Advertising strategy generation method and system

Abstract

This invention provides an advertising strategy generation method, system, and program that generate the optimal advertising strategy for advertising request queries composed of multiple natural languages. [Solution] The advertising strategy generation method includes the steps of: receiving an advertising request query from an advertiser terminal S100; inputting the advertising request query into a generative AI model and generating a first detailed query and a second detailed query to achieve the advertising request query S200; determining a data source to store the data necessary to achieve each detailed query S300; obtaining search results corresponding to each detailed query from each data source S400; integrating the search results to generate a final search result for the advertising request query S500; and inputting the final search result into a generative AI model to generate an optimal advertising strategy for the advertising request query S600.

Description

Technical Field

[0001] The present disclosure relates to a method for generating an advertising strategy and a system thereof. More specifically, it relates to a method for generating an advertising strategy that recommends an optimal advertising strategy for an advertising request query composed of natural language and a system thereof.

Background Art

[0002] In conventional advertising systems, when selecting various online or offline advertising media, an advertising strategy is formulated based on the results of direct human data analysis. High-cost unit DOOH (Digital Out-of-Home) products have a mechanism that is difficult for small and medium-sized advertisers to use. Also, there are limitations in the linkage and effect measurement between digital out-of-home advertising and online advertising, and there is a problem that it is difficult to simultaneously satisfy the efficiency and cost-effectiveness of advertising.

[0003] Therefore, there is a need for a technology that recommends an optimal advertising strategy for an advertising request query of an advertiser composed of natural language.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Technical problems to be solved by some embodiments of the present disclosure are to provide an advertising strategy generation method and a system thereof that generate an optimal advertising strategy for an advertising request query composed of natural language.

[0006] Another technical problem that some embodiments of this disclosure aim to solve is to provide an advertising strategy generation method and system that delivers advertisements based on the optimally generated advertising strategy.

[0007] The technical issues of this disclosure are not limited to those described above, and other technical issues not mentioned will be clearly understood by a person of the ordinary skill in the art of this disclosure from the following description. [Means for solving the problem]

[0008] An advertising strategy generation method according to some embodiments of the present disclosure for solving the technical problems described above may include, in a method performed by a computing system, the steps of: receiving an advertising request query from an advertiser terminal; inputting the advertising request query into a generative AI model to generate a first detail query and a second detail query to achieve the advertising request query; determining a first data source for storing data necessary to achieve the first detail query and a second data source for storing data necessary to achieve the second detail query; obtaining a first search result corresponding to the first detail query from the first data source and a second search result corresponding to the second detail query from the second data source; integrating the first and second search results to generate a final search result for the advertising request query; and inputting the final search result into the generative AI model to generate an optimal advertising strategy for the advertising request query.

[0009] In some embodiments, the step of generating the first and second detailed queries may include the step of extracting keywords from the ad request query and the step of generating the first and second detailed queries using the keywords.

[0010] In some embodiments, the step of determining the first and second data sources may include comparing the index of each of the first and second detail queries with the index of each data source, and determining the first and second data sources using the results of the comparison.

[0011] In some embodiments, the step of obtaining the first and second search results may include performing a search for the first detail query against the first data source and performing a search for the second detail query against the second data source in parallel.

[0012] In some embodiments, the step of generating the final search result may include a step of filtering out data that overlaps between the first search result and the second search result.

[0013] In some embodiments, the filtering step may include a step of pre-processing the first search result and the second search result by attribute, a step of mapping the pre-processed first search result and the pre-processed second search result based on common attributes, and a step of merging the first search result and the second search result if the attribute values ​​corresponding to the common attributes are the same.

[0014] In some embodiments, the step of generating the final search result may include: calculating a first importance score, which is the degree of importance of the first search result, based on a first weight predefined for the first data source; calculating a second importance score, which is the degree of importance of the second search result, based on a second weight predefined for the second data source; and integrating the first and second search results based on the first and second importance scores.

