Search

KR-102963357-B1 - METHOD FOR GENERATING ADVERTISEMENT CONTENT AND SYSTEM THEREFOR

KR102963357B1KR 102963357 B1KR102963357 B1KR 102963357B1KR-102963357-B1

Abstract

A method for creating advertising content and a system thereof are provided. A method for generating advertising content according to one embodiment of the present disclosure is a method performed by a computing system, comprising: receiving keyword-based brand evaluation data from a user terminal; inputting the brand evaluation data into a first artificial intelligence model, which is a pre-trained large-scale language model, and obtaining a brand evaluation prompt; inputting the brand evaluation prompt into a second artificial intelligence model, which is a pre-trained image generation model, and obtaining a brand evaluation image; integrating the brand evaluation prompt and the brand evaluation image and generating brand recognition data for a consumer's brand; inputting the brand recognition data into the first artificial intelligence model and generating an advertising prompt template for generating an advertising image; generating gap analysis data between the brand recognition data and the target data using the brand recognition data and the target data aimed at by the brand; generating an advertising prompt corrected by the advertising prompt template using each component included in the advertising prompt template and the gap analysis data; inputting the advertising prompt, brand context data expressing the product of the brand, and pre-set advertising situation data into the first artificial intelligence model and generating a text advertising material; and inputting the advertising prompt, the brand context data, and the advertising situation data It may include the steps of inputting into the second artificial intelligence model and generating an image advertising material, and integrating the text advertising material and the image advertising material and automatically generating advertising content.

Inventors

  • 김대희
  • 선형조
  • 양승만
  • 지준형

Assignees

  • 주식회사 드래프타입

Dates

Publication Date
20260511
Application Date
20250604

Claims (9)

