Method and system for exposing advertisement to result obtained using artificial intelligence agent in artificial intelligence augmented search

Abstract

The present invention relates to a method and a system for exposing an advertisement to a result obtained using an artificial intelligence agent in an artificial intelligence augmented search. The method is characterized by comprising the steps in which: an advertiser terminal registers an advertiser artificial intelligence agent in an artificial intelligence agent storage of a search engine platform; a user terminal transmits a query to an interface module of the search engine platform; an advertiser artificial intelligence agent allocation module of the search engine platform calls an advertiser artificial intelligence agent having high relevance to the query; a result generation artificial intelligence agent module of the search engine platform generates a response including advertisement information by using the called artificial intelligence agent; and the result generation artificial intelligence agent module of the search engine platform generates a final search result by combining the response including the advertisement information with an existing search result of an external data agent.

Description

Method and system for displaying advertisements on results utilizing AI agents in AI augmented search

[0001] The present invention relates to a method and system for displaying advertisements in results utilizing an artificial intelligence agent in artificial intelligence augmented search, and more specifically, to an advertisement display method and system that naturally integrates and provides search results and advertising information in response to a user query through an artificial intelligence agent directly registered by a domain owner (company, institution, etc.).

[0002]

[0003] Generally, advertisements provided by search engines (keyword ads, banner ads) are displayed separately from search results, which leads to: i) hindering user experience; ii) causing inconvenience due to the display of irrelevant or excessive advertisements; and iii) difficulties in updating advertiser information in real time.

[0004] Meanwhile, a Large Language Model (LLM) is an artificial intelligence model capable of understanding and generating language by learning from vast amounts of text data. As one of the core technologies enabling conversational AI, it performs various Natural Language Processing (NLP) tasks to understand the meaning and context of a user's question and generate specific and contextual responses.

[0005] However, i) it is difficult to fully reflect the latest information or specialized data from specific domains (companies, universities, institutions, etc.) with simple LLM alone, and ii) from the perspective of search engine operators, how to use advertising as a revenue model in the newly changing AI search paradigm is emerging as a key challenge, so research and development to solve this is necessary.

[0006]

[0007] The problem that the present invention aims to solve is to provide a method and system for displaying advertisements in results utilizing an artificial intelligence agent in artificial intelligence augmented search.

[0008] The problems that the present invention aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below.

[0009]

[0010] The method of the present invention for solving the above-mentioned problem may be characterized by comprising: a step in which an advertiser terminal registers an advertiser AI agent in the AI ​​agent storage of a search engine platform; a step in which a user terminal transmits a query to an interface module of the search engine platform; a step in which an advertiser AI agent with high relevance to the query is called from an advertiser AI agent allocation module of the search engine platform; a step in which a response including advertising information is generated using the called AI agent from a result generation AI agent module of the search engine platform; and a step in which a final search result is generated by combining the response including advertising information with an existing search result of an external data agent from a result generation AI agent module of the search engine platform.

[0011] The advertiser AI agent allocation module of the above-mentioned search engine platform may be characterized by selecting an advertiser AI agent highly relevant to the query using at least one of vector embedding or keyword analysis technology.

[0012] The result generation AI agent model of the above-mentioned search engine platform may be characterized by generating a response containing advertising information in at least one of the following methods: real-time data collection through an external database using a called advertiser AI agent, document search based on search augmentation generation technology, response generation through a natural language processing model, and a workflow including a natural language processing model.

[0013] The result generation AI agent model of the above-mentioned search engine platform may be characterized by using a called advertiser AI agent to determine the similarity between a query embedding and an ad document embedding registered by the advertiser AI agent, selecting ad documents with high similarity to the query, and using the called advertiser AI agent to input refined ad information and an ad placement guide into a natural language processing model as a prompt to generate a response containing ad information.

[0014] The result-generating artificial intelligence agent model of the above-mentioned search engine platform may be characterized by summarizing selected advertising documents and refining them into advertising information.