[0015] An advertising strategy generation system according to some embodiments of the present disclosure for solving the technical problems described above includes a communication interface, a memory on which a computer program is loaded, and one or more processors on which the computer program is executed, wherein the computer program may include instructions to perform the following operations: receiving an advertising request query from an advertiser terminal; inputting the advertising request query into a generative AI model and generating a first detail query and a second detail query to achieve the advertising request query; determining a first data source for storing data necessary to achieve the first detail query and a second data source for storing data necessary to achieve the second detail query; obtaining a first search result corresponding to the first detail query from the first data source and a second search result corresponding to the second detail query from the second data source; integrating the first and second search results to generate a final search result for the advertising request query; and inputting the final search result into the generative AI model to generate an optimal advertising strategy for the advertising request query.

[0016] A computer program stored on a computer-readable recording medium according to some embodiments of the present disclosure for solving the technical problems described above can be coupled with a computing device and perform the following steps: receiving an advertising request query from an advertiser terminal; inputting the advertising request query into a generative AI model to generate a first detail query and a second detail query to achieve the advertising request query; determining a first data source to store the data necessary to achieve the first detail query and a second data source to store the data necessary to achieve the second detail query; obtaining a first search result corresponding to the first detail query from the first data source and a second search result corresponding to the second detail query from the second data source; integrating the first and second search results to generate a final search result for the advertising request query; and inputting the final search result into the generative AI model to generate an optimal advertising strategy for the advertising request query. [Brief explanation of the drawing]

[0017] [Figure 1] This is a system configuration diagram illustrating the configuration and operation of an advertising system according to some embodiments of the present disclosure. [Figure 2] This is a flowchart illustrating the operation of an advertising strategy generation method according to some embodiments of the present disclosure. [Figure 3] This is a detailed flowchart illustrating the detailed operation of the advertising strategy generation method according to some embodiments of the present disclosure, as described with reference to Figure 2. [Figure 4] This is a detailed flowchart illustrating the detailed operation of the advertising strategy generation method according to some embodiments of the present disclosure, as described with reference to Figure 2. [Figure 5] This figure illustrates the operation of an advertising strategy generation method according to some embodiments of the present disclosure. [Figure 6]A detailed flowchart for explaining the detailed operations of an advertising strategy generation method according to some embodiments of the present disclosure, described with reference to FIG. 2. [Figure 7] A diagram for explaining a method of generating a final search result for an advertising request query according to some embodiments of the present disclosure. [Figure 8] A detailed flowchart for explaining the detailed operations of an advertising strategy generation method according to some embodiments of the present disclosure, described with reference to FIG. 2. [Figure 9] A configuration diagram of the hardware of a computing device according to some embodiments of the present disclosure. **Modes for Carrying Out the Invention**

[0018] Hereinafter, various embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. The advantages and features of the present disclosure, and the methods for achieving them, will become clear by referring to the embodiments described in detail below together with the accompanying drawings. However, the technical idea of the present disclosure is not limited to the following embodiments, and can be realized in various different forms. The following embodiments are merely to complete the technical idea of the present disclosure and are provided to fully inform those with ordinary knowledge in the technical field to which the present disclosure belongs of the scope of the present disclosure. The technical idea of the present disclosure is defined only by the scope of the claims.

[0019] When explaining various embodiments of the present disclosure, if it is determined that a specific explanation related to a related known configuration or function obscures the gist of the present disclosure, the detailed explanation thereof will be omitted.

[0020] Unless otherwise defined, the terms (including technical and scientific terms) used in the following embodiments are used in a meaning commonly understood by those with ordinary knowledge in the technical field to which the present disclosure belongs, but this may change depending on the intention or precedent of those skilled in the relevant field, the emergence of new technologies, etc. The terms used in the present disclosure are for explaining the embodiments and do not limit the scope of the present disclosure.

[0021] In the following embodiments, singular expressions include plural concepts unless explicitly identified as singular in context. Similarly, plural expressions include singular concepts unless explicitly identified as plural in context.