  1. In a method performed by a computing system, A step of receiving keyword-based brand evaluation data from a user terminal; A step of inputting the above brand evaluation data into a first artificial intelligence model, which is a pre-trained large-scale language model, and obtaining a brand evaluation prompt; A step of inputting the above-mentioned brand evaluation prompt into a second artificial intelligence model, which is a pre-trained image generation model, and acquiring a brand evaluation image; A step of integrating the above-mentioned brand evaluation prompt and the above-mentioned brand evaluation image, and generating brand awareness data regarding the consumer's brand; A step of inputting the above brand recognition data into the above first artificial intelligence model and generating an advertising prompt template for generating an advertising image; A step of generating gap analysis data between the brand recognition data and the target data using the brand recognition data and the target data aimed for by the brand; A step of generating an advertising prompt corrected by using each component included in the advertising prompt template and the gap analysis data; A step of inputting the above advertising prompt, brand context data expressing the product of the above brand, and pre-set advertising situation data into the above first artificial intelligence model, and generating text advertising material; The step of inputting the above advertising prompt, the above brand context data, and the above advertising situation data into the second artificial intelligence model and generating image advertising material; and The method includes the step of integrating the text ad creative and the image ad creative and automatically generating ad content, The step of obtaining the above-mentioned brand evaluation image is, A step of transmitting a first brand evaluation image generated based on the brand evaluation prompt to the user terminal; A step of receiving feedback data regarding the first brand evaluation image from the user terminal; and The method includes the step of adopting the first brand evaluation image as the brand evaluation image only when the feedback data is positive feedback indicating that the first brand evaluation image reflects the perception of the brand, and The step of generating the above brand recognition data is, A step of inputting the above-mentioned brand evaluation image into a pre-trained Vision Language Model and obtaining image description text data in which visual elements for pre-defined concept types are combined; A step of performing clustering on each component included in the image description text data and each component included in the brand evaluation prompt according to predefined semantic classification criteria; A step of assigning weights to the clusters based on the frequency of occurrence of keywords included in the clusters generated as a result of performing the clustering above; and The method includes the step of adopting clusters with weights greater than or equal to a threshold as the brand recognition data. The step of generating the above gap analysis data is, A step of inputting the above target data into the above first artificial intelligence model and classifying the components included in the above target data according to the above predefined semantic classification criteria; A step of deriving a gap element between a first component included in the brand recognition data corresponding to the semantic classification criteria and a second component included in the target data corresponding to the semantic classification criteria; and The above gap elements include the step of generating structured data according to the above semantic classification criteria, and The step of generating the above advertising prompt is, The step of inputting the above advertising prompt template into the above first artificial intelligence model and classifying each component included in the above advertising prompt template according to the above predefined semantic classification criteria; and A step comprising adjusting the components included in the advertising prompt template corresponding to the first classification criterion generated according to the result of the above classification according to the data expressing the gap elements included in the gap analysis data corresponding to the first classification criterion. How to create ad content.
  2. In Article 1, The above receiving step is, A step of further receiving user profile data from the user terminal, wherein the user profile data includes at least one of information regarding the user's age group, gender, and occupation group; and The method includes the step of generating segment tags by mapping the above brand evaluation data to the above user profile data, The step of generating the above brand recognition data is, The method includes the step of integrating the brand evaluation prompt and the brand evaluation image for each segment tag and generating brand recognition data for each segment tag. The above target data is, Distinguished by the above segment tags, How to create ad content.
  3. In Article 1, The above-mentioned first artificial intelligence model is, The above brand evaluation data is mapped to coordinates in the embedding space, and a first embedding vector is generated, Extract a second embedding vector having a similarity to the first embedding vector above a threshold value, and The above second embedding vector is classified according to a predefined conceptual type, and trained to generate the brand evaluation prompt by combining visual elements corresponding to embedding vectors for the above concept types, How to create ad content.
  4. delete
  5. In Article 1, The above brand recognition data is, Information including the above semantic classification criteria, clusters corresponding to the above semantic classification criteria, a set of keywords included in the clusters, and the frequency of occurrence of the set of keywords, How to create ad content.
  6. In Article 1, The above brand recognition data is, The above weights are reflected for the above clusters for each segment tag, and The above segment tag is, including at least one of the user's age group, gender, and occupation group, How to create ad content.
  7. delete
  8. delete
  9. Communication interface; Memory where a computer program is loaded; and The computer program described above includes one or more processors on which it is executed, The above computer program is, An operation of receiving keyword-based brand evaluation data from a user terminal; The operation of inputting the above brand evaluation data into a first artificial intelligence model, which is a pre-trained large-scale language model, and obtaining a brand evaluation prompt; The operation of inputting the above-mentioned brand evaluation prompt into a second artificial intelligence model, which is a pre-trained image generation model, and acquiring a brand evaluation image; An operation to integrate the above-mentioned brand evaluation prompt and the above-mentioned brand evaluation image, and to generate brand awareness data regarding the consumer's brand; The operation of inputting the above brand recognition data into the above first artificial intelligence model and generating an advertising prompt template for generating an advertising image; An operation to generate gap analysis data between the brand recognition data and the target data using the brand recognition data and the target data aimed for by the brand; The operation of generating an advertising prompt corrected by the advertising prompt template using each component included in the advertising prompt template and the gap analysis data; The operation of inputting the above advertising prompt, brand context data expressing the product of the above brand, and pre-set advertising situation data into the above first artificial intelligence model, and generating text advertising material; The operation of inputting the above advertising prompt, the above brand context data, and the above advertising situation data into the above second artificial intelligence model and generating image advertising material; and Includes instructions that integrate the text ad creative and the image ad creative and perform the operation of automatically generating ad content, The operation of acquiring the above-mentioned brand evaluation image is, The operation of transmitting a first brand evaluation image generated based on the brand evaluation prompt to the user terminal; The operation of receiving feedback data regarding the first brand evaluation image from the user terminal; and The method includes the operation of adopting the first brand evaluation image as the brand evaluation image only when the feedback data is positive feedback indicating that the first brand evaluation image reflects the perception of the brand, and The operation of generating the above brand recognition data is, The operation of inputting the above-mentioned brand evaluation image into a pre-trained Vision Language Model and obtaining image description text data in which visual elements corresponding to pre-defined concept types are combined; An operation to perform clustering on each component included in the image description text data and each component included in the brand evaluation prompt, according to predefined semantic classification criteria; An operation of assigning weights to the clusters based on the frequency of occurrence of keywords included in the clusters generated as a result of performing the clustering above; and The operation includes adopting clusters with weights greater than or equal to a threshold as the brand recognition data, The operation of generating the above gap analysis data is, The operation of inputting the above target data into the above first artificial intelligence model and classifying the components included in the above target data according to the above predefined semantic classification criteria; An operation to derive a gap element between a first component included in the brand recognition data corresponding to the semantic classification criteria and a second component included in the target data corresponding to the semantic classification criteria; and It includes the operation of generating the above gap elements into structured data according to the above semantic classification criteria, and The operation of generating the above advertising prompt is, The operation of inputting the above-mentioned advertising prompt template into the above-mentioned first artificial intelligence model and classifying each component included in the above-mentioned advertising prompt template according to the above-mentioned predefined semantic classification criteria; and A method comprising adjusting the components included in the advertising prompt template corresponding to the first classification criterion generated according to the result of the above classification, according to data expressing gap elements included in the gap analysis data corresponding to the first classification criterion. Advertising content creation system.