[0015] The result generation artificial intelligence agent module of the above-mentioned search engine platform may be characterized by generating a final search result by contextually combining existing search results from an external data agent with a response containing advertising information, and the external data agent may generate existing search results for a query from the external web.

[0016] The artificial intelligence ad inspection module of the search engine platform may further include the step of transmitting a filtered response to the user terminal after determining falsehood, illegality, and excessiveness in the final search result.

[0017] The artificial intelligence ad inspection module of the above-mentioned search engine platform may be characterized as a natural language processing model that learns the final search result as a feature value and the response filtered from the final search result to include false, illegal, and excessive advertising expressions as a label value.

[0018] The ad revenue module of the search engine platform may further include a step of claiming ad revenue from the advertiser terminal based on the exposure of ad information of the response provided to the user terminal.

[0019] The system of the present invention for solving the aforementioned problem comprises: a search engine platform; an advertiser terminal that registers an advertiser AI agent in the search engine platform; and a user terminal that inputs a query into the search engine platform and receives results using the advertiser AI agent. The search engine platform may be characterized by including: an AI agent storage in which an advertiser AI agent is registered by the advertiser terminal; an interface module that receives a query input by the user terminal; an advertiser AI agent allocation module that calls an advertiser AI agent with high relevance to the query; a result generating AI agent module that generates a response including advertising information for the query using the called AI agent; and a result generating AI agent module that generates a response including advertising information for the query using the called AI agent, and combines the response including advertising information with an existing search result of an external data agent to generate a final search result.

[0020]

[0021] Unlike keywords or banner advertisements in general search engines, this invention incorporates advertisements into the actual answers to user queries, thereby reducing resistance and offering the advantage of naturally integrating advertisements and search results.

[0022] In addition, the present invention has the advantage of enabling autonomous, real-time updates by advertisers, as advertisers can receive immediate information updates (latest news, new product launches, price changes, events, etc.) through an artificial intelligence agent.

[0023] In addition, the present invention has the advantage of having a revenue model suitable for the AI ​​era, as search engine operators can secure stable AI search advertising revenue through billing, license fees, etc., based on the registration and exposure of AI agents.

[0024] Furthermore, the present invention has the advantage of improving the user experience (UX) by increasing search satisfaction through the exposure of advertisements that match the context of the query, rather than indiscriminate or irrelevant advertisements.

[0025] The effects of the present invention are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below.

[0026]

[0027] FIG. 1 is a configuration diagram showing a system for performing the method of the present invention.

[0028] FIG. 2 is a flowchart showing the process in which the method of the present invention is performed.

[0029] FIG. 3 is a conceptual diagram showing the operation of the method of the present invention.

[0030]

[0031] The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the present invention, and the present invention is defined only by the scope of the claims.

[0032] The terms used in this specification are for describing the embodiments and are not intended to limit the invention. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. The terms "comprises" and / or "comprising" used in this specification do not exclude the presence or addition of one or more other components in addition to the components mentioned. Throughout the specification, the same reference numerals refer to the same components, and "and / or" includes each of the mentioned components and all combinations of one or more. Although terms such as "first," "second," etc., are used to describe various components, these components are not limited by these terms. These terms are used merely to distinguish one component from another. Therefore, the first component mentioned below may be the second component within the technical scope of the invention.

[0033] Unless otherwise defined, all terms used herein (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0034]

[0035] Hereinafter, the system (100) and method (200) of the present invention will be described with reference to the drawings. FIG. 1 is a configuration diagram showing a system for performing the method of the present invention, FIG. 2 is a flowchart showing a process in which the method of the present invention is performed, and FIG. 3 is a conceptual diagram showing the operation of the method of the present invention.

[0036] In describing the system (100) and method (200) of the present invention below, the natural language processing model can be interpreted as an artificial intelligence model capable of learning text data to understand and generate language, and, for example, may be an LLM, sLLM, SLM, etc., but is not limited thereto and should be broadly interpreted as a comprehensive concept that includes all engines that perform various natural language processing (NLP) tasks.

