JP2022002025A - Information processing device and information processing method - Google Patents

Abstract

To create an artificial intelligence model which is customized for a user using the artificial intelligence model.SOLUTION: An information processing device 1 includes an information acquisition unit 141 which acquires selection information for specifying a plurality of selection models selected from a plurality of artificial intelligence models each having a different process content, a creation unit 142 which creates an artificial intelligence model set containing the plurality of selection models corresponding to the selection information, a data acquisition unit 143 which acquires a plurality of pieces of input data to be input to the artificial intelligence model set and inputs the acquired plurality of pieces of input data to the artificial intelligence model set, and a data output unit 144 which outputs output data output by the artificial intelligence model set to which the input data is input.SELECTED DRAWING: Figure 3

Description

The present invention relates to an information processing apparatus and an information processing method using an artificial intelligence model.

Conventionally, there is known a technique for creating an artificial intelligence model that executes a desired process by performing deep learning using a learning data set (see, for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2020-61088

When creating an artificial intelligence model using conventional technology, if the learning data sets used for deep learning are equivalent, an artificial intelligence model that performs the same operation is created. In this case, there has been a problem that it is difficult for a user who uses an artificial intelligence model to differentiate from other users who use an equivalent artificial intelligence model.

Therefore, the present invention has been made in view of these points, and an object of the present invention is to provide an information processing device and an information processing method capable of creating a customized artificial intelligence model for a user who uses the artificial intelligence model. ..

The information processing apparatus according to the first aspect of the present invention has an information acquisition unit for acquiring selection information for specifying a plurality of selection models selected from a plurality of artificial intelligence models having different processing contents, and the selection information. A creation unit that creates an artificial intelligence model set including the corresponding plurality of selection models, and a plurality of input data to be input to the artificial intelligence model set are acquired, and the acquired plurality of input data are used in the artificial intelligence model set. It has a data acquisition unit for input and a data output unit for outputting output data output by the artificial intelligence model set into which the input data is input.

The data acquisition unit selects one or more input data to be used by each of the plurality of selection models from the acquired plurality of input data, and inputs the selected one or more input data to each of the plurality of selection models. You may.

The information acquisition unit presents the plurality of artificial intelligence models to an external device in association with feature information indicating the characteristics of each of the plurality of artificial intelligence models, presents the feature information to the external device, and then presents the feature information to the external device. The selection information may be acquired from.

The information acquisition unit further acquires the purpose of use of the artificial intelligence model set from the external device, and presents the plurality of artificial intelligence models selected from a plurality of artificial intelligence model candidates based on the purpose of use to the external device. You may.

Before presenting the plurality of artificial intelligence models to the external device, the information acquisition unit presents a plurality of purpose-of-use candidates, which are candidates for the purpose of use, to the external device, and selects from the plurality of purpose-of-use candidates. You may obtain one or more of the above-mentioned purposes of use.

The information acquisition department identifies the type of business of the company for which credit risk is specified, the range of sales of the company for which credit risk is specified, the size of the company for which credit risk is specified, or the credit risk. The plurality of artificial intelligence models may be presented to the external device in association with the plurality of feature information having at least one of the range of locations of the target enterprises.

The information acquisition unit associates the plurality of artificial intelligence models with the plurality of characteristic information having different types of financial assets to be invested, trading conditions of the financial assets to be invested, or at least one of the purposes of investment. It may be presented to an external device.

The creation unit may determine the charge for using the artificial intelligence model set based on the plurality of selection models indicated by the selection information.

The information processing method according to the second aspect of the present invention includes a step of acquiring selection information for identifying a plurality of selection models selected from a plurality of artificial intelligence models having different processing contents, which are executed by a computer. A step of creating an artificial intelligence model set including the plurality of selection models corresponding to the selection information, a plurality of input data to be input to the artificial intelligence model set are acquired, and the acquired plurality of input data are used as the artificial intelligence model. It has a step of inputting to the set and a step of outputting the output data output by the artificial intelligence model set into which the input data is input.

According to the present invention, there is an effect that an artificial intelligence model customized for a user who uses the artificial intelligence model can be created.

