Device and method

WO2026083561A1PCT designated stage Publication Date: 2026-04-23NTT DOCOMO INC
View PDF 5 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
NTT DOCOMO INC
Filing Date
2024-10-17
Publication Date
2026-04-23

Smart Images

  • Figure JP2024037064_23042026_PF_FP_ABST
    Figure JP2024037064_23042026_PF_FP_ABST
Patent Text Reader

Abstract

The purpose of the present invention is to provide a device and a method for generating personas capable of meeting various needs. In a persona generation device 100 according to the present disclosure, a virtual persona generation unit 103 projects a requested persona requested by an operator into a latent space. For example, the virtual persona generation unit 103 projects a requested persona encoded using a VAE into the latent space. The virtual persona generation unit 103 then acquires at least one piece of information contained in the latent space on the basis of requested persona information. This at least one piece of information is at least one of the continuous values (points) in the latent space. The virtual persona generation unit 103 then generates a virtual persona on the basis of the at least one piece of information. That is, the virtual persona generation unit 103 performs decoding using a VAE to generate information constituting a persona from at least one point in the latent space. This information is the virtual persona.
Need to check novelty before this filing date? Find Prior Art

Claims

1. An apparatus comprising: a latent space processing unit that projects a requested persona from an operator into a latent space; an acquisition unit that acquires at least one piece of information contained in the latent space based on the requested persona; and a virtual persona generation unit that generates a virtual persona based on the at least one piece of information.

2. The apparatus according to claim 1, wherein the acquisition unit acquires information in the latent space that is within a range based on the requested persona.

3. The apparatus according to claim 2, wherein the range is defined as a region of a reference radius centered on the requested persona information, and the acquisition unit acquires more of the information from the range closest to the center.

4. An apparatus comprising: a latent space processing unit that projects a requested persona requested by an operator into a latent space; a calculation unit that determines the ratio of the overlapping portion between the requested persona region and at least one base persona region based on the overlapping region between the requested persona region and at least one base persona region; and a virtual persona generation unit that generates a virtual persona based on the ratio of the base persona region in the overlapping portion.

5. The apparatus according to claim 4, wherein the virtual persona generation unit generates a virtual persona by adding the percentage to each item constituting the base persona in the base persona area.

6. The apparatus according to claim 4, wherein the base persona region is generated by projecting a plurality of user profiles into a latent space, grouping them, and based on the information of the grouped user profiles, and at least one base persona region is labeled using a predetermined base persona.

7. A decision-making device comprising: a prompt generation unit that generates prompt instruction information to cause a generating AI to determine the compatibility between a user assumed by the virtual persona and the decision target, using a virtual persona and a decision target generated by the device described in claim 1; a transmission unit that transmits the prompt instruction information to the generating AI; and a response result acquisition unit that acquires a response result from the generating AI.

8. The determination device according to claim 7, wherein the object of determination is an advertisement, and the generating AI is made to determine whether or not a user assumed by the virtual persona will respond to the advertisement.

9. A method for generating a persona in a device, comprising: a latent space processing step of projecting a requested persona requested by an operator into a latent space; an acquisition step of acquiring at least one piece of information contained in the latent space based on the requested persona information; and a virtual persona generation step of generating a virtual persona based on the at least one piece of information.

10. A determination method for generating prompts in a device, comprising: a prompt generation step of generating prompt instruction information that causes a generating AI to determine the suitability of a user assumed by the virtual persona with the determination target, using a pre-configured virtual persona and a determination target; a transmission step of transmitting the prompt instruction information to the generating AI; and a response result acquisition step of obtaining a response result from the generating AI.

Citation Information

Patent Citations

  • Management device for a plurality of identities belonging to same person, authentication method, control method and method for utilization

    JP2009169921A

  • Consumer-driven advertising system

    JP2014522015A

  • Information providing device, information providing method, and information providing program

    JP2023000314A

  • Information processing apparatus and program

    JP2023044741A

  • Information processing device, information processing method, and information processing program

    JP2024013501A