[0037]

[0038] The system (100) of the present invention may include a search engine platform (10); an advertiser terminal (20) that registers an advertiser artificial intelligence agent on the search engine platform (10); and a user terminal (30) that inputs a query on the search engine platform (10) and receives results using the advertiser artificial intelligence agent.

[0039] The search engine platform (10) may include: an AI agent storage (11) in which an advertiser AI agent is registered by an advertiser terminal (20); an interface module (12) that receives a query from a user terminal (30); an advertiser AI agent allocation module (13) that calls an advertiser AI agent that is highly relevant to the query; a result generation AI agent module (14) that uses the called AI agent to generate a response containing advertising information for the query and combines the response containing advertising information with the existing search results of an external data agent to generate a final search result; an AI ad inspection module (15) that determines falsehood, illegality, and excessiveness in the search results and transmits the filtered response to the user terminal (30); and an ad revenue module (16) that charges ad revenue to the advertiser terminal (20) based on the exposure of ad information in the response provided to the user terminal (30).

[0040] Accordingly, the method (200) of the present invention comprises the steps of: registering an advertiser AI agent in the AI ​​agent storage (11) of the search engine platform (10) by an advertiser terminal (20) (210); transmitting a query to the interface module (12) of the search engine platform (10) by a user terminal (30) (220); calling an advertiser AI agent that is highly relevant to the query in the advertiser AI agent allocation module (13) of the search engine platform (10) (230); generating a response containing advertising information using the called AI agent in the result generation AI agent module (14) of the search engine platform (10) (240); and generating a final search result by combining the response containing advertising information with the existing search result of an external data agent in the result generation AI agent module (14) of the search engine platform (10) (250). The method may include: a step (260) in which an artificial intelligence ad inspection module (15) of a search engine platform (10) determines falsehood, illegality, and excessiveness in the final search result and transmits the filtered response to a user terminal (30); and a step (270) in which an ad revenue module (16) of a search engine platform (10) claims ad revenue to an advertiser terminal (20) based on the exposure of ad information of the response provided to the user terminal (30).

[0041] The step (210) in which the advertiser terminal (20) registers an advertiser AI agent in the AI ​​agent storage (11) of the search engine platform (10) may be a step in which the advertiser terminal (20) registers an advertiser AI agent in the search engine platform (10) for promotion, which is capable of collecting real-time data through an external database or an internal database of the advertiser under the control of the advertiser terminal (20), searching for documents (advertisement documents) based on Retrieval Augmented Generation (RAG) technology, generating responses integrated with natural language processing models, and processing workflows including natural language processing models. Accordingly, the advertiser AI agent can collect and learn data in real-time under the control of the advertiser terminal (20), and thus can be updated according to the advertiser's intentions.

[0042] The step (220) in which a user terminal (30) transmits a query to an interface module (12) of a search engine platform (10) may be a step in which a user who wishes to use the search engine inputs a query to be searched through their terminal. In this case, the query may be composed only of natural language, but is not limited thereto, and the query may include input forms other than natural language text, such as images or videos.

[0043] For example, a query can be composed of a sentence such as, "Please provide a professional review of a new whitening cosmetic product or overseas price trends."

[0044] The step (230) of calling an advertiser AI agent that is highly relevant to the query in the advertiser AI agent allocation module (13) of the search engine platform (10) may be a step of selecting and calling an advertiser AI agent corresponding to the user's query so that appropriate advertising information is included in the response to the user's query in the search engine platform.

[0045] In this case, the advertiser AI agent allocation module (13) of the search engine platform (10) can select an advertiser AI agent that is highly relevant to the query by using at least one of vector embedding or keyword analysis technology.