It is a figure which shows the structure of an information processing system. It is a conceptual diagram which shows the relationship between an artificial intelligence model and an artificial intelligence model set. It is a figure which shows the structure of an information processing apparatus. It is a figure which shows an example of the feature table which shows the feature of an artificial intelligence model. This is the screen for selecting the purpose of use. This is an example of a screen displayed on the customer terminal after the purpose of use is selected. This is an example of a screen displayed on the customer terminal after "Industry" is selected. It is a flowchart which shows the flow of the process executed by an information processing apparatus.

[Overview of information processing system S]
FIG. 1 is a diagram showing a configuration of an information processing system S. The information processing system S includes an information processing device 1 and a plurality of customer terminals 2 (customer terminals 2-1 to customer terminals 2-n, n is an integer of 2 or more). The information processing device 1 and the plurality of customer terminals 2 are connected to a network N such as the Internet or a public communication network.

The information processing device 1 is a device for providing a customer with an artificial intelligence (AI) model set customized according to the customer's needs, for example, a computer. The information processing apparatus 1 may be configured by one computer or may be configured by a plurality of computers.

The artificial intelligence model set is composed of a plurality of artificial intelligence models in which rules or mathematical formulas for outputting output data based on the input data are described by machine learning. The information processing device 1 corresponds to the input data by inputting data to each of the plurality of artificial intelligence models and executing arithmetic processing based on the rules or mathematical formulas of the artificial intelligence model. Multiple output data can be calculated. The information processing device 1 provides the customer terminal 2 with one or more output data calculated by inputting a plurality of input data acquired from the customer terminal 2 into an artificial intelligence model suitable for processing each input data.

The information processing device 1 can use a large number of artificial intelligence models having different functions, and one or more artificial intelligence models selected from a large number of artificial intelligence models in response to a request from a customer who uses the customer terminal 2 can be used. Construct an artificial intelligence model set that includes. When the information processing apparatus 1 receives the input data from the customer terminal 2, the information processing apparatus 1 provides the customer terminal 2 with the result of calculation using the artificial intelligence model set corresponding to the customer of the customer terminal 2.

FIG. 2 is a conceptual diagram showing the relationship between the artificial intelligence model and the artificial intelligence model set. Although AI models a to AI models i are shown in FIG. 2, the number of artificial intelligence models possessed by the information processing apparatus 1 is arbitrary. The AI model set α includes AI models a, d, and f, and is used for company A. The AI model set β includes AI models b, e, f, and i, and is used for company B.

The plurality of artificial intelligence models differ in at least one of the methods of machine learning (for example, reinforcement learning, supervised learning, etc.), the content of input data, or the content of output data. The plurality of artificial intelligence models are configured to output different data even if the same data is input, for example, due to different machine learning methods.

When multiple artificial intelligence models are models that output data indicating the degree of credit risk of a company based on input data, for example, the multiple artificial intelligence models are the type of industry and credit of the company for which credit risk is specified. At least one of the range of sales of the company for which risk is specified, the size of the company for which credit risk is specified, or the range of the location of the company for which credit risk is specified corresponds to different processing. do. When multiple artificial intelligence models are models that output data for supporting investment in financial assets based on input data, for example, the plurality of artificial intelligence models are the types of financial assets to be invested and investment. Corresponds to processing in which at least one of the trading conditions or the purpose of investment of the target financial asset is different.

An artificial intelligence model set is characterized by a combination of multiple artificial intelligence models that make up the artificial intelligence model set. In the case shown in FIG. 2, if the AI model set α aims to identify the credit risk of a small and medium-sized enterprise that manufactures and sells daily necessities, the AI model set α is a company that manufactures and sells daily necessities. The AI model a suitable for the analysis of SMEs and the AI model d suitable for the analysis of small and medium-sized enterprises are included. If the AI model set β aims to identify the credit risk of a large company that provides a communication service, the AI model set β is an AI model b suitable for analysis of the company that provides the communication service. , And AI model e suitable for analysis of large enterprises.

The purpose of use of the artificial intelligence model set is arbitrary and can be applied to various areas where artificial intelligence is used. The purpose of using the artificial intelligence model set is, for example, risk management such as management of corporate credit risk, management of equipment failure risk, management of human resource trouble occurrence risk, management of investment in financial assets, management of capital investment. Or, management related to investment such as management of investment in companies.

