Information Processing Apparatus, Information Processing Method, and Program

Abstract

Make it possible to easily identify products expressed in a format different from the product names registered in the database in advance. 【Solution means】 An information processing apparatus 1 includes an acquisition unit 131 that acquires text information, an element identification unit 132 that identifies a product identification element for identifying a product that is the subject of an inquiry in the text information, a conversion unit 133 that generates a group of converted elements, which are character strings obtained by converting each character string constituting the product identification element into a character string that matches the format registered in the product database, a query generation unit 134 that generates a search query for searching a product database in which a reference product identification element, which is an element for identifying a product, and a product code are associated based on the group of converted elements, and an output unit 135 that outputs a product code associated with a reference product identification element corresponding to the group of converted elements included in the search query in the product database.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program.

Background Art

[0002] A system for searching for products that match the product names included in text data is known (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In exchanges such as emails, products may be specified in a form different from the actual product name. In such cases, in the prior art, it has been impossible to identify products expressed in a form different from the product names registered in the database in advance, and there has been a problem that it takes time for the person in charge who received the product specification to identify the product.

[0005] Therefore, the present invention has been made in view of these points, and an object thereof is to make it possible to easily identify products expressed in a form different from the product names registered in the database in advance.

Means for Solving the Problems

[0006] In the information processing apparatus according to the first aspect of the present invention, an acquisition unit that acquires text information, an element identification unit that identifies a group of product identification elements for identifying a product that is the subject of an inquiry in the text information, and a conversion unit that generates a group of converted elements, which is a character string obtained by converting each character string constituting the group of product identification elements so as to match the format registered in the product database, and a query generation unit that generates a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, and a product code are associated based on the group of converted elements, and an output unit that outputs a product code associated with the reference product identification element group corresponding to the group of converted elements in the product database.

[0007] The reference product identification element group includes information indicating the attributes of the product, and the query generation unit may identify the element attributes of the group of converted elements and generate the search query for searching for the attributes of the product corresponding to the element attributes and the group of converted elements and the product code associated with the reference product identification element group.

[0008] The conversion unit may generate the group of converted elements by inputting an output instruction indicating that the conversion unit outputs each character string constituting the group of product identification elements and the group of converted elements corresponding to the character string to a language model and causing the language model to output the group of converted elements corresponding to the character string.

[0009] The conversion unit may determine whether the group of product identification elements matches the format registered in the product database, and generate the group of converted elements when it is determined that they do not match.

[0010] The text information includes customer identification information that is information for identifying a customer who makes an inquiry about a product, and when there are a plurality of product codes associated with the reference product identification element group corresponding to the group of converted elements in the product database, the output unit may output a product code selected from the plurality of product codes based on the attributes of the customer indicated by the customer identification information.

[0011] The acquisition unit acquires customer attribute information indicating the attributes of the customer indicated by the customer identification information. The information processing apparatus further includes a model storage unit that stores a learned model using, as teacher data, learned customer attribute information indicating the attributes of the customer for learning, a plurality of learning product codes, and the correct product code among the plurality of learning product codes. When there are a plurality of product codes associated with the reference product identification element group corresponding to the converted element group in the product database, the output unit may input the customer attribute information and the plurality of product codes into the learned model, and select the product code output by the learned model.

[0012] The information processing apparatus further includes a rule information storage unit that stores rule information indicating rules for selecting products to be presented from among the products registered in the product database. When there are a plurality of product codes associated with the reference product identification element group corresponding to the converted element group in the product database, the output unit may output the product code selected from the plurality of product codes based on the rules indicated by the rule information.

[0013] The rule information is described in natural language. The information processing apparatus further includes a model storage unit that stores a learned model using, as teacher data, the rule information, a plurality of learning product codes, one or more reference product identification element groups associated with each of the plurality of learning product codes, and the correct product code. When there are a plurality of product codes associated with the reference product identification element group corresponding to the converted element group in the product database, the output unit may input the rule information and the plurality of product codes into the learned model, and output the product code output by the learned model.

[0014] When there are a plurality of product codes associated with the reference product identification element group corresponding to the converted element group in the product database, the information processing apparatus may further include a question generation unit that generates a question for specifying which product code to select from among the plurality of product codes.

[0015] It further has a determination unit that analyzes the meaning of the text included in the text information and determines whether the analyzed text information includes the content of an inquiry regarding a product. When the determination unit determines that the content of an inquiry regarding a product is included, the element identification unit may identify the product specific element group.

[0016] It may further have a learning unit that generates a collating model, which is a learned model that learns, as teacher data, the character strings constituting each of the product specific element groups and the converted element groups generated by the conversion unit for the character strings, and outputs the converted element groups when the character strings constituting each of the product specific element groups are input.