[0046] For example, the query "Tell me professional reviews or overseas price trends for new whitening cosmetic products" is tokenized to extract keywords such as "whitening cosmetic" or generate embedding vectors for them, and text from advertising documents collected by the advertiser AI agent is tokenized to extract keywords such as "whitening cosmetic" or generate embedding vectors for them. By determining the similarity of keywords or embedding vectors using methods such as cosine similarity or K-NN similarity, an advertiser AI agent with high similarity can be selected.

[0047] The step (240) of generating a response containing advertising information using an artificial intelligence agent called from the result generation artificial intelligence agent module (14) of the search engine platform (10) may be a step of generating a response using an artificial intelligence agent selected so that advertising information is included in the response to the user's query.

[0048] In detail, the result generating artificial intelligence agent module (14) can generate a response containing advertising information in at least one of the following ways: collecting real-time data through an external database using a called advertiser artificial intelligence agent, searching for advertising documents based on search augmentation generation technology (RAG), generating a response through a natural language processing model, and a workflow including a natural language processing model.

[0049] In this case, the advertiser artificial intelligence agent is controlled by the advertiser terminal (20) to collect data related to the advertisement and search for documents, and since a natural language processing model can be trained, the response containing the advertisement information may be a response that reflects the advertiser's advertising intent.

[0050] For example, the result generating AI agent module (14) can determine the similarity between the query embedding and the ad document embedding registered by the advertiser AI agent using the called advertiser AI agent, and select an ad document that is highly similar to the query.

[0051] In this case, the advertising document registered by the advertiser AI agent may be a document searched by the advertiser AI agent based on real-time data collection through an external (or internal, within the advertising agency) database or search augmentation generation technology under the control of the advertiser terminal (20), and may include advertising information, but is not limited thereto.

[0052] In detail, the result generating AI agent module (14) can select advertising documents by calculating the cosine similarity between natural language query embeddings and the embeddings of advertising documents registered by the advertiser AI agent. In this case, user query embeddings can be vectorized using a model utilizing a transformer. Advertising document embeddings are documents provided by the advertiser and may include brand information, promotional benefits, product specifications, etc., and can be vectorized by splitting the bundled text (Chunking) and embedding each document to build an index.

[0053] Next, the result generating artificial intelligence agent module (14) can calculate the similarity between the query embedding and the advertisement document embedding and select the advertisement document whose similarity is greater than or equal to a threshold.

[0054] For example, when the query is "tell me professional reviews of new whitening cosmetic products or overseas price trends," the result generating AI agent module (14) generates an embedding vector for tokenized "whitening cosmetic," etc., and after splitting the bundled text of the advertising document in the called advertiser AI agent, analyzes and infers the meaning and intent, and sets a "whitening cosmetic brand index" for advertising documents containing content (text) about whitening cosmetic brands, a "whitening cosmetic promotion index" for advertising documents containing content (text) about whitening cosmetic promotions, and a "whitening cosmetic product specification index" for advertising documents containing content (text) about whitening cosmetic product specifications, and then vectorizes them, and then determines the similarity between the query embedding and the advertising document embedding to select a specific advertising document.

[0055] Next, the result generating AI agent module (14) can generate a response containing advertising information by using the called advertiser AI agent to input a query and selected advertising documents, refined advertising information, and an advertising placement guide as prompts into a natural language processing model (e.g., a generative AI model such as ChatGPT).

[0056] In this case, the result-generating AI agent model (14) can use the called advertiser AI agent to summarize selected advertising documents and refine them into advertising information, but is not limited thereto.

[0057] For example, the result generating AI agent model (14) can use the called advertiser AI agent to summarize text about "brand name of whitening cosmetics," "promotional benefits when purchasing whitening cosmetics," and "specific efficacy of whitening cosmetics" from selected advertising documents and refine it into advertising information.

[0058] Accordingly, the advertising information may be text containing information regarding "brand name," "promotional benefits," and "product specifications," for example, "Company A," "25% autumn discount," and "industry's first natural substance extract contained" (result of search augmentation generation technology).