The information processing device 1 provides information indicating the characteristics of each of the plurality of artificial intelligence models to the plurality of customer terminals 2, and the customer who intends to use the artificial intelligence model set uses the information processing device 1 via the customer terminal 2. The selection information using one or more artificial intelligence models desired to be used is transmitted to the information processing apparatus 1. Upon receiving the selection information, the information processing apparatus 1 creates an artificial intelligence model set for the customer who has transmitted the selection information. After that, when the information processing apparatus 1 receives the input data from the customer terminal 2, the received input data is input to the artificial intelligence model set for the customer, and the output data output from the artificial intelligence model set is sent to the customer terminal 2. offer.

As described above, the information processing apparatus 1 has one or more artificial intelligences from a plurality of artificial intelligence models so as to be able to provide the optimum artificial intelligence model set to the customer according to the purpose of use of the data output from the artificial intelligence model set. Select a model set. Therefore, according to the information processing apparatus 1, the customer who uses the artificial intelligence model set can use the artificial intelligence model set that is most suitable for the company, and the probability that the competitor will use the same artificial intelligence model set is determined. Since it can be lowered, it will be possible to effectively utilize the artificial intelligence model to achieve the purpose while trying to differentiate from the competitors.

[Configuration of information processing device 1]
FIG. 3 is a diagram showing the configuration of the information processing apparatus 1. The information processing device 1 includes a communication unit 11, a storage unit 12, an artificial intelligence processing unit 13, and a control unit 14. The artificial intelligence processing unit 13 has an artificial intelligence model group 131 and an output unit 132. The control unit 14 includes an information acquisition unit 141, a creation unit 142, a data acquisition unit 143, and a data output unit 144. The information processing device 1 may have all the configurations in one device, or may be distributed in a plurality of devices.

The communication unit 11 has a communication interface for transmitting / receiving data to / from a plurality of customer terminals 2 via the network N. The communication unit 11 inputs the selection information received from the customer terminal 2 to the information acquisition unit 141. Further, the communication unit 11 inputs the input data received from the customer terminal 2 to the data acquisition unit 143. Further, the communication unit 11 transmits the output data input from the data output unit 144 to the customer terminal 2.

The storage unit 12 includes a storage medium such as a ROM (Read Only Memory), a RAM (Random Access Memory), and a hard disk. The storage unit 12 stores a program executed by the control unit 14. Further, the storage unit 12 stores various information including information for specifying the artificial intelligence model selected by the customer in association with the customer who uses the customer terminal 2.

Further, the storage unit 12 stores a feature table including feature information indicating the features of each of the plurality of artificial intelligence models in association with each of the plurality of artificial intelligence models. The feature information is used so that the customer who wants to use the artificial intelligence model set can select one or more artificial intelligence models to be used by himself / herself. The feature information is, for example, information indicating the use of the artificial intelligence model, information indicating the type of data used as input data to the artificial intelligence model, information indicating the type of data output by the artificial intelligence model, or the artificial intelligence model. Is information indicating the frequency with which data is output.

FIG. 4 is a diagram showing an example of a feature table showing the features of the artificial intelligence model. In the feature table, the artificial intelligence model name and the feature of each artificial intelligence model are associated with each other. FIG. 4 shows the characteristics of a company that is good at analyzing artificial intelligence models, but the contents of the characteristics are not limited to this and are arbitrary.

In the case of an artificial intelligence model that can be used to estimate credit risk, the characteristic information can be, for example, the industry of the company for which credit risk is identified, the range of sales of the company for which credit risk is identified, and credit risk. The size of the company to be specified or the range of the location of the company to be specified for credit risk. In the case of an artificial intelligence model that can be used to support an investment in a financial asset, the characteristic information is, for example, the type of financial asset to be invested, the trading conditions of the financial asset to be invested, or the purpose of investment.

The artificial intelligence processing unit 13 executes processing using the artificial intelligence model set. In the example shown in FIG. 3, the information processing apparatus 1 has a plurality of artificial intelligence processing units 13 (artificial intelligence processing units 13-1, 13-2, ...) Corresponding to each of the plurality of customers, but information processing is performed. The number of artificial intelligence processing units 13 included in the device 1 is arbitrary. The artificial intelligence processing unit 13 has, for example, a processor and a memory, and executes arithmetic processing based on rules or mathematical formulas defined for each of a plurality of artificial intelligence models.

