Information processing device

Abstract

To appropriately select an evaluator who evaluates a product or service.SOLUTION: An information processing device determines an evaluator who evaluates a product or service provided by a provider from among a plurality of evaluator's candidates. The information processing device has a control unit which is configured to: learn a user model using first data about the provider and second data about the evaluator who evaluated the product or service provided by the provider in the past as input data and third data indicating a usefulness degree of evaluation performed by the evaluator in the past as output data; and select the evaluator matched to the provider from among the plurality of evaluators by using the user model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to ratings for products or services. [Background technology]

[0002] Systems for evaluating human resources are known. For example, Patent Document 1 discloses an invention related to a system for managing a list of human resources based on technical proficiency and extracting human resources with skills that meet customer requirements. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-099652 [Patent Document 2] Japanese Patent Application Laid-Open No. 2004-046564 Summary of the Invention [Problem to be solved by the invention]

[0004] The present disclosure aims to appropriately select evaluators who evaluate products or services. [Means for solving the problem]

[0005] One aspect of an embodiment of the present disclosure is An information processing device that determines an evaluator who will evaluate a product or service provided by a provider from among a plurality of evaluator candidates, the information processing device having a control unit that executes the following: learning a user model using first data about the provider and second data about the evaluator who has previously evaluated the product or service provided by the provider as input data, and third data indicating the usefulness of the evaluation made by the evaluator in the past as output data; and using the user model to select the evaluator who is suitable for the provider from among the plurality of evaluators.

[0006] Other aspects include a method executed by the information processing device, a program for causing a computer to execute the method, or a computer-readable storage medium non-temporarily storing the program. [Effects of the Invention]

[0007] According to the present disclosure, it is possible to appropriately select evaluators who evaluate products or services. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram for explaining an overview of a system according to an embodiment. [Figure 2] FIG. 2 is a diagram illustrating the module configuration of the server device 1. [Figure 3] FIG. 2 is a diagram for explaining data stored in a storage unit 12. [Figure 4] FIG. 3 is a diagram for explaining the flow of processing in the control unit 11. [Figure 5] 4 is a flowchart of a process executed by the server device 100. [Figure 6] 4 is a flowchart of a process executed by the server device 100. DETAILED DESCRIPTION OF THE INVENTION

[0009] There is a known system in which a third party evaluates the products and services provided by a provider. For example, when a provider starts to provide a new product or service, the provider allows a third party to use the product or service in advance and provides feedback on the evaluation, thereby improving the quality of the product or service. It can be done. In the following description, a person who provides a product or service will be referred to as a "servicer" or "provider," and a person who evaluates the product or service will be referred to as a "tester" or "evaluator."

[0010] When using testers to evaluate products or services, it is important to select the right testers. In the past, testers were selected based on their personal attributes, such as their preferences and areas of expertise. However, depending on the attributes of the target product or service, there are cases where it is not possible to select an appropriate tester. For example, if the product being evaluated is a food ingredient, testers may be selected from among people who have the attribute of being good at cooking. However, if the product being evaluated is kelp, even if a tester is good at cooking, they should not be selected if they do not normally make dashi. As described above, conventional methods of selecting testers based on a single attribute (for example, area of ​​expertise) can result in cases where effective testers cannot be selected. The information processing device according to this embodiment solves such a problem.

[0011] An information processing apparatus according to one embodiment is an information processing apparatus that determines an evaluator who will evaluate a product or service provided by a provider from among a plurality of evaluator candidates. Specifically, the control unit has the following: learning a user model using first data about the provider and second data about the evaluator who previously evaluated the product or service provided by the provider as input data; and third data indicating the usefulness of the evaluation made by the evaluator in the past as output data; and using the user model to select the evaluator who is suitable for the provider from among the multiple evaluators.

[0012] The provider is the person who provides the product or service in question. The provider is typically the developer or inventor of the product, but the provider can also be a business entity. The first data is data relating to the provider (servicer), and the second data is data relating to the evaluator (tester).