[0059] Next, the result-generating AI agent model (14) can generate a response including advertising information by using the called advertiser AI agent to input a query (e.g., "Tell me professional reviews of new whitening cosmetics or overseas price trends"), advertising information (e.g., Company A, 25% discount for autumn, industry's first natural substance extract), and an advertising placement guide (e.g., generate a response including brand, promotion benefits, and product specifications) as prompts into a natural language processing model (generating a response through the natural language processing model of the called advertiser AI agent, workflow including the natural language processing model, etc.). In this case, the advertising placement guide may be a command to generate a response by selecting advertising information (brand name, promotion benefits, product specifications) of a specific category among advertising information of various categories (e.g., brand name, promotion benefits, product specifications, product sales period, etc.), but is not limited thereto.

[0060] Accordingly, responses containing advertising information may include, but are not limited to, "Company A's whitening cosmetics are currently offering a 25% discount for the autumn season and are characterized by containing natural substance extracts for the first time in the industry."

[0061] The step (250) of generating a final search result by combining a response containing advertising information with an existing search result of an external data agent in a result generating artificial intelligence agent module (14) of a search engine platform (10) may be a step in which the result generating artificial intelligence agent module (14) receives an existing search result from an external data agent via an API method, etc., and combines it with a response containing advertising information.

[0062] In this case, the external data agent can generate (obtain) existing search results that are the results of searching for a query on an external web (for example, a search engine web such as Google), and can be generated by extracting professional reviews of whitening cosmetics or articles or blog responses from the web regarding overseas prices.

[0063] In other words, since the invoked advertiser AI agent has a significant bias toward responding with advertising information, existing search results are also received through an external data agent to compensate for user satisfaction with the response, and they are combined contextually to generate the final search result.

[0064] For example, the result generation artificial intelligence agent module (14) can combine a response containing advertising information and an existing search result according to contextual similarity. That is, in the method (200) of the present invention, the text of the response containing advertising information and the existing search result can be split, and then contextual similarity can be determined by determining the similarity of the tokens embedded with the tokens and combining them contextually to generate a final search result, but is not limited thereto.

[0065] For example, the final search result for the query "Tell me professional reviews of new whitening cosmetic products or overseas price trends" can be generated by combining an existing search result with a response containing advertising information such as "Company A's whitening cosmetic is currently offering a 25% autumn sale and is characterized by containing natural substance extracts for the first time in the industry": "Company A's product has high moisturizing and anti-pigmentation effects due to natural substances, Company B's product is low-cost and particularly effective for exfoliation, and Company C's product is preferred by young women in their 20s for its hydration, and globally, the price of whitening cosmetics was expected to continuously decline until 2024 but has recently been on an upward trend, with the increase being particularly significant for high-end brand lines." Additionally, tokens with high contextual similarity, such as "Company A's whitening cosmetic - Company A's product," "Industry's first to contain natural substance extracts - extracts," and "Autumn sale 25% - Global price of whitening cosmetics," can be contextually combined to [Company A's product contains natural substance extracts for the first time in the industry] for moisturization and anti-pigmentation It is possible to derive a final search result in which advertising information (results using the called advertiser AI agent) is contextually combined with general information (existing search results) such as, "Company B's product is low-cost and particularly effective for exfoliation care, and Company C's product is preferred by young women in their 20s for its moisturizing properties, and globally, the price of whitening cosmetics continued to decline until 2024 but has recently been on an upward trend, with the increase being particularly large for high-end brand lines, and currently, [Company A's product is offering a 25% discount for the autumn season.]"

[0066] Accordingly, the method (200) of the present invention has the advantage of naturally integrating advertisements and search results by incorporating advertisements into actual answers to user queries, unlike keywords or banner advertisements of general search engines, thereby reducing resistance.