The artificial intelligence model group 131 selects one or more artificial intelligence models to be used from the plurality of artificial intelligence models based on the selection information input from the creation unit 142. The artificial intelligence model group 131 inputs the data output by each of the selected one or more artificial intelligence models to the output unit 132. FIG. 3 shows a case where the model M-a and the model M-d are used and the model M-b and the model M-c are not used among the plurality of artificial intelligence models possessed by the artificial intelligence model group 131. There is.

The output unit 132 outputs the result of the processing executed by the artificial intelligence model group 131 as output data. The output unit 132 may output one or more data output by one or more artificial intelligence models as it is, or may output data obtained by executing an operation based on one or more data. The data output by the output unit 132 is input to the data output unit 144.

The control unit 14 has, for example, a CPU (Central Processing Unit). The control unit 14 functions as an information acquisition unit 141, a creation unit 142, a data acquisition unit 143, and a data output unit 144 by executing the program stored in the storage unit 12.

The information acquisition unit 141 acquires various information from the customer terminal 2 via the communication unit 11. The information acquisition unit 141 acquires, for example, the selection information received from the customer terminal 2 by the communication unit 11. As described above, the selection information is information for identifying a plurality of selection models selected by the customer from a plurality of artificial intelligence models having different processing contents. The selection information is, for example, information including a model ID which is identification information for identifying a plurality of selection models. The information acquisition unit 141 inputs the acquired selection information to the creation unit 142.

For example, the information acquisition unit 141 presents a plurality of artificial intelligence models in association with the feature information to the customer terminal 2 which is an example of the external device, presents the feature information to the customer terminal 2, and then selects selection information from the customer terminal 2. get. By presenting the feature information to the customer terminal 2 in this way, the information acquisition unit 141 makes it easier for the customer using the customer terminal 2 to select the artificial intelligence model, but the information acquisition unit 141 presents the feature information to the customer terminal 2. The selection information may be acquired without presenting it to.

The information acquisition unit 141 further acquires the purpose of use of the artificial intelligence model set from the customer terminal 2 before presenting the plurality of artificial intelligence models to the customer terminal 2, and selects from a plurality of artificial intelligence model candidates based on the purpose of use. A plurality of artificial intelligence models may be presented to the customer terminal 2. For example, before presenting a plurality of artificial intelligence models to the customer terminal 2, the information acquisition unit 141 presents a plurality of purpose-of-use candidates, which are candidates for the purpose of use, to the customer terminal 2, and is selected from the plurality of purpose-of-use candidates. Acquire one or more purposes of use.

When the purpose of using the artificial intelligence model set is to investigate the credit risk of a company, the information acquisition unit 141 describes the type of business of the company for which the credit risk is specified and the range of sales of the company for which the credit risk is specified. , The size of the company for which credit risk is specified, or at least one of the range of locations of the company for which credit risk is specified is different. Present.

When the purpose of using the artificial intelligence model set is to invest in financial assets, the information acquisition unit 141 has at least one of the types of financial assets to be invested, the trading conditions of the financial assets to be invested, or the purpose of investment. A plurality of artificial intelligence models are presented to an external device in association with a plurality of different feature information. By presenting the feature information corresponding to the purpose of use in this way, the information acquisition unit 141 makes it easier for the customer to select an artificial intelligence model suitable for the purpose of use.

5 to 7 are diagrams showing an example of a screen presented to the customer terminal 2 by the information acquisition unit 141 so that the customer can select the artificial intelligence model. FIG. 5 is a screen for selecting a purpose of use. FIG. 5 shows a plurality of things that can be realized (that is, purpose of use) by using the artificial intelligence model set that can be provided by the information processing apparatus 1. The customer can determine the purpose of use by selecting the desired purpose of use from the plurality of displayed purposes of use and pressing "OK".