[0017] In the information processing apparatus according to the second aspect of the present invention, an acquisition unit that acquires text information, an element identification unit that identifies a product specific element group for identifying a product that is the subject of an inquiry in the text information, a conversion unit that generates a converted element group, which is a character string obtained by converting the character strings constituting each of the product specific element groups so as to match the format registered in the product database, a query generation unit that generates a search query for searching a product database in which a reference product specific element group, which is an element for identifying a product, and a product code are associated with the converted element group, a first output unit that outputs a plurality of product codes associated with the reference product specific element groups corresponding to the converted element groups included in the search query in the product database, a rule information storage unit that stores rule information described in natural language for a rule for selecting a product to be presented from among the products registered in the product database, a model storage unit that stores a language model that outputs an answer to the output instruction when the output instruction is input, and a second output unit that outputs the product codes output by the first output unit, includes the rule information stored by the rule information storage unit, and outputs the product codes resulting from inputting an output instruction to the effect of outputting the product codes conforming to the rule information among the plurality of product codes to the language model.

[0018] It may further include a learning unit that generates a learned model by learning, as teacher data, a plurality of product codes output by the first output unit, rule information stored in the rule information storage unit, and product codes output by the second output unit.

[0019] In the information processing method according to the third aspect of the present invention, the computer executes steps of: acquiring text information; identifying a group of product identification elements for identifying a product that is the subject of an inquiry in the text information; generating a group of converted elements, which are strings obtained by converting the strings constituting each of the group of product identification elements so as to match the format registered in a product database; generating a search query for searching a product database that associates a reference product identification element group, which is an element for identifying a product, with a product code, based on the group of converted elements; and outputting a product code associated with the reference product identification element group corresponding to the group of converted elements included in the search query in the product database.

[0020] In the program according to the fourth aspect of the present invention, the computer is caused to execute steps of: acquiring text information; identifying a group of product identification elements for identifying a product that is the subject of an inquiry in the text information; generating a group of converted elements, which are strings obtained by converting the strings constituting each of the group of product identification elements so as to match the format registered in a product database; generating a search query for searching a product database that associates a reference product identification element group, which is an element for identifying a product, with a product code, based on the group of converted elements; and outputting a product code associated with the reference product identification element group corresponding to the group of converted elements included in the search query in the product database.

[0021] In the information processing method according to the fifth aspect of the present invention, a step of acquiring text information executed by a computer, a step of identifying a group of product identification elements for identifying a product targeted for inquiry in the text information, and a step of generating a converted element group, which is a character string obtained by converting each character string constituting the group of product identification elements into a character string that matches the format registered in a product database, a step of generating a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, and a product code are associated based on the converted element group, a first output step of outputting a plurality of product codes associated with the reference product identification element group corresponding to the converted element group included in the search query in the product database, referring to a rule information storage unit that stores rule information describing in natural language a rule for selecting a product to be presented from among the products registered in the product database, and including the plurality of product codes output in the first output step and the rule information stored in the rule information storage unit, and inputting an output instruction indicating that a product code conforming to the rule information among the plurality of product codes is to be output to a language model that outputs an answer to the output instruction, and a second output step of outputting a product code resulting from the input and output as a result.

[0022] In the program according to the sixth aspect of the present invention, the computer is caused to execute steps of: acquiring text information; identifying a group of product identification elements for identifying a product that is the subject of an inquiry in the text information; generating a group of converted elements, which are character strings obtained by converting each character string constituting the group of product identification elements into a character string that matches the format registered in a product database; generating a search query for searching a product database that associates a reference product identification element group, which is an element for identifying a product, with a product code, based on the group of converted elements; a first output step of outputting a plurality of product codes associated with the reference product identification element group corresponding to the group of converted elements included in the search query in the product database; and a second output step of referring to a rule information storage unit that stores rule information describing, in natural language, a rule for selecting a product to be presented from among the products registered in the product database, inputting an output instruction to output a product code that conforms to the rule information among the plurality of product codes output in the first output step and the rule information stored in the rule information storage unit to a language model that outputs a response to the output instruction, and outputting the product code resulting from the input.

Advantages of the Invention

[0023] According to the present invention, it is possible to easily identify a product expressed in a format different from the product name registered in the database in advance.

Brief Description of the Drawings

[0024]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0025] [Overview of Information Processing System S] FIG. 1 and FIG. 2 are diagrams for explaining the overview of the information processing system S according to the embodiment. The information processing system S is a system for supporting business activities. More specifically, the information processing system S is a system for identifying a product that is the subject of an inquiry when receiving an inquiry from a customer U1 and providing information about the identified product to the customer U1. In the following, unless otherwise specified, "product" shall include services. The information processing system S includes an information processing apparatus 1, an information terminal 2, and a product database.

