit, 103 control unit, 1100 screen, 121 first DB, 122 second DB, 123 third DB, 124 fourth DB, 125 fifth DB, 131 receiving control unit, 132 transmitting control unit, 133 input unit, 134 conversion unit, 135 extraction unit, 136 detection unit, 137 display unit, 138 learning unit, 1000 screen, 1030 difference emphasis icon, 1050 company product number column.

Claims

1. A program for operating a computer having a processor and a storage unit in a system for managing part numbers of materials used in a transaction between a first user and a second user, comprising: Even if the material is the same, the part number managed by the first user and the part number managed by the second user are not necessarily the same, the storage unit is configured to store a history in which a part number of the material of the first user is associated with a part number of the material of the second user, receiving an input of a product number of the first user; determining a display mode of a candidate part number of the second user for a material corresponding to the input part number of the first user, using the input part number of the first user and the history; displaying the product number candidates to the second user in the determined display mode; A program that, in the determining step, calculates a prediction score indicating the accuracy of the product number for the second user using the product number of the first user and the history, and determines the display mode of the candidate product number according to the prediction score.

2. In the displaying step, the product number candidates are displayed to the second user together with the prediction scores. The program according to claim 1.

3. extracting the number of times the second user's product number was selected relative to the first user's product number based on the history; Run the command, 2. The program according to claim 1, wherein in the displaying step, the number of times each of the part number candidates is displayed to the second user together with the part number candidates.

4. 4. The program according to claim 3, wherein in the determining step, a part number of the second user having a large number of times is extracted as a candidate for a part number of the second user for the material.

5. A program for operating a computer having a processor and a memory unit in a system for managing part numbers of materials used in a transaction between a first user and a second user, comprising: Even if the material is the same, the part number managed by the first user and the part number managed by the second user are not necessarily the same, the storage unit is configured to store a history in which a part number of the material of the first user is associated with a part number of the material of the second user, receiving an input of a product number of the first user; converting the history into a first feature vector; converting the input product number of the first user into a second feature vector; determining a display mode of a candidate part number of the second user for a material corresponding to the input part number of the first user, using the input part number of the first user and the history; displaying the product number candidates to the second user in the determined display mode; a program for extracting candidates for part numbers for the second user by searching the first feature vector using the second feature vector in the determining step.

6. In the determining step, a candidate product number of the second user is extracted using a trained model that outputs a candidate product number of the second user when the second feature vector is input and the second feature vector obtained by converting the input product number of the first user; The program according to claim 5 , wherein the trained model is a model trained to output candidates for product numbers of the second user when the second feature vector is input using the history.

7. receiving a selection of the product number candidates from the second user; The program according to claim 1 , wherein the product number candidate for which the selection has been accepted is determined to be a product number corresponding to the input product number of the first user.

8. 8. The program according to claim 7, further comprising a step of inputting the determined product number into a document used by the second user.

9. A program for operating a computer having a processor and a memory unit in a system for managing part numbers of materials used in a transaction between a first user and a second user, comprising: Even if the material is the same, the part number managed by the first user and the part number managed by the second user are not necessarily the same, the storage unit is configured to store a history in which a part number of the material of the first user is associated with a part number of the material of the second user, receiving an input of a product number of the first user; determining a display mode of a candidate part number of the second user for a material corresponding to the input part number of the first user, using the input part number of the first user and the history; displaying the product number candidates to the second user in the determined display manner; accepting an input of a first specification sheet describing a product number of the first user from the first user; extracting a second specification which is a past specification corresponding to the part number candidate extracted in the determining step; extracting a difference between the first specification and the second specification; A program that executes, in the displaying step, displaying the second specification sheet on which the part number candidates are described to the second user so that the difference from the first specification sheet can be seen.

10. In a system for managing part numbers of materials used in a transaction between a first user and a second user, an information processing device is provided that includes a processor and a storage unit, Even if the material is the same, the part number managed by the first user and the part number managed by the second user are not necessarily the same, the storage unit is configured to store a history in which a part number of the material of the first user is associated with a part number of the material of the second user, receiving an input of a product number of the first user; determining a display mode of a candidate part number of the second user for a material corresponding to the input part number of the first user, using the input part number of the first user and the history; displaying the extracted part number candidates to the second user; In the determining step, an information processing device calculates a prediction score indicating the correctness of the product number for the second user using the product number of the first user and the history, and determines the display mode of the candidate product number based on the prediction score.

11. A method for operating a computer having a processor and a storage unit in a system for managing part numbers of materials used in a transaction between a first user and a second user, the method comprising: Even if the material is the same, the part number managed by the first user and the part number managed by the second user are not necessarily the same, the storage unit is configured to store a history in which a part number of the material of the first user is associated with a part number of the material of the second user, receiving an input of a product number of the first user; determining a display mode of a candidate part number of the second user for a material corresponding to the input part number of the first user, using the input part number of the first user and the history; displaying the extracted part number candidates to the second user; A method in which, in the determining step, a prediction score indicating the correctness of the product number for the second user is calculated using the product number of the first user and the history, and a display mode of the candidate product number is determined based on the prediction score.

Images