[0013] The first data is, for example, a set of multiple attributes that the provider has. Examples of the attributes that the provider has include the type of business the provider operates in, and the categories of products and services that the provider provides. The second data is, for example, a collection of multiple attributes that the evaluator has, such as the evaluator's educational background, occupation, field of expertise, hobbies, preferences, and past behavior history (such as visited spots).

[0014] The third data is data indicating the usefulness of the evaluations made by the evaluator in the past. For example, suppose that an evaluator has previously evaluated a certain product or service and left useful information for the provider. In this case, the third data describes that "the evaluator left a highly useful evaluation."

[0015] The usefulness indicated by the third data may be calculated based on predetermined data. For example, suppose that a provider attempts to improve the product or service that it provides based on the evaluation performed by the evaluator. Also, suppose that sales or profits change as a result. In this case, the amount of change in sales or profits can be the usefulness of the evaluation. The usefulness may be a value calculated based on a predetermined standard (for example, the sales or profit amount mentioned above), or may be a score given by the provider in response to the evaluation made by the evaluator.

[0016] The control unit generates a user model using the first to third data as learning data. For example, the user model is a model that outputs an expected usefulness of the evaluation when data regarding the attributes of the provider and the evaluator are input. Whether an evaluator can leave a valid review depends on whether the attributes of the evaluator (e.g., area of ​​expertise) match the attributes of the evaluator expected by the provider. Therefore, the user model learns the relationship between the attributes of the evaluator and the usefulness of the reviews that the evaluator has given in the past for a specific provider. In other words, the user model is a model that outputs the usefulness of the reviews given by the evaluator to the provider (in other words, the compatibility between the provider and the evaluator). By using the user model, it is possible to predict how effective an evaluation will be for each of multiple evaluators, and also to screen evaluators who are expected to provide more effective evaluations.

[0017] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. The configurations of the following embodiments are examples, and the present disclosure is not limited to the configurations of the embodiments.

[0018] (First embodiment) An overview of a server device according to a first embodiment will be described. The server device according to this embodiment is a device that selects evaluators who evaluate products or services provided by providers of the products or services.

[0019] An overview of the processing performed by the server device will be described with reference to FIG. First, the provider provides a first product or service to the evaluator, who then evaluates it. The evaluator then provides the provider with feedback on the evaluation results, including impressions from using the first product or service and suggestions for improvement.

[0020] The provider who receives the evaluation results will implement improvements to the first product or service in accordance with the content of the evaluation. Whether or not an evaluation made by an evaluator is beneficial to a provider can be determined by the results of improving the product or service in accordance with the contents of the evaluation. For example, if the sales of a product or service increase as a result of improving the product or service in accordance with the improvement suggestions included in the evaluation, the evaluation can be said to have been beneficial to the provider.

[0021] In this embodiment, the provider transmits the improvement results (i.e., information for calculating the usefulness of the evaluation) to the server device, and the server device learns the usefulness of the evaluation based on the results. For example, the server device learns the relationship between multiple attributes possessed by the provider, multiple attributes possessed by the rater, and the usefulness of the evaluation using a machine learning model.

[0022] Consider a case where a provider releases a second product or service following a first product or service. In this case, the provider preferably selects an evaluator who will provide a more beneficial evaluation for the second product or service. In this embodiment, the server device uses a trained machine learning model to select an evaluator who is predicted to be able to leave a beneficial evaluation for the provider. This makes it possible to provide the provider with information such as "which evaluator should be asked to provide an evaluation to obtain a more beneficial evaluation."

[0023] [Device configuration] FIG. 2 is a diagram showing an example of the configuration of the server device 1. As shown in FIG. The server device 1 is, for example, a computer such as a server device, a personal computer, a smartphone, a mobile phone, a tablet computer, or a personal digital assistant. The device 1 includes a control unit 11 , a storage unit 12 , and an input / output unit 13 .