[0067] The step (260) in which the artificial intelligence ad inspection module (15) of the search engine platform (10) determines falseness, illegality, and excessiveness in the final search result and transmits the filtered response to the user terminal (30) may be a step of providing the filtered response to the user by deleting or modifying false, illegal, or exaggerated content from the ad information of the final search result before providing the response to the user.

[0068] In this case, the artificial intelligence ad inspection module (15) may be a natural language processing model trained with the final search result as a feature value and the response filtered by deleting or modifying advertising expressions that are false, illegal, and excessive in the final search result as a label value, and may provide the output response to the user terminal (30) by inputting the final search result into the artificial intelligence ad inspection module (15), but is not limited thereto. Alternatively, the artificial intelligence ad inspection module (15) may be an artificial intelligence natural language processing model trained with text bundles of correlational advertising expressions that are false, illegal, and excessive and expressions modified therefrom. Additionally, the artificial intelligence ad inspection module (15) may detect expressions with high falsehood, illegality, and excessiveness in the final search result and ultimately determine whether it is an advertising expression with falsehood, illegality, and excessiveness by comprehensively considering the similarity with pre-set expressions with high falsehood, illegality, and excessiveness and whether the expression is listed in an official document that was published in the existing search result.

[0069] Accordingly, the artificial intelligence ad inspection module (15) can output a response that deletes or modifies expressions that are highly false, illegal, and excessive, such as the expression “first in the industry,” in the final search results, such as [Company A’s product contains natural substance extracts for the first time in the industry], which have high moisturizing and anti-pigmentation effects, Company B’s product is low-cost and particularly effective for exfoliation care, and Company C’s product is preferred by young women in their 20s for its moisturizing effect, and the price of whitening cosmetics worldwide has been continuously declining until 2024 and has recently been on an upward trend, with the increase in high-end brand lines being particularly large, and currently [Company A’s product is offering a 25% discount for autumn.]

[0070] The step (270) in which the ad revenue module (16) of the search engine platform (10) charges ad revenue to the advertiser terminal (20) based on the exposure of ad information in the response provided to the user terminal (30) may be a step of charging revenue to the advertiser terminal (20) based on the number of times ad information is exposed in the response provided to the user by the advertiser artificial intelligence agent. In addition, in the method (200) of the present invention, the search engine platform (10) may also charge licensing fees for the registration, update, etc. of the advertiser artificial intelligence agent.

[0071] As described above, the system (100) and method (200) of the present invention have the advantage of: i) that advertisers can immediately update information (latest news, new product launches, price changes, events, etc.) through an artificial intelligence agent, thereby enabling autonomous and real-time updates by advertisers; ii) that search engine operators can secure stable artificial intelligence search advertising revenue through billing, license fees, etc. based on the registration and exposure of the artificial intelligence agent, thereby having a revenue model suitable for the artificial intelligence era; and iii) that user experience (UX) is improved by increasing search satisfaction through the exposure of advertisements that match the context of the query, rather than indiscriminate advertisements or advertisements unrelated to the content of the question.

[0072]

[0073] The method (200) of the present invention described above may be implemented as a program (or application) to be executed in combination with a server, which is hardware, and stored on a medium.

[0074] The aforementioned program may include code encoded in a computer language such as C, C++, JAVA, or machine language, which can be read by the computer's processor (CPU) through the computer's device interface, in order for the computer to read the program and execute the methods implemented in the program. Such code may include functional code related to functions that define the necessary functions for executing the methods, and may include control code related to execution procedures necessary for the computer's processor to execute the functions according to a predetermined procedure. Additionally, such code may further include memory reference code regarding where (address) additional information or media necessary for the computer's processor to execute the functions should be referenced in the computer's internal or external memory. In addition, if the processor of the computer needs to communicate with any other computer or server located remotely in order to execute the above functions, the code may further include communication-related code regarding how to communicate with any other computer or server located remotely using the communication module of the computer, and what information or media to transmit or receive during communication.