[0026] The information processing apparatus 1 is an apparatus for identifying a product that is the subject of an inquiry received from the customer U1. More specifically, the information processing apparatus 1 acquires text information including the content of the inquiry about the product, and identifies the product that is the subject of the inquiry in the text information. For example, the information processing apparatus 1 acquires text information including the content of requesting an estimate for the product. The information processing apparatus 1 may further generate an answer to the content for which the identified product has been inquired. The customer U1 is a customer who purchases, etc., products provided by the business operator to which the user U2 belongs.

[0027] The information terminal 2 is a terminal used by the user U2. The information terminal 2 is, for example, a smartphone, a tablet, or a personal computer. The user U2 is, as an example, a person in charge of responding to inquiries from the customer U1 or a person in charge of conducting sales or marketing to the customer U1.

[0028] The product database is a database for searching products. Although details will be described later, for the product database, elements for specifying products and product codes are associated. The product database receives a search query for searching products and outputs a product code corresponding to the content of the search query. In FIG. 1, an example in which the product database and the information processing apparatus 1 are configured separately is shown, but the product database may be included in the information processing apparatus 1.

[0029] The processing of the information processing system S will be described with reference to FIG. 2. The information processing apparatus 1 acquires text information. The text information is, as an example, an email sent from the customer U1. The text information includes a product specific element group. The product specific element group includes one or more product specific elements. The product specific element is information for specifying a product that is the subject of an inquiry in the text information. The product specific element is, as an example, information indicating a product name, a seller of the product, a manufacturer, a type of the product, a usage period, etc. Note that the information processing apparatus 1 may acquire the text information from an external mail server (not shown) or may acquire the text information from the information terminal 2.

[0030] The information processing apparatus 1 specifies the product specific element group included in the text information. As an example, the information processing apparatus 1 inputs the text information into an extraction model and causes the extraction model to output the product specific element group included in the text information. The extraction model is a learned model that is learned to take the text information as an input and output the product specific element group included in the input text information.

[0031] The information processing apparatus 1 generates a converted element group based on a product specific element group. The converted element group is a character string obtained by converting the character string constituting the product specific element group so as to match the format registered in the product database (hereinafter referred to as the "predetermined format"). As an example, the information processing apparatus 1 inputs the product specific element group into a collation language model and outputs the converted element group.

[0032] The information processing apparatus 1 generates a search query based on the converted element group. The search query is a query for searching the product database. Specifically, the information processing apparatus 1 identifies an element attribute corresponding to each of the converted elements constituting the converted element group. The element attribute is an attribute that the product specific element has. The information processing apparatus 1 generates a search query based on the converted element group and the element attribute. As an example, when the converted element group is "M Soft" and "OfficeXXX", the information processing apparatus 1 identifies that the element attributes of "M Soft" and "OfficeXXX" are the vendor and the product name, respectively, and the information processing apparatus 1 generates "Vendor:M Soft, Product name:OfficeXXX" as the search query.

[0033] The information processing apparatus 1 searches the product database based on the search query and acquires a product code that matches the condition indicated by the search query. The information processing apparatus 1 causes the acquired product code to be displayed on the information terminal 2.

[0034] By configuring the information processing system S to acquire text information, convert the product specific element group extracted from the acquired text information into a predetermined format, and generate a search query for searching for a product code associated with the converted element group, it becomes possible to identify a product to be inquired about based on the information contained in the text information, and it is possible to easily identify a product expressed in a format different from the product name registered in the database in advance.

[0035] [Configuration of Information Processing Apparatus 1] FIG. 3 is a block diagram showing the configuration of the information processing apparatus 1. The information processing apparatus 1 includes a communication unit 11, a storage unit 12, and a control unit 13. The control unit 13 includes an acquisition unit 131, an element identification unit 132, a conversion unit 133, a query generation unit 134, an output unit 135, a question generation unit 136, a determination unit 137, and a learning unit 138.

[0036] The communication unit 11 is a communication interface for transmitting and receiving data to and from other devices via a network. The storage unit 12 is a storage medium including a ROM (Read Only Memory), a RAM (Random Access Memory), an SSD (Solid State Drive), a hard disk drive, and the like. The storage unit 12 stores in advance a program executed by the control unit 13. The storage unit 12 functions as a model storage unit and a rule information storage unit.

[0037] The storage unit 12 stores an extraction model. The extraction model is a large language model and has learned learning text information and a product identification element group included in the learning text information as teacher data.

[0038] The storage unit 12 stores a clustering language model. The clustering language model is a large language model. As an example, the clustering language model may learn an output instruction including a learning product identification element group and a converted element group obtained by converting the product identification element group as teacher data. The output instruction includes content indicating conversion of the input product identification element group into a predetermined format.

[0039] The control unit 13 is a processor such as a CPU (Central Processing Unit). The control unit 13 functions as an acquisition unit 131, an element identification unit 132, a conversion unit 133, a query generation unit 134, an output unit 135, a question generation unit 136, a determination unit 137, and a learning unit 138 by executing a program stored in the storage unit 12.