[0024] The server device 1 trains a machine learning model using data related to the provider, data related to the evaluator, and the usefulness of past evaluations by the evaluator as learning data. The trained machine learning model is also used to select an evaluator who can provide a useful evaluation for a certain provider.

[0025] The server device 1 can be configured as a computer having a processor (CPU, GPU, etc.), a main memory device (RAM, ROM, etc.), and an auxiliary memory device (EPROM, hard disk drive, removable media, etc.). The auxiliary memory device stores an operating system (OS), various programs, various tables, etc., and by executing the programs stored therein, various functions (software modules) that match predetermined purposes, as described below, can be realized. However, some or all of the functions may be realized as hardware modules using hardware circuits such as ASICs, FPGAs, etc.

[0026] The control unit 11 is a computing unit that executes predetermined programs to realize various functions of the server device 1. The control unit 11 can be realized by, for example, a hardware processor such as a CPU. The control unit 11 may also be configured to include RAM, ROM (Read Only Memory), cache memory, etc.

[0027] The control unit 11 is configured to have three software modules: a data acquisition unit 111, a learning unit 112, and an evaluation unit 113. Each software module may be realized by the control unit 11 (CPU) executing a program stored in the storage unit 12, which will be described later.

[0028] The data acquisition unit 111 acquires data (learning data) for training a machine learning model. In this embodiment, the learning data includes three types: data related to the provider (provider data), data related to the evaluator (evaluator data), and data indicating the usefulness of evaluations previously performed by the evaluator (evaluation result data).

[0029] Provider data is a collection of multiple attributes related to a provider. The provider data may include attributes of the provider himself / herself as well as attributes of the products and services provided by the provider. The provider data may include, for example, the provider's user identifier, the category (field) of the products and services provided by the provider, the industry, etc. In the following description, a collection of multiple attributes will be referred to as attribute information. The provider data may be generated by the data acquisition unit 111 based on the provider's attribute information input by the provider himself / herself, or may be acquired from another device that manages user information.

[0030] The evaluator data is a collection of multiple attributes related to the evaluator. The evaluator data includes the attributes of the evaluator himself / herself. The evaluator data may include, for example, the evaluator's user identifier, the evaluator's personal information (gender, age, educational background, etc.), the evaluator's field of expertise, specialization, preferences, areas of interest, personal evaluation of the evaluator, and the evaluator's personal trends. The evaluator data may be generated by the data acquisition unit 111 based on attribute information of the evaluator input by the evaluator himself / herself or a third party, or may be acquired from another device that manages user information. Examples of data input by a third party include the evaluator's expertise and personal evaluation.

[0031] Evaluation result data is data that indicates the usefulness of an evaluation made by a specific evaluator for a product or service provided by a specific provider. For example, if a certain evaluator previously evaluated a certain product or service, and the provider attempts to improve the product or service based on this evaluation, If this results in an increase in sales, average customer spending, profits, etc., then the evaluation can be said to be beneficial to the provider.

[0032] The data acquisition unit 111 may directly import the evaluation result data, or may generate the evaluation result data based on data relating to evaluations that have been performed in the past. For example, the data acquisition unit 111 can acquire a history of the results of an evaluator's attempts to improve a product or service in accordance with an improvement suggestion included in an evaluation made by a specific evaluator in the past (hereinafter referred to as improvement history data), and generate evaluation result data after calculating the usefulness of the evaluation based on the history. The improvement history data may indicate, for example, the changes in sales or profits of the product or service before and after the improvement suggestion. The improvement history data may be input by the provider via the input / output unit 13.

[0033] In this embodiment, the evaluation result data includes a numerical value indicating the usefulness of the evaluation. The numerical value may be, for example, a dimensionless number obtained by normalizing the usefulness of the evaluation to a predetermined range.