[0075] The above-mentioned storage medium refers to a medium that stores data semi-permanently and is readable by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, examples of the above-mentioned storage medium include, but are not limited to, ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. That is, the above-mentioned program may be stored on various recording media on various servers that the computer can access, or on various recording media on the user's computer. Additionally, the above-mentioned medium may be distributed across networked computer systems, and computer-readable code may be stored in a distributed manner.

[0076] The steps of the method or algorithm described in connection with embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, a hard disk, a removable disk, a CD-ROM, or any form of computer-readable recording medium well known in the art to which the present invention belongs.

[0077] Although embodiments of the present invention have been described above with reference to the attached drawings, those skilled in the art will understand that the present invention may be implemented in other specific forms without altering its technical concept or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive.

Claims

1. A step in which the advertiser terminal registers the advertiser AI agent in the search engine platform's AI agent storage; A step in which a user terminal transmits a query to an interface module of the search engine platform; A step of calling an advertiser AI agent highly relevant to the query in the advertiser AI agent allocation module of the above-mentioned search engine platform; A step of generating a response including advertising information using an artificial intelligence agent called from the result generation artificial intelligence agent module of the search engine platform; and A method characterized by including the step of generating a final search result by combining a response containing advertising information from an artificial intelligence agent module of the search engine platform with an existing search result from an external data agent.

2. In Paragraph 1, A method characterized by an advertiser AI agent allocation module of the above-mentioned search engine platform selecting an advertiser AI agent that is highly relevant to the query using at least one of vector embedding or keyword analysis technology.

3. In Paragraph 1, The result-generating artificial intelligence agent model of the above-mentioned search engine platform is, A method characterized by using a called advertiser AI agent to generate a response containing advertising information in at least one of the following methods: real-time data collection through an external database, document exploration based on search augmentation generation technology, response generation through a natural language processing model, and a workflow including a natural language processing model.

4. In Paragraph 3, The result-generating artificial intelligence agent model of the above-mentioned search engine platform is, Using the called advertiser AI agent, determine the similarity between the query embedding and the ad document embedding registered by the advertiser AI agent, and select ad documents with high similarity to the query, and A method characterized by using a called advertiser AI agent to input a query and selected ad documents, refined ad information, and an ad placement guide as prompts into a natural language processing model to generate a response containing ad information.

5. In Paragraph 4, The result-generating artificial intelligence agent model of the above-mentioned search engine platform is, A method characterized by summarizing selected advertising documents and refining them into advertising information.

6. In Paragraph 1, The result generation AI agent module of the above-mentioned search engine platform generates a final search result by contextually combining existing search results from an external data agent with a response containing advertising information, and A method characterized by an external data agent generating existing search results for a query from an external web.

7. In Paragraph 1, A method characterized by further including the step of transmitting to the user terminal a response filtered by an artificial intelligence ad inspection module of the search engine platform, which determines falsehood, illegality, and excessiveness in the final search result.

8. In Paragraph 7, A method characterized in that the artificial intelligence ad inspection module of the above-mentioned search engine platform is a natural language processing model that learns the final search result as a feature value and the response filtered from the final search result to include false, illegal, and excessive advertising expressions as a label value.

9. In Paragraph 1, A method characterized by further including the step of an ad revenue module of the search engine platform claiming ad revenue from an advertiser terminal based on the exposure of ad information of a response provided to the user terminal.

10. Search engine platform; An advertiser terminal that registers an advertiser AI agent on the above-mentioned search engine platform; and It includes a user terminal that inputs a query into the above-mentioned search engine platform and receives results using an advertiser AI agent, and The above search engine platform is, AI agent storage where an advertiser AI agent is registered by the advertiser terminal; An interface module that receives a query from the above user terminal; An advertiser AI agent assignment module that calls an advertiser AI agent highly relevant to the query; and A system characterized by including a result generating AI agent module that generates a response containing advertising information for a query using a called AI agent, and generates a final search result by combining the response containing advertising information with existing search results from an external data agent.

Images