[0022] Furthermore, terms such as First, Second, A, B, (a), (b), etc., used in the following embodiments are used to distinguish one component from another, and the terms do not limit the nature, order, or sequence of the component in question.

[0023] Various embodiments of this disclosure will be described in detail below with reference to the attached drawings.

[0024] The configuration and operation of advertising systems according to some embodiments of this disclosure will be described below with reference to Figure 1. Figure 1 is a system configuration diagram illustrating the configuration and operation of advertising systems according to some embodiments of this disclosure.

[0025] Referring to Figure 1, the advertising system may consist of an advertising strategy generation system 1, an advertiser terminal 20, and a data source 30. The advertising strategy generation system 1 may consist of a service server 10, a generative AI model 11, and a database 12. However, the scope of this disclosure is not limited thereto. In some cases, the advertising system may also consist of modules / devices / systems not shown in Figure 1. Alternatively, the event search system may be configured in a way that excludes at least some of the components (1 to 32) shown in Figure 1.

[0026] The advertiser terminal 20 sends an advertising request query written in natural language to the advertising strategy generation system 1. The advertiser terminal 20 is a terminal used by the advertiser and receives the advertising request query input from the advertiser.

[0027] When the advertising strategy generation system 1 receives an advertising request query, the service server 10 inputs the advertising request query into the generative AI model 11 and can generate a detailed query to achieve the advertising request query. The generative AI model can generate the detailed query to achieve the advertising request query using CoT (Chain of Thought) technology. CoT technology is designed to explicitly show the step-by-step logical thought process when the generative AI model 11 solves a complex problem, and can show the intermediate reasoning process rather than simply deriving the result. A detailed explanation is omitted as this is common knowledge to those of the ordinary art in the field to which the inventions disclosed herein belong. This will be explained in detail with reference to Figure 3.

[0028] The generative AI model 11 is an artificial intelligence model that generates advertising strategies in response to the advertising request queries based on those queries. It can understand advertiser requests and generate new results through large-scale data learning and deep learning techniques. Typical examples of generative AI models 11 include GPT and BERT in text generation. Since these are well known to those ordinarily skilled in the art to which the inventions disclosed herein belong, a detailed explanation is omitted.

[0029] The service server 10 can determine the data source 30 that stores the necessary data to accomplish the multiple detailed queries generated. Each data source 31, 32 may be a database that stores different types of data. For example, the first data source 31 may be a database that stores demographic data, and the second data source 32 may be a database that stores sales data for each store. Data source 30 may include the first data source 31 and the second data source 32, as well as a database that stores a variety of data. This will be explained in more detail with reference to Figure 4.

[0030] The service server 10 can perform a search for detailed queries corresponding to each data source and obtain search results for each detailed query. In other words, the service server 10 can obtain appropriate search results for each detailed query from each data source 30. This will be explained in detail with reference to Figure 5.

[0031] Subsequently, the service server 10 can integrate each search result and generate the final search result for the ad request query. This will be explained in detail with reference to Figures 6 and 7.

[0032] The service server 10 can input the final search results into the generative AI model 11 and generate the optimal advertising strategy for the advertising request query. The generative AI model 11 can summarize the generated final search results in a user-friendly manner, create an advertising delivery strategy and related reports, and send them to the advertiser terminal 20 and the advertising execution system (not shown).

[0033] Database 12 may be an internal device of the advertising strategy generation system 1 that stores various data generated during the advertising strategy generation process, such as advertising request queries received from advertiser terminals 20 and optimal advertising strategy data for those advertising request queries.

[0034] Each of the components (10, 11, and 12) of the advertising strategy generation system 1 described above can be implemented on at least one computing device. For example, all functions of the advertising strategy generation system 1 may be implemented on a single computing device, or the first function of the advertising strategy generation system 1 may be implemented on the first computing device and the second function on the second computing device. Alternatively, specific functions of the advertising strategy generation system 1 may be implemented on multiple computing devices.

[0035] Computing devices can include any device that possesses computing capabilities; see Figure 9 for an example of such a device. Because computing devices are a collection of interacting components (e.g., memory, processors, etc.), they can sometimes be called "computing systems." Of course, the term "computing system" also encompasses the concept of a collection of interacting computing devices.