[0034] The learning unit 112 trains a machine learning model (an example of a user model, hereinafter referred to as a provider model) using the attributes (provider attributes and evaluator attributes) included in the data acquired by the data acquisition unit 111 and the usefulness of the evaluation as training data. The provider model is stored in the storage unit 12, which will be described later. By performing the learning, the learning unit 112 can obtain a machine learning model that has learned the relationship between the attributes of the provider, the attributes of the evaluator, and the usefulness of the evaluation.

[0035] The evaluation unit 113 uses the trained provider model to select evaluators who are predicted to leave useful reviews for a specific provider. Specifically, the evaluation unit 113 generates combinations of the target provider with multiple evaluators registered in the system, inputs the provider attributes and the evaluator attributes for each combination into the provider model, and obtains a predicted value of the usefulness of the review as an output. The evaluation unit 113 outputs information about evaluators whose predicted usefulness of the review exceeds a predetermined value.

[0036] The storage unit 12 is a means for storing information, and is configured with storage media such as RAM, a magnetic disk, a flash memory, etc. The storage unit 12 stores programs executed by the control unit 11, data used by the programs, etc.

[0037] The storage unit 12 stores the provider data, evaluator data, evaluation result data, and provider model acquired by the data acquisition unit 111.

[0038] Here, examples of provider data, evaluator data, and evaluation result data will be described. Fig. 3(A) shows an example of provider data. The provider data includes a set of multiple attributes related to the provider. In the illustrated example, the provider data includes the user identifier (user ID) of the provider, data related to the category to which the product or service provided by the provider belongs, the industry, the type of service, the phase, the purpose, etc. These data may be input by the provider.

[0039] FIG. 3(B) is an example of the evaluator data. The evaluator data includes a set of multiple attributes related to the evaluator. In the illustrated example, the evaluator data includes data related to the evaluator's user identifier (user ID), sex, age, educational background, specialty, preferences, etc. Note that the evaluator data may include other data related to the evaluator's attributes than those shown in the example. For example, the evaluator data may include personal interests, trends, schedules (conditions for participating in the evaluation), remuneration, etc. Also, for example, the evaluator's personal characteristics (friendliness, personality, relationships, character, etc.) may be evaluated by a third party, and the results may be used to determine the evaluator's personal characteristics. The results of may be included in the evaluator data.

[0040] FIG. 3(C) is an example of the evaluation result data. As described above, the evaluation result data is data indicating the usefulness of evaluations made by evaluators in the past. In the illustrated example, the evaluation result data includes an identifier (evaluation ID) of the evaluation made in the past, the user ID of the evaluator, the user ID of the provider, the content of the evaluation, and the usefulness. The usefulness may be a dimensionless number or a numerical value related to the usefulness of the evaluation, as long as it is a numerical value indicating the usefulness of the evaluation. The usefulness may be calculated by the data acquisition unit 111 based on the improvement history data, or may be a score assigned by the provider himself / herself. FIG. 3D will be described later in the second embodiment.

[0041] Returning to Figure 2, we continue the explanation. The input / output unit 13 is a means for receiving input operations performed by an operator and presenting information to the operator. Specifically, the input / output unit 13 includes devices for input such as a mouse and a keyboard, and devices for output such as a display and a speaker. The input / output devices may be integrally configured with, for example, a touch panel display.

[0042] The specific hardware configuration of the server device 1 may include omissions, substitutions, and additions of components as appropriate depending on the embodiment. For example, the control unit 11 may include multiple hardware processors. The hardware processor may be configured with a microprocessor, FPGA, GPU, etc. Furthermore, input / output devices other than those illustrated (for example, an optical drive, etc.) may be added. Furthermore, the server device 1 may be configured with multiple computers. In this case, the hardware configurations of the computers may or may not be the same.

[0043] Next, the flow of processing performed by the control unit 11 of the server device 1 will be described with reference to FIG. First, the data acquisition unit 111 acquires data necessary for generating provider data, evaluator data, and evaluation result data.

[0044] In this embodiment, the data acquisition unit 111 acquires data related to the attributes of the target provider, and generates the provider data shown in FIG. 3(A) based on the data. At this time, data conversion or integration may be performed. The data acquisition unit 111 may accept input of data via the input / output unit 13. Note that the data acquisition unit 111 may directly acquire the provider data via the input / output unit 13.