[0040] The acquisition unit 131 acquires text information. The acquisition unit 131 may acquire text information from the information terminal 2 or may acquire text information from an external device.

[0041] The element identification unit 132 identifies a group of product identification elements based on the text information acquired by the acquisition unit 131. The element identification unit 132 inputs the text information into an extraction model and outputs a group of product identification elements.

[0042] The conversion unit 133 generates a group of post-conversion elements based on the group of product identification elements. The conversion unit 133 inputs an output instruction described in natural language into a collating language model, and outputs a group of post-conversion elements corresponding to the product identification elements included in the input instruction to the collating language model, thereby generating post-conversion elements. The output instruction includes a character string constituting each of the groups of product identification elements and an instruction to output a group of post-conversion elements corresponding to the character string.

[0043] The query generation unit 134 generates a search query for searching a product database based on the group of post-conversion elements. FIG. 4 is a diagram showing an example of the process of the query generation unit 134. An example of the process of the query generation unit will be described with reference to FIG. 4. As an example, the storage unit 12 stores a template T configured to generate a search query by applying a group of product identification elements, and the query generation unit 134 generates a search query by applying the group of post-conversion elements to the template T.

[0044] The query generation unit 134 identifies an element attribute group of the group of post-conversion elements. As an example, the storage unit 12 stores an element identification model learned to output the element attribute of each of the post-conversion elements constituting the input group of post-conversion elements with the group of post-conversion elements as an input. The query generation unit 134 inputs the group of post-conversion elements into the element identification model and identifies the element attribute of each of the post-conversion elements constituting the group of post-conversion elements. In the example of FIG. 4, the query generation unit 134 identifies the element attributes of "vendor", "product name", and "type" for each of the groups of post-conversion elements of "M Software", "OfficeXXX", and "Enterprise".

[0045] The query generation unit 134 generates a search query for retrieving the product code associated with the identified element attributes and the element attributes and reference product identification element group corresponding to the product identification element group. The query generation unit 134 applies the identified conversion element group and element attributes to the template T to generate a search query.

[0046] In the product database, the reference product identification element group and the product code are associated. The reference product identification element group is a reference product identification element group. The reference product identification element group includes, as an example, information having element attributes such as product name, vendor, license type, license period, version, etc. FIG. 5 is a diagram showing an example of the data structure of the product database. In the product database shown in FIG. 5, the product name, vendor, license type, and license period are associated with the product code as reference product identification elements.

[0047] The output unit 135 searches the product database based on the search query and identifies the product code corresponding to the reference product identification element group corresponding to the converted element group. The output unit 135 outputs the product code associated with the reference product identification element group corresponding to the converted element included in the search query in the product database. The output unit 135 searches the product database based on the search query and acquires the product code corresponding to the converted element group. The output unit 135 outputs the acquired product code to the information terminal 2. Specifically, the output unit 135 may cause the information terminal 2 to display a screen for displaying the acquired product code by transmitting the acquired product code to the information terminal 2.

[0048] By configuring the information processing apparatus 1 to convert the product identification element group extracted from the text information into a predetermined format and generate a search query for retrieving the product code associated with the converted element group, it is possible to easily identify a product expressed in a format different from the product name registered in the database in advance.

[0049] Note that the number of elements in the product identification element group and the converted element group does not have to correspond one-to-one. That is, the conversion unit 133 may generate a converted element group including complemented elements based on the product identification element group. In this case, the alignment language model learns teacher data associating the product identification element group for learning and the converted element group including the complemented product identification elements. As an example, the alignment language model may be learned to output "M Software", "OfficeXXX", and "Enterprise" for the input product identification element groups "M Software" and "OfficeXXX".

[0050] When there is insufficient information for identifying a product code in the text information, a plurality of product candidates may be identified. In such a case, the information processing apparatus 1 may be configured to identify a product based on the attributes of the customer U1.

[0051] In order for the information processing apparatus 1 to be able to identify a product based on the attributes of the customer U1, the text information may include customer identification information. The customer identification information is information for identifying the customer U1 who makes an inquiry about a product. As an example, the customer identification information is the email address of the customer U1, the name of the customer U1, or the user ID of the customer U1. The element identification unit 132 identifies the customer identification information included in the text information. As an example, the extraction model may be configured to output the customer identification information included in the text information.

[0052] The acquisition unit 131 acquires customer attribute information indicating the attributes of the customer U1 indicated by the customer identification information. As an example, the storage unit 12 stores profile information associating the customer identification information and the customer attribute information. FIG. 6 is a diagram showing an example of the data structure of the profile information stored in the storage unit 12. The acquisition unit 131 acquires the customer attribute information corresponding to the customer identification information identified by the element identification unit 132 in the profile information. In the profile information shown in FIG. 6, the customer attribute information includes the number of employees and the industry type of the customer.