[0036] In some embodiments, the components (1 to 30) of the event search system can communicate over a network. Here, the network can be any type of wired / wireless network, such as a Local Area Network (LAN), Wide Area Network (WAN), mobile radio communication network, or Wibro (Wireless Broadband Internet).

[0037] In the following explanation, for the sake of clarity, we will assume that all stages / operations of the method described below are performed on the service server 10 mentioned above. Therefore, if the subject of a particular stage / operation is omitted, it can be understood that it is performed on the service server 10. However, in a real environment, some stages / operations of the method described below may also be performed on other computing devices.

[0038] The operation of the advertising strategy generation method according to some embodiments of the present disclosure will be described below with reference to Figure 2. Figure 2 is a flowchart illustrating the operation of the advertising strategy generation method according to some embodiments of the present disclosure.

[0039] Referring to Figure 2, the service server 10 receives an advertising request query from the advertiser terminal 20 (S100). The advertising request query may be an advertising request item composed in natural language. For example, the advertising request query may be "I want to efficiently deliver cosmetic advertisements to women in their 20s and 30s in Gangnam-gu, Seoul." However, the scope of this disclosure is not limited to this.

[0040] Subsequently, the service server 10 inputs the advertising request query into the generative AI model 11 and uses CoT technology to generate a first detailed query and a second detailed query to achieve the advertising request query (S200).

[0041] The following describes how to generate detailed queries according to some embodiments of the present disclosure, with reference to Figure 3. Figure 3 is a detailed flowchart illustrating the detailed operation of the advertising strategy generation method according to some embodiments of the present disclosure, as described with reference to Figure 2.

[0042] Referring to Figure 3, the service server 10 extracts keywords from the advertising request query (S210). For example, if the advertising request query is "We want to efficiently deliver cosmetics advertisements to women in their 20s and 30s in Gangnam-gu, Seoul," the extracted keywords may be "Gangnam-gu, Seoul," "women in their 20s and 30s," "cosmetics advertisements," and "advertising efficiency."

[0043] Subsequently, the service server 10 inputs the keywords into the generative AI model 11 to generate the first and second detailed queries (S220). For example, the generative AI model 11 can generate a region-based query, "Percentage of women in their 20s and 30s in Seoul Gangnam District," from the first keywords, "Seoul Gangnam District" and "women in their 20s and 30s." The generative AI model 11 can generate a consumer-based query, "Consumption patterns in the Gangnam District commercial area," from the second keyword, "Cosmetics advertising." The generative AI model 11 can generate a trend-based query from the third keyword, "Advertising efficiency." The trend-based query could be related to commercial area and cosmetics purchasing trends, or to efficient advertising media.

[0044] Let's refer to Figure 2 again for further explanation.

[0045] When the first and second detailed queries are generated (S200), the service server 10 determines a first data source 31 to store the data necessary to accomplish the first detailed query and a second data source 32 to store the data necessary to accomplish the second detailed query (S300). Each data source 31, 32 may be a database that stores different types of data.

[0046] The following describes how to determine a data source according to some embodiments of the present disclosure, with reference to Figure 4. Figure 4 is a detailed flowchart illustrating the detailed operation of the advertising strategy generation method according to some embodiments of the present disclosure, as described with reference to Figure 2.

[0047] Referring to Figure 4, the service server 10 compares the index of each of the first detailed query and the detailed query with the index of each data source 31,32 (S310). The index of each data source 31,32 is data relating to the characteristics of each data source 31,32, and for example, in the case of a region-based query, it may be decided to perform a search on a data source that stores region-related data.

[0048] The service server 10 uses the results of the comparison to determine the first data source 31 and the second data source 32 (S320). For example, in the case of a region-based query containing the keyword "Seoul Gangnam District," it may be determined to perform a search on region-related data sources. In the case of a consumer-based query containing the keyword "women in their 20s and 30s," it may be determined to perform a search on demographic-related data sources. In the case of a trend-related query containing the keyword "cosmetics," it may be determined to perform a search on data sources that store consumer behavior and sales data.