Description

Method for Generating Advertising Content and System Thereof The present disclosure relates to a method and system for generating advertising content. More specifically, it relates to a method and system for generating advertising content that reflects the gap between brand perception data and brand target data of consumers encountering the advertisement. Existing advertising production systems faced difficulties in ensuring efficiency and consistency because the entire process, ranging from brand image research to strategy formulation and ad creative production, was fragmented and based on manual procedures. In particular, brand awareness surveys were primarily conducted using keyword selection or fragmentary sentence responses, making it difficult to fully grasp the situational context, emotional atmosphere, and visual imagery that consumers actually perceive. Furthermore, even when respondents used the same keywords, differences in interpretation limited accurate comparison and analysis of brand image, leading to the problem of failing to provide reliable data for strategic advertising messages or creative composition. Furthermore, even when advertising strategies are established based on brand image research results, an interpretive gap exists between the strategy and the production stages, making it difficult to maintain brand consistency and leading to issues where deliverables vary depending on the individual designer's subjectivity. In particular, producing materials for diverse formats (social media, banners, videos, etc.) requires excessive time and cost, and there are practical limitations in designing sophisticated strategies and customizing materials to reflect differing brand perceptions across target audiences. FIG. 1 is a system configuration diagram for explaining the configuration and operation of an advertising content provision system according to some embodiments of the present disclosure. FIG. 2 is a flowchart for explaining the operation of an advertising content generation method according to some embodiments of the present disclosure. FIG. 3 is a diagram illustrating a method for receiving brand evaluation data from a user terminal according to some embodiments of the present disclosure. FIG. 4 is a drawing for explaining a method of obtaining a brand evaluation image according to some embodiments of the present disclosure. FIG. 5 is a detailed flowchart for explaining the detailed operation of an advertising content generation method according to some embodiments of the present disclosure, described with reference to FIG. 2. FIG. 6 is a detailed flowchart for explaining the detailed operation of an advertising content generation method according to some embodiments of the present disclosure, described with reference to FIG. 2. FIG. 7 is a drawing for explaining a method for generating brand recognition data according to some embodiments of the present disclosure. FIG. 8 is a drawing for illustrating an example of brand recognition data according to some embodiments of the present disclosure. FIG. 9 is a drawing for illustrating an example of brand recognition data according to some embodiments of the present disclosure. FIG. 10 is a detailed flowchart for explaining the detailed operation of an advertising content generation method according to some embodiments of the present disclosure, described with reference to FIG. 2. FIG. 11 is a drawing for illustrating an example of target data that a brand aims for according to some embodiments of the present disclosure. FIG. 12 is a drawing showing an example of a gap element between brand recognition data and target data according to some embodiments of the present disclosure. FIG. 13 is a drawing illustrating an example of gap analysis data between brand recognition data and target data according to some embodiments of the present disclosure. FIG. 14 is a detailed flowchart for explaining the detailed operation of an advertising content generation method according to some embodiments of the present disclosure, described with reference to FIG. 2. FIG. 15 is a drawing for explaining a method for generating an advertising prompt according to some embodiments of the present disclosure. FIG. 16 is a drawing for illustrating an example of advertising content generated according to some embodiments of the present disclosure. FIG. 17 illustrates an exemplary computing device capable of implementing systems according to some embodiments of the present disclosure. Hereinafter, various embodiments of the present disclosure will be described in detail with reference to the attached drawings. The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the attached drawings. However, the technical concept of the present disclosure is not limited to the following embodiments but can be implemented in various different forms. The follo