[0045] The data acquisition unit 111 also acquires data related to the attributes of the evaluators and generates the evaluator data shown in FIG. 3(B) based on the acquired information. At this time, data conversion or integration may be performed. The data acquisition unit 111 may receive data input via the input / output unit 13. The data acquisition unit 111 may also directly acquire the evaluator data via the input / output unit 13. Furthermore, the data acquisition unit 111 may acquire part of the data for generating the evaluator data from an external device.

[0046] Furthermore, the data acquisition unit 111 acquires the above-mentioned improvement history data, calculates the usefulness of the evaluation based on the data, and then generates evaluation result data. The improvement history data may include, for example, numerical values ​​that can be used to evaluate the usefulness (for example, the amount or rate of change in sales or profit before and after the improvement). The improvement history data may be received from an external device such as a sales management server. The data acquisition unit 111 may also allow the provider to directly input the usefulness via the input / output unit 13.

[0047] Note that the storage unit 12 may store data that serves as a reference for calculating the degree of usefulness, and the data acquisition unit 111 may use this data to calculate the degree of usefulness. An example of such criteria is, for example: The more sales increase before and after the improvement, the higher the profitability is calculated. The more profit there is before and after the improvement, the higher the profitability is calculated. The more customers there are before and after the improvement, the higher the profitability will be calculated. The more customer satisfaction increases before and after the improvement, the higher the benefit is calculated. Examples include:

[0048] In this embodiment, the data used to calculate the usefulness of the evaluation includes, for example, the trends in sales and profits of products and services before and after the improvement proposal, but the usefulness of the evaluation may also be calculated using other criteria. The usefulness of an evaluation indicates the degree to which a provider has achieved positive results by receiving the evaluation. Therefore, the indicator does not necessarily have to be monetary, etc., as long as it is possible to calculate the degree to which a provider has achieved positive results by receiving the evaluation. The degree of usefulness may also be calculated based on factors other than commercial considerations. For example, if the evaluation results in an improvement in the provider's skills or abilities, the degree of usefulness for the evaluation may be increased, assuming that a positive result has been achieved.

[0049] The provider data, evaluator data, and evaluation result data generated by the data acquisition unit 111 are stored in the storage unit 12.

[0050] Next, the learning unit 112 learns the provider model based on the three types of stored data. The learning unit 112 extracts corresponding multiple pieces of attribute information from the provider data and the evaluator data to generate input data, and extracts the usefulness from the evaluation result data to generate output data. The learning unit 112 also learns the provider model using these as learning data. The input of the learning data may be repeated at a predetermined interval. This makes it possible to obtain a provider model that outputs a predicted value of the usefulness of the evaluation when multiple attributes held by the provider and multiple attributes held by the evaluator are input.

[0051] The evaluation unit 113 selects an evaluator suitable for a specific provider, triggered by an instruction from the operator of the server device 1. The evaluation unit 113 prompts the operator to specify the target provider and extracts corresponding records from the provider data. The evaluation unit 113 also extracts records of multiple evaluators (i.e., evaluator candidates) that can be combined with the target provider from the evaluator data. The evaluation unit 113 inputs attribute information for each combination of provider and evaluator into a provider model and obtains a predicted value of usefulness as output. As a result, a predicted value of usefulness is obtained for each evaluator candidate. The evaluation unit 113 then outputs information about evaluators whose obtained usefulness exceeds a predetermined value.

[0052] [flowchart] Next, the process executed by the server device 1 according to this embodiment will be described. The processing executed by the server device 1 can be divided into a phase in which a provider model is trained (training phase) and a phase in which prediction is made based on the trained provider model (prediction phase). 5 is a flowchart of the learning phase executed by the server device 1. The illustrated process is started by an operation by the operator of the server device 1.