FIG. 6 is an example of a screen displayed on the customer terminal 2 after the purpose of use is selected. FIG. 6 shows a plurality of attributes for determining the characteristics of the artificial intelligence model set used by the customer, and when the customer selects the attribute, the screen shifts to the screen of FIG. 7. FIG. 6A is a screen displayed when "I want to investigate the credit risk of a company" is selected as the purpose of use on the screen shown in FIG. 5, and is displayed as an attribute related to the selection of the artificial intelligence model. "Industry", "sales", "scale", and "location" are shown. FIG. 6B is a screen displayed when "I want to invest in financial assets" is selected as the purpose of use on the screen shown in FIG. 5, and "Financial assets" is displayed as an attribute related to the selection of the artificial intelligence model. "Type", "Trading conditions", "Purpose of investment", and "Allowable risk" are shown.

FIG. 7 is an example of a screen displayed on the customer terminal 2 after the “industry” is selected on the screen of FIG. 6A. In FIG. 7, a plurality of industries to which each artificial intelligence model is suitable are shown together with the artificial intelligence model name. When "OK" is pressed after one or more industries are selected from a plurality of industries, the customer terminal 2 uses the information indicating the selected industry or the artificial intelligence model name as one of the selection information in the information processing device 1. Send and return to the screen of FIG. 6 (a). The customer terminal 2 repeats such processing, and outputs selection information indicating an artificial intelligence model selected in each attribute such as "industry", "sales", "scale", and "location" in the information processing device 1. Send to.

Returning to FIG. 3, the configuration and operation of each other unit of the control unit 14 that operates after the information acquisition unit 141 acquires the selection information will be described. The creation unit 142 creates an artificial intelligence model set including a plurality of selection models corresponding to the selection information notified from the information acquisition unit 141. The creation unit 142 constructs the artificial intelligence processing unit 13 in association with the customer ID for identifying the customer, for example. In the example shown in FIG. 3, the creating unit 142 creates the artificial intelligence processing unit 13-1 for the customer who has selected the artificial intelligence model M-a and the artificial intelligence model M-d. The creation unit 142 may create an artificial intelligence model set by storing the artificial intelligence model name selected by the customer in the storage unit 12.

The creation unit 142 may determine the charge for using the artificial intelligence model set based on the plurality of selection models indicated by the selection information. The creation unit 142 determines the price of the artificial intelligence model set, for example, by adding the unit prices of each of the plurality of artificial intelligence models constituting the artificial intelligence model set. The unit price of the artificial intelligence model is, for example, the number of data input to the artificial intelligence model, the number of data output by the artificial intelligence model, or the time required for the information processing apparatus 1 to execute processing using the artificial intelligence model. It is decided based on.

After the creation unit 142 creates the artificial intelligence model set, the data acquisition unit 143 acquires a plurality of input data to be input to the artificial intelligence model set, and inputs the acquired plurality of input data to the artificial intelligence model set. The data acquisition unit 143 acquires the input data from, for example, the customer terminal 2, but may acquire the input data from another terminal used by the customer.

The data acquisition unit 143 inputs input data to the artificial intelligence model group 131 of the artificial intelligence processing unit 13 corresponding to the customer who uses the customer terminal 2 that has transmitted the acquired input data among the plurality of artificial intelligence processing units 13. do. At this time, the data acquisition unit 143 selects one or more input data to be used by each of the plurality of selection models from the plurality of acquired input data, and inputs the selected one or more input data to each of the plurality of selection models. ..

Specifically, the input data includes type information indicating the type of the input data, and the data acquisition unit 143 specifies the type of the input data based on the type information and associates it with the type to store the storage unit 12. Input data to the artificial intelligence model stored in. The type of input data is represented by the difference in the nature of the data, such as sales data, number of employees data, address data, and the like. For example, the data acquisition unit 143 inputs sales data into an artificial intelligence model that uses sales data, and inputs employee number data into an artificial intelligence model that uses employee number data. In this way, the data acquisition unit 143 automatically selects the artificial intelligence model for inputting the input data according to the type of the input data, so that the customer who uses the artificial intelligence model set can distinguish each artificial intelligence model. There is no need to create input data.

The data output unit 144 outputs the output data output by the artificial intelligence model set to which the input data is input. Specifically, the data output unit 144 transmits the data input from the output unit 132 to the customer terminal 2 as output data via the communication unit 11. The customer terminal 2 displays the received output data.