[0053] When there are multiple product codes associated with the reference product identification element group corresponding to the converted element in the product database, the output unit 135 outputs the product code selected from the multiple product codes based on the attributes of customer U1 indicated by the customer identification information.

[0054] As an example, the information processing apparatus 1 selects a product code using a learned model that has been learned to output a product code to be presented to customer U1 based on the attributes of customer U1. The storage unit 12 stores the learned model learned using, as teacher data, the learning customer attribute information indicating the attributes of customer U1 for learning, a plurality of product codes for learning, and the correct product code among the plurality of product codes for learning. When the learned model inputs the customer attribute information and a plurality of product codes, it outputs a product code.

[0055] When there are multiple product codes associated with the reference product identification element group corresponding to the converted element group in the product database, the output unit 135 inputs the customer attribute information and the plurality of product codes into the learned model, and causes the information terminal 2 to display a screen for displaying the product code output by the learned model.

[0056] By configuring the output unit 135 to select a product code based on the attributes of customer U1, there is an effect that it becomes possible to select a product that conforms to the attributes of customer U1 who has made an inquiry about the product.

[0057] There may be a case where rules for determining the product to be proposed to customer U1 are defined in advance. In such a case, the information processing apparatus 1 may be configured to select a product code based on the defined rules.

[0058] In this case, the storage unit 12 stores rule information indicating rules for selecting products to be presented from among the products registered in the product database. FIG. 7 is a diagram showing an example of the data structure of the rule information stored in the storage unit 12. The rules in the rule information are described in natural language. The rule information is set for each business operator to which the customer U1 belongs, as an example. The rule information may be set for each product code.

[0059] When there are a plurality of product codes associated with the corresponding reference product identification element group in the product database for the converted element, the output unit 135 outputs the product code selected from the plurality of product codes based on the rules indicated by the rule information. In this case, the learned model has learned the rule information, a plurality of learning product codes, one or more reference product identification element groups associated with each of the plurality of learning product codes, and the correct product code as teacher data, and inputs the rule information and the plurality of product codes to output the product code.

[0060] When there are a plurality of product codes associated with the corresponding reference product identification element group in the product database for the converted element group, the output unit 135 inputs the rule information and the plurality of product codes to the learned model, and outputs the product code output by the learned model.

[0061] Note that the information processing apparatus 1 may be configured to select, using a large language model, the product code that matches the rule indicated by the rule information from among the plurality of product codes. In this case, the storage unit 12 stores a selection language model. The selection language model is learned to output a product code by taking, as an input, an output instruction including a plurality of product codes, the rule indicated by the rule information, and an indication to output a product conforming to the rule among the plurality of product codes. The selection language model in this case is a learning output instruction and is learned based on teacher data associating a product code, the rule indicated by the rule information, an indication to output a product conforming to the rule among the plurality of product codes, and the correct product code.

[0062] The output unit 135 inputs an output instruction including a plurality of product codes specified based on a search query, a rule indicated by rule information, and an indication to output products conforming to the rule among the plurality of product codes to a selected language model, and causes the product codes to be output as an answer to the output instruction.

[0063] With the information processing apparatus 1 configured in this way, it becomes possible to identify products conforming to the rules defined for an inquiry.

[0064] The information processing apparatus 1 may be configured to determine whether the product identification element group extracted from the text information matches the format registered in the product database, and generate a converted element group when they do not match.

[0065] In this case, the conversion unit 133 determines whether the product identification element group identified by the element identification unit 132 matches a predetermined format, and generates a converted element group when it is determined that they do not match. As an example, the storage unit 12 stores the allowed format for each element attribute. As an example, the storage unit 12 defines the number of digits and data format for each element attribute. The conversion unit 133 determines whether each product identification element included in the product identification element group identified by the element identification unit 132 matches a predetermined format. When the conversion unit 133 determines that the product identification element group does not match the predetermined format, the conversion unit 133 generates a converted element group based on the product identification element group.

[0066] By configuring the information processing apparatus 1 to determine the format of the product identification element group extracted by the information processing apparatus 1 and generate a converted element group when the formats do not match, it becomes possible to identify product codes with simple processing.

[0067] When there are multiple product codes corresponding to a group of product-specific elements, the information processing apparatus 1 may be configured to generate a question for the user U2 to identify the product code. To do this, when there are multiple product codes associated with the reference product-specific element group corresponding to the converted element group in the product database, the question generation unit 136 generates a question for identifying which product code to select from among the multiple product codes.

[0068] Specifically, the storage unit 12 stores a question generation model. The question generation model is a learned model that is learned to output a question asking the user U2 for the elements for identifying which product when the product-specific element groups of multiple products are input. The question generation model is a large language model. As an example, the question generation model is learned using, as teacher data, the product-specific element groups associated with each of the multiple product codes for learning and the question sentences for learning. As an example, when the product names, vendors, and types associated with each of the multiple product codes identified based on a search query are the same, and the license periods are different between "1 year" and "6 months", the question generation model generates a question asking "Is the license period 1 year?"