[0053] First, in step S11, the data acquisition unit 111 collects data on the attributes of the provider. The data may be acquired via the input / output unit 13 or from an external device. Next, in step S12, the data acquisition unit 111 calculates the following data based on the collected data: Generate provider data as described in 3(A). Provider data is generated for all providers for whom attribute data is collected.

[0054] In step S13, the data acquisition unit 111 collects data on the attributes of the evaluators. The data may be acquired via the input / output unit 13 or from an external device. Next, in step S14, the data acquisition unit 111 generates evaluator data such as that described with reference to FIG. 3(B) based on the collected data. The evaluator data is generated for all evaluators for whom attribute data has been collected.

[0055] Next, in step S15, the data acquisition unit 111 acquires improvement history data. As described above, the improvement history data is a history of the results of an evaluator improving a product or service in accordance with an improvement proposal included in an evaluation made by a specific evaluator in the past. The improvement history data may include the evaluator, the provider, the identifier of the target product or service, the content of the improvement proposal, and the improvement results (such as the progress of sales and profits of the product or service). The improvement history data may include multiple records.

[0056] Next, in step S16, the data acquisition unit 111 generates evaluation result data based on the improvement history data. In this step, the data acquisition unit 111 calculates the usefulness based on the acquired improvement results, for example, and then generates evaluation result data as described with reference to FIG. 3(C).

[0057] Next, in step S17, the learning unit 112 learns the provider model based on the evaluator data, provider data, and evaluation result data (benefits). For example, the learning unit 112 converts values ​​stored in multiple fields included in each data into features and inputs them as learning data into the provider model, thereby learning the provider model. This allows the provider model to learn the relationship between what attributes an evaluator has and how beneficial an evaluation they can leave for a specific provider.

[0058] 6 is a flowchart of the prediction phase executed by the server device 1. The illustrated process is started by an operation by the operator of the server device 1.

[0059] First, in step S21, the evaluation unit 113 generates combinations of evaluators and providers. In this step, information specifying the target provider is acquired via the input / output unit 13, and then evaluator candidates are acquired. The evaluator candidates may be filtered based on minimum conditions (for example, conditions presented by the provider).

[0060] Next, in step S22, the evaluation unit 113 inputs attribute information corresponding to the evaluator and provider into the provider model for each combination of the evaluator and provider, and obtains a predicted value of the usefulness as an output. This step allows the predicted value of the usefulness for the combination of the evaluator and provider to be obtained.

[0061] Next, in step S23, the evaluation unit 113 selects an evaluator who has obtained a predicted value of usefulness that exceeds a predetermined value, and outputs information related to that evaluator (e.g., user ID, personal information, contact information, etc.) as an evaluator who is suitable for the target provider.

[0062] As described above, the server device according to this embodiment uses the usefulness of past evaluations by evaluators as learning data to learn a provider model, and selects an evaluator who will evaluate a new product or service using the provider model. In general, whether an evaluator can leave a useful evaluation for a certain product or service depends on whether the evaluator has the attributes expected by the provider. The server device according to this embodiment selects an evaluator who will evaluate a new product or service using the provider model. Since learning is performed using the attributes that the provider has, it is possible to select an evaluator with attributes that match the provider.

[0063] (Second embodiment) In the first embodiment, the provider model is trained using information such as the expertise and preferences of the evaluator. However, the provider model can also be trained using other information. In the second embodiment, the provider model is trained using data on the past behavior of the evaluator.

[0064] For example, if an evaluator has frequently visited a specific spot (e.g., a baseball stadium) in the past, the evaluator can be considered to have an attribute related to the spot (e.g., an attribute of being a baseball fan). In the second embodiment, the server device 1 acquires the past behavior history of the evaluator and learns the provider model based on the acquired data.