The customer who uses the customer terminal 2 may change the selected artificial intelligence model if he / she is not satisfied with the content of the output data after seeing the displayed output data. When the information acquisition unit 141 receives a request to change the artificial intelligence model from the customer terminal 2, the information acquisition unit 141 displays the screens shown in FIGS. 6 and 7 on the customer terminal 2 to acquire new selection information, and the information acquisition unit 142 acquires new selection information. Creates an artificial intelligence model set consisting of newly selected artificial intelligence models. By repeating such processing, the customer can complete the desired artificial intelligence model set.

FIG. 8 is a flowchart showing the flow of processing executed by the information processing apparatus 1. FIG. 8 starts from the time when a customer who wants to use the artificial intelligence model set performs an operation for creating the artificial intelligence model set on the customer terminal 2.

First, as shown in FIG. 5, the information acquisition unit 141 presents a candidate for the purpose of use to the customer terminal 2 (S11). The information acquisition unit 141 acquires information indicating the purpose of use selected by the customer on the customer terminal 2 (S12). Subsequently, as shown in FIGS. 6 and 7, the information acquisition unit 141 presents a plurality of artificial intelligence model names selected based on the purpose of use to the customer terminal 2 together with their respective features (S13). The information acquisition unit 141 acquires selection information indicating an artificial intelligence model selected by the customer (S14).

Subsequently, the creation unit 142 creates an artificial intelligence model set based on the selection information acquired by the information acquisition unit 141 (S15). Then, the creation unit 142 stores one or more artificial intelligence model names constituting the customer's artificial intelligence model set in the storage unit 12 in association with the customer ID (S16).

After the artificial intelligence model set is created, the data acquisition unit 143 acquires input data from the customer (S17) and inputs the acquired input data to the artificial intelligence processing unit 13 (S18). The data output unit 144 outputs the output data output by the artificial intelligence processing unit 13 so that the customer can confirm it (S19).

After that, when the information acquisition unit 141 acquires the model change instruction for changing the artificial intelligence model set from the customer terminal 2, the information processing apparatus 1 updates the artificial intelligence model set by executing the processing after S13 again. do.

[Upgrade (update) of artificial intelligence model]
The information processing device 1 may update the artificial intelligence model at any timing. When the information processing device 1 updates the artificial intelligence model, the artificial intelligence model set including the updated artificial intelligence model is also updated in conjunction with it. In this way, when the artificial intelligence model set is composed of a plurality of artificial intelligence models and it is necessary to update a part of the artificial intelligence model set, the artificial intelligence model set is updated without updating each of them. The target artificial intelligence model may be updated. Therefore, it is possible to suppress the cost required for upgrading the version of the artificial intelligence model set.

Some customers do not want to upgrade the artificial intelligence model. Therefore, when the artificial intelligence model is updated, the information processing device 1 (for example, the information acquisition unit 141) updates the artificial intelligence model for the customer who is using the artificial intelligence model set including the updated artificial intelligence model. You may notify the name and update details and accept instructions to perform the update of the artificial intelligence model set. The information processing device 1 may accept an instruction to update a part of the updated artificial intelligence models. By operating the information processing device 1 in this way, the customer can update the artificial intelligence model after determining the necessity.

[Effect of information processing device 1]
As described above, the information processing apparatus 1 corresponds to the information acquisition unit 141 for acquiring selection information for specifying a plurality of selection models selected from a plurality of artificial intelligence models having different processing contents, and the selection information. It has a creation unit 142, which creates an artificial intelligence model set including a plurality of selection models. With the information processing device 1 configured in this way, customers who want to perform various analyzes using the artificial intelligence model set can use the artificial intelligence model set customized based on their own needs. Will be possible.

When a large number of artificial intelligence models constituting the artificial intelligence model set are prepared for each attribute, the information processing apparatus 1 can create a large number of types of artificial intelligence model sets. For example, if there are 10 types of artificial intelligence models for each of the five attributes, the information processing apparatus 1 can create an artificial intelligence model set of 100,000 patterns. Therefore, it is unlikely that an artificial intelligence model set for one customer will match an artificial intelligence model set for another customer, and customers may try to differentiate themselves from their competitors by choosing an artificial intelligence model. It will be possible.