[0069] The question generation unit 136 refers to the product database and acquires the product-specific element groups corresponding to each of the multiple product codes identified based on the search query. The question generation unit 136 inputs the product-specific element groups corresponding to each of the acquired multiple product codes into the question generation model to output a question sentence. The question generation unit 136 causes the generated question sentence to be displayed on the information terminal 2. The acquisition unit 131 acquires the answer obtained from the information terminal 2 as text information. Hereinafter, the information processing apparatus 1 performs processing for identifying the product-specific element group and the converted element group included in the text information obtained as an answer to the question and identifying the product code based on the already acquired converted element group.

[0070] The information processing apparatus 1 is configured to generate a question for specifying a product code, so that it is possible to specify the product code even when product specifying elements are lacking in the text information.

[0071] The information processing apparatus 1 may be configured to determine whether the text information is an inquiry regarding a product and specify the product to be inquired about. In this case, the determination unit 137 analyzes the meaning of the text included in the text information and determines whether the analyzed text information includes the content of an inquiry regarding a product. As an example, when the determination unit 137 includes a character string that means an inquiry such as "Please provide a quotation", it determines that the content of an inquiry regarding a product is included. When the determination unit 137 determines that the content of an inquiry regarding a product is included, the element specifying unit 132 specifies a product specifying element group.

[0072] Note that the storage unit 12 stores a determination model that is a learned model learned to determine whether the text information is an inquiry regarding a product with the text information as an input, and the determination model may be input with the text information to determine whether the text information is an inquiry regarding a product.

[0073] By configuring the information processing apparatus 1 to determine whether the text information is an inquiry regarding a product, it is possible to avoid performing a process of specifying a product code for text information that is not intended for inquiry.

[0074] The information processing apparatus 1 may be configured to generate a converted element group by a learned model that has learned the converted element group output by the clustering language model.

[0075] The learning unit 138 learns a name-matching model by using, as teacher data, a group of product-specific elements and a group of converted elements output by the name-matching language model with the group of product-specific elements as input. That is, a name-matching model is generated by learning, as teacher data, the character strings constituting each of the groups of product-specific elements and the group of converted elements generated by the conversion unit 133 for each of the character strings. The name-matching model is a learned model that outputs a group of converted elements when a character string constituting each of the groups of product-specific elements is input. The learning unit 138 generates a name-matching model by causing a machine learning model not intended for natural language processing to learn teacher data. The conversion unit 133 inputs the group of product-specific elements to the name-matching model generated by the learning unit 138 and outputs a group of converted elements.

[0076] The information processing apparatus 1 is configured to output a group of converted elements by using a learned model that is a non-language model that has learned the group of converted elements output by the name-matching language model, making it possible to generate a group of converted elements with simple processing.

[0077] The information processing apparatus 1 may be configured to select a product code by using a learned model that has learned the product codes output by the selection language model. In this case, the learning unit 138 generates a selection model that has learned, as teacher data, an association between a plurality of product codes specified based on a search query, rule information, and the product codes output by the selection language model based on an output instruction including the plurality of product codes and the rule information. The selection model outputs a product code when a plurality of product codes and rule information are input. The learning unit 138 generates a selection model by causing a machine learning model not intended for natural language processing to learn teacher data. The output unit 135 inputs the plurality of product codes specified based on the search query and the rule information to the selection model and outputs a product code.

[0078] By configuring the information processing apparatus 1 to select a product code by using a selection model that is a non-language model that has learned the information output by the selection language model, it becomes possible to select a product code to be presented to the customer U1 with simple processing.

[0079] [Flow of Processing in Information Processing Apparatus 1] FIG. 8 is a flowchart showing the flow of processing in the information processing apparatus 1. The flowchart shown in FIG. 8 starts from the point when text information becomes acquirable. The acquisition unit 131 acquires text information (S01). The element identification unit 132 identifies a product identification element group based on the text information (S02). The conversion unit 133 converts the product identification element group into a converted element group (S03).

[0080] The query generation unit 134 generates a search query based on the converted element group (S04). The output unit 135 searches the product database with the generated search query and identifies a product code (S05). The output unit 135 outputs the product code (S06). Then, the information processing apparatus 1 ends the processing.

[0081] [Effect of Information Processing Apparatus 1] The information processing apparatus 1 is configured to acquire text information, convert a product identification element group extracted from the acquired text information into a predetermined format, and generate a search query for searching a product code associated with the converted element group, so that a product expressed in a format different from a product name registered in a database in advance can be easily identified.