[0065] In the second embodiment, the data acquisition unit 111 is configured to be able to acquire data relating to past behavioral history (hereinafter, behavioral history data) for each of a plurality of evaluators. Behavioral history data is data related to the history of past behaviors of an evaluator. Examples of such data include the location history of a mobile device owned by the evaluator, the website browsing history of the evaluator, and the history of online purchases by the evaluator. The behavioral history data may be acquired from a device (such as a smartphone) associated with the evaluator, or from an external device that manages the device's location history, website access logs, and product purchase history.

[0066] In the second embodiment, the data acquisition unit 111 identifies spots previously visited by the evaluator, objects previously interacted with online by the evaluator, and items previously purchased by the evaluator, based on the acquired behavioral history data. Furthermore, additional attributes of the evaluator are determined based on the identified visited spots, objects interacted with, and purchased items.

[0067] 3(D) is an example of evaluator data in the second embodiment. In the second embodiment, the evaluator data further includes attributes based on visit history, attributes based on interaction history, and attributes based on purchase history. Attributes based on visit history are attributes determined based on spots visited by the rater in the past. Attributes based on interaction history are attributes determined based on objects with which the rater has interacted online in the past. Attributes based on purchase history are attributes determined based on items purchased by the rater in the past. These attributes may be obtained, for example, by classifying a collection of visited spots, objects interacted with, and purchased items using a machine learning model. For example, if an evaluator has visited many automobile-related stores and has interacted with many automobile-related web pages, the evaluator may be given the attribute "has extensive knowledge of automobiles."

[0068] In the second embodiment, the learning unit 112 and the evaluation unit 113 perform training and prediction of the provider model using the additional attributes described above in addition to the attributes described in the first embodiment. This configuration makes it possible to further improve the accuracy of prediction.

[0069] (Variation) The above-described embodiment is merely an example, and the present disclosure can be modified and implemented as appropriate within the scope that does not deviate from the gist of the disclosure. For example, the processes and means described in this disclosure can be freely combined and implemented as long as no technical contradiction occurs.

[0070] Furthermore, in the description of the embodiment, attribute information of the provider and the evaluator is used as the training data, but other data relating to the provider and the evaluator may also be used as the training data.

[0071] Furthermore, a process described as being performed by one device may be shared and executed by multiple devices. Alternatively, a process described as being performed by different devices may be executed by a single device. In a computer system, the hardware configuration (server configuration) by which each function is realized can be flexibly changed.

[0072] The present disclosure can also be realized by providing a computer program implementing the functions described in the above embodiments to a computer, and having one or more processors in the computer read and execute the program. Such a computer program may be provided to the computer via a non-transitory computer-readable storage medium connectable to the computer's system bus or via a network. Non-transitory computer-readable storage media include, for example, any type of disk, such as a magnetic disk (e.g., a floppy disk, a hard disk drive (HDD), etc.), an optical disk (e.g., a CD-ROM, a DVD disk, a Blu-ray disk), a read-only memory (ROM), a random access memory (RAM), an EPROM, an EEPROM, a magnetic card, a flash memory, an optical card, or any type of medium suitable for storing electronic instructions. [Explanation of symbols]

[0073] 1. Server device 11 Control section 12...Storage section 13...Input / output section

Claims

1. An information processing device that determines an evaluator who evaluates a product or service provided by a provider from among a plurality of evaluator candidates, learning a user model using, as input data, first data related to the provider and second data related to the evaluator who previously evaluated a product or service provided by the provider, and output data, third data indicating the usefulness of the evaluation previously made by the evaluator; selecting the evaluator that matches the provider from the plurality of evaluators using the user model; a control unit that executes The usefulness is a value calculated based on the results of the provider's past attempts to improve the product or service that the provider provides, based on evaluations made by the evaluator in the past. Information processing device.

2. The usefulness is a score given by the provider to an evaluation previously made by the evaluator. The information processing device according to claim 1 .

3. the second data is a set of a plurality of attributes possessed by the evaluator; The information processing device according to claim 1 .

4. The user model is a model that, when the first data and the second data are input, outputs a predicted value of the usefulness of the evaluation made by the evaluator to the provider. The information processing device according to claim 1 .

Images