In addition, when creating a customized artificial intelligence model, POC (Proof of Concept) is performed to confirm whether the desired artificial intelligence model is completed, and then the artificial intelligence model is used in the actual environment. It was necessary to conduct a proof-of-concept operation in parallel with the existing system to confirm whether it could withstand. There is a problem that it takes a long time and a high cost to carry out this POC and the proof-of-concept.

On the other hand, when the information processing apparatus 1 is used, it is not necessary to create a new artificial intelligence model for each customer in order to provide a different artificial intelligence model set for each customer. Therefore, since the POC time to be performed when creating or updating the artificial intelligence model can be significantly shortened, the information processing apparatus 1 suppresses the cost for using the artificial intelligence model. Can be done.

Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes can be made within the scope of the gist. be. For example, all or part of the device can be functionally or physically distributed / integrated in any unit. Also included in the embodiments of the present invention are new embodiments resulting from any combination of the plurality of embodiments. The effect of the new embodiment produced by the combination has the effect of the original embodiment together.

1 Information processing device 2 Customer terminal 11 Communication unit 12 Storage unit 13 Artificial intelligence processing unit 14 Control unit 131 Artificial intelligence model group 132 Output unit 141 Information acquisition unit 142 Creation unit 143 Data acquisition unit 144 Data output unit

Claims (9)

An information acquisition unit that acquires selection information for identifying multiple selection models selected from multiple artificial intelligence models with different processing contents, and an information acquisition unit.
A creation unit that creates an artificial intelligence model set including the plurality of selection models corresponding to the selection information, and a creation unit.
A data acquisition unit that acquires a plurality of input data to be input to the artificial intelligence model set and inputs the acquired plurality of input data to the artificial intelligence model set.
A data output unit that outputs output data output by the artificial intelligence model set to which the input data is input, and a data output unit.
Information processing device with. The data acquisition unit selects one or more input data to be used by each of the plurality of selection models from the acquired plurality of input data, and inputs the selected one or more input data to each of the plurality of selection models. do,
The information processing apparatus according to claim 1. The information acquisition unit presents the plurality of artificial intelligence models to an external device in association with feature information indicating the characteristics of each of the plurality of artificial intelligence models, presents the feature information to the external device, and then presents the feature information to the external device. Obtain the selection information from
The information processing apparatus according to claim 1 or 2. The information acquisition unit further acquires the purpose of use of the artificial intelligence model set from the external device, and presents the plurality of artificial intelligence models selected from a plurality of artificial intelligence model candidates based on the purpose of use to the external device. do,
The information processing apparatus according to claim 3. Before presenting the plurality of artificial intelligence models to the external device, the information acquisition unit presents a plurality of purpose-of-use candidates, which are candidates for the purpose of use, to the external device, and selects from the plurality of purpose-of-use candidates. To obtain one or more of the above-mentioned purposes of use,
The information processing apparatus according to claim 4. The information acquisition department identifies the type of business of the company for which credit risk is specified, the range of sales of the company for which credit risk is specified, the size of the company for which credit risk is specified, or the credit risk. Presenting the plurality of artificial intelligence models to the external device in association with the plurality of feature information having at least one of the different range of locations of the target company.
The information processing apparatus according to claim 4 or 5. The information acquisition unit associates the plurality of artificial intelligence models with the plurality of characteristic information having different types of financial assets to be invested, trading conditions of the financial assets to be invested, or at least one of the purposes of investment. Present to an external device,
The information processing apparatus according to claim 4 or 5. The creation unit determines a fee for using the artificial intelligence model set based on the plurality of selection models indicated by the selection information.
The information processing apparatus according to any one of claims 1 to 7. Computer runs,
A step to acquire selection information for identifying a plurality of selection models selected from a plurality of artificial intelligence models having different processing contents, and a step.
A step of creating an artificial intelligence model set including the plurality of selection models corresponding to the selection information, and
A step of acquiring a plurality of input data to be input to the artificial intelligence model set and inputting the acquired plurality of input data to the artificial intelligence model set.
A step of outputting the output data output by the artificial intelligence model set to which the input data is input, and a step of outputting the output data.
Information processing method with.

Images