[0082] As described above, the present invention has been described using embodiments, but the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of the gist. For example, all or part of the apparatus can be functionally or physically distributed and integrated in any unit. Also, new embodiments resulting from any combination of a plurality of embodiments are included in the embodiments of the present invention. The effects of the new embodiments resulting from the combination have the effects of the original embodiments combined.

Description of Reference Numerals

[0083] 1 Information Processing Apparatus 2 Information Terminal 11 Communication Unit 12 Memory unit 13 Control unit 131 Acquisition unit 132 Element identification unit 133 Conversion unit 134 Query generation unit 135 Output unit 136 Question generation unit 137 Judgment unit 138 Learning unit

Claims

1. an acquisition unit that acquires text information including customer identification information that is information for identifying a customer making an inquiry about a product; an element specifying unit that specifies a group of product specifying elements for specifying a product that is the subject of an inquiry in the text information; a conversion unit that generates a converted element group, which is a character string obtained by converting a character string constituting each of the product specifying element groups so that the character string matches a format registered in a product database; a query generation unit that generates a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, and a product code are associated with each other based on the converted element group; an output unit that outputs a product code associated with the reference product identifying element group that corresponds to the converted element group included in the search query in the product database; having the output unit, when there are a plurality of product codes associated with the corresponding reference product identification element group in the product database for the converted element group, outputs a product code selected from the plurality of product codes based on a customer attribute indicated by the customer identification information. Information processing device.

2. An acquisition unit that acquires text information; a determination unit that analyzes the meaning of text included in the text information and determines whether the analyzed text information includes the content of an inquiry about a product; an element specifying unit that specifies a group of product specifying elements for specifying a product that is the subject of the inquiry in the text information when the determination unit determines that the content of the inquiry includes a product; a conversion unit that generates a converted element group, which is a character string obtained by converting a character string constituting each of the product specifying element groups so that the character string matches a format registered in a product database; a query generation unit that generates a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, and a product code are associated with each other based on the converted element group; an output unit that outputs a product code associated with the reference product identifying element group that corresponds to the converted element group included in the search query in the product database; An information processing device having the above configuration.

3. An acquisition unit that acquires text information; an element specifying unit that specifies a group of product specifying elements for specifying a product that is the subject of an inquiry in the text information; a conversion unit that generates a converted element group, which is a character string obtained by converting a character string constituting each of the product specifying element groups so that the character string matches a format registered in a product database; a query generation unit that generates a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, and a product code are associated with each other based on the converted element group; an output unit that outputs a product code associated with the reference product identifying element group that corresponds to the converted element group included in the search query in the product database; a learning unit that learns character strings constituting each of the product identification element groups and converted element groups generated by the conversion unit for the character strings as training data, and generates a name matching model that is a trained model that outputs a converted element group when a character string constituting each of the product identification element groups is input; An information processing device having the above configuration.

4. An acquisition unit that acquires text information; an element specifying unit that specifies a group of product specifying elements for specifying a product that is the subject of an inquiry in the text information; a conversion unit that generates a converted element group, which is a character string obtained by converting a character string constituting each of the product specifying element groups so that the character string matches a format registered in a product database; a query generation unit that generates a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, and a product code are associated with each other based on the converted element group; an output unit that outputs a product code associated with the reference product identifying element group that corresponds to the converted element group included in the search query in the product database; a rule information storage unit that stores rule information indicating a rule for selecting a product to be presented from among the products registered in the product database, the rule information being written in a natural language; a model storage unit that stores a trained model trained using the rule information, a plurality of learning product codes, one or more reference product identification element groups associated with each of the plurality of learning product codes, and a correct product code as training data; and having When the converted element group has a plurality of product codes associated with the corresponding reference product identification element group in the product database, the output unit inputs the rule information and the plurality of product codes to the trained model, and outputs the product code output by the trained model. Information processing device.

5. The reference product identification element group includes information indicating product attributes, the query generation unit identifies an element attribute of the converted element group, and generates the search query for searching for an attribute of the product corresponding to the element attribute and the converted element group, and a product code associated with the reference product identification element group. The information processing device according to claim 1 .

6. the conversion unit inputs, to a language model, character strings constituting each of the product specifying element groups and an output instruction to output a converted element group corresponding to the character strings, and outputs the converted element group corresponding to the character strings, thereby generating a converted element group. The information processing device according to claim 1 .

7. the conversion unit determines whether the product specification element group matches a format registered in the product database, and generates the converted element group when it determines that the product specification element group does not match a format registered in the product database. The information processing device according to claim 1 .

8. The acquisition unit acquires customer attribute information indicating attributes of the customer indicated by the customer identification information, The information processing device further includes a model storage unit that stores a trained model trained using training customer attribute information indicating attributes of training customers, a plurality of training product codes, and a correct product code among the plurality of training product codes as training data; When the converted element group has a plurality of product codes associated with the corresponding reference product identification element group in the product database, the output unit inputs the customer attribute information and the plurality of product codes into the trained model, and selects the product code output by the trained model. The information processing device according to claim 1 .

9. a rule information storage unit that stores rule information indicating a rule for selecting a product to be presented from the products registered in the product database; the output unit, when there are a plurality of product codes associated with the corresponding reference product specifying element group in the product database for the converted element group, outputs a product code selected from the plurality of product codes based on a rule indicated by the rule information. The information processing device according to claim 1 .

10. a question generating unit configured to generate a question for specifying which of the plurality of product codes is to be selected when there are a plurality of product codes associated with the reference product specifying element group corresponding to the converted element group in the product database; The information processing device according to claim 1 .

11. An acquisition unit that acquires text information; an element specifying unit that specifies a group of product specifying elements for specifying a product that is the subject of an inquiry in the text information; a conversion unit that generates a converted element group, which is a character string obtained by converting a character string constituting each of the product specifying element groups so that the character string matches a format registered in a product database; a query generation unit that generates a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, and a product code are associated with each other based on the converted element group; a first output unit that outputs a plurality of product codes associated with the reference product identifying element group that corresponds to the converted element group included in the search query in the product database; a rule information storage unit that stores rule information in which rules for selecting products to be presented from among the products registered in the product database are written in a natural language; a model storage unit that stores a language model that outputs a response to an output instruction when the output instruction is input; a second output unit that outputs a product code that includes the multiple product codes output by the first output unit and the rule information stored in the rule information storage unit, and that is output as a result of inputting an output instruction to the language model to output a product code according to the rule information from among the multiple product codes; An information processing device having the above configuration.

12. a learning unit that generates a trained model trained using the multiple product codes output by the first output unit, the rule information stored in the rule information storage unit, and the product codes output by the second output unit as training data; The information processing device according to claim 11.

13. The computer executes acquiring text information including customer identification information for identifying a customer making an inquiry about a product; identifying a group of product identification elements for identifying a product that is the subject of an inquiry in the text information; generating a converted element group, the converted element group being a string obtained by converting character strings constituting each of the product specifying element groups so that the character strings match a format registered in a product database; generating a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, is associated with a product code based on the converted element group; outputting a product code associated with the reference product identifying element group corresponding to the converted element group included in the search query in the product database; having In the outputting step, when there are a plurality of product codes associated with the corresponding reference product identification element group in the product database for the converted element group, a product code selected from the plurality of product codes is output based on the customer attribute indicated by the customer identification information. Information processing methods.

14. On the computer, acquiring text information including customer identification information for identifying a customer making an inquiry about a product; identifying a group of product identification elements for identifying a product that is the subject of an inquiry in the text information; generating a converted element group, the converted element group being a string obtained by converting character strings constituting each of the product specifying element groups so that the character strings match a format registered in a product database; generating a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, is associated with a product code based on the converted element group; outputting a product code associated with the reference product identifying element group corresponding to the converted element group included in the search query in the product database; Run the command, In the outputting step, when there are a plurality of product codes associated with the corresponding reference product identification element group in the product database for the converted element group, a product code selected from the plurality of product codes is output based on the customer attribute indicated by the customer identification information. program.

15. The computer executes obtaining text information; identifying a group of product identification elements for identifying a product that is the subject of an inquiry in the text information; generating a converted element group, the converted element group being a string obtained by converting character strings constituting each of the product specifying element groups so that the character strings match a format registered in a product database; generating a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, is associated with a product code based on the converted element group; a first output step of outputting a plurality of product codes associated with the reference product identifying element group corresponding to the converted element group included in the search query in the product database; a second output step of referring to a rule information storage unit that stores rule information, in which rules for selecting products to be presented from products registered in a product database are described in natural language, and, upon inputting an output instruction to output a product code from among the multiple product codes output in the first output step and the rule information stored in the rule information storage unit in accordance with the rule information, outputting a product code output as a result of inputting the product code into a language model that outputs a response to the output instruction; An information processing method comprising the steps of:

16. On the computer, obtaining text information; identifying a group of product identification elements for identifying a product that is the subject of an inquiry in the text information; generating a converted element group, the converted element group being a string obtained by converting character strings constituting each of the product specifying element groups so that the character strings match a format registered in a product database; generating a search query for searching a product database in which a reference product identification element group, which is an element for identifying a product, is associated with a product code based on the converted element group; a first output step of outputting a plurality of product codes associated with the reference product identifying element group corresponding to the converted element group included in the search query in the product database; a second output step of referring to a rule information storage unit that stores rule information, in which rules for selecting products to be presented from products registered in a product database are described in natural language, and, upon inputting an output instruction to output a product code from among the multiple product codes output in the first output step and the rule information stored in the rule information storage unit in accordance with the rule information, outputting a product code output as a result of inputting the product code into a language model that outputs a response to the output instruction; A program for executing the above.

Images