[0049] Let's refer to Figure 2 again for further explanation.

[0050] The service server 10 obtains a first search result corresponding to the first detailed query from the first data source 31 and a second search result corresponding to the second detailed query from the second data source 32 (S400). The service server 10 can perform a search on the first data source 31 for the first detailed query and obtain the first search result as a result of the search. The service server 10 can perform a search on the first data source 31 for the second detailed query and obtain the second search result as a result of the search.

[0051] The operation of the advertising strategy generation method according to some embodiments of the present disclosure will be described below with reference to Figure 5. Figure 5 is a diagram illustrating the operation of the advertising strategy generation method according to some embodiments of the present disclosure.

[0052] Referring to Figure 5, we assume that the ad request query 500 received from advertiser terminal 20 is "We want to efficiently deliver cosmetics advertisements to women in their 20s and 30s in Gangnam-gu, Seoul."

[0053] The service server 10 inputs the ad request query 500 into the generative AI model 11 and can generate a region-based first detail query 510, a consumer-based second detail query 520, and a trend-based third detail query 530 to fulfill the ad request query 500.

[0054] The first detailed query 510 could be a query related to region-related data, such as "population distribution within Gangnam-gu, Seoul." The second detailed query 520 could be a query related to consumer-related data, such as "purchasing patterns of cosmetics consumers within Gangnam-gu, Seoul, and trends in cosmetics sales." The third detailed query 530 could be a query related to trend-related data, such as "latest search keywords related to cosmetics and cosmetics purchasing channels."

[0055] The service server 10 may determine a first data source 511 that stores the data necessary to achieve the first detailed query 510. The first data source 511 is the data necessary to achieve the region-based first detailed query 510 and may be a data source that stores demographic data.

[0056] The service server 10 may determine a second data source 521 that stores the data necessary to achieve the second detailed query 520. The second data source 521 is the data necessary to achieve the consumer-based second detailed query 520 and may be a data source that stores consumer behavior data and sales data, etc. Consumer behavior data may include information related to consumers' purchasing patterns, preferred media for each consumer, preferred brands for each consumer, etc.

[0057] The service server 10 may determine a third data source 531 that stores the data necessary to achieve the third detailed query 530. The third data source 531 is the data necessary to achieve the trend-based third detailed query 530 and may be a data source that stores trade area data and trend data. Trade area data may include information on the location and population flow of major trade areas. Trend data may include information on the latest search keywords, seasonally popular products, etc.

[0058] The service server 10 performs a search on the first data source 511 for the first detailed query 510 and obtains the first search result as the search result. For example, the service server 10 can perform a search on the first data source 511 for the population ratio within Gangnam-gu, Seoul, and obtain the age-specific population ratio within Gangnam-gu as the search result.

[0059] The service server 10 performs a search on the second data source 521 for the second detailed query 520 and obtains the second search result as the search result. For example, the service server 10 can perform a search on the second data source 521 for cosmetics purchasing patterns in Gangnam-gu and obtain age-specific cosmetics purchasing patterns in Gangnam-gu as the search result.

[0060] The service server 10 performs a search on the third data source 531 for the third detailed query 530 and obtains the third search result as the search result. For example, the service server 10 can search the third data source 531 for the latest cosmetics-related keywords searched within Gangnam-gu and obtain the latest cosmetics-related keywords as the search result.

[0061] In one embodiment, the service server 10 can perform a search for the first detailed query 510 on the first data source 511 and a search for the second detailed query 520 on the second data source 521. That is, the service server 10 can perform searches for the first detailed query 510 and the second detailed query 520 on the first data source 31 and the second data source 32 in parallel.

[0062] According to this embodiment, by processing queries simultaneously against multiple data sources, search speed can be increased and the overall process efficiency can be maximized. Therefore, this embodiment has the effect of saving resources such as time and cost.

[0063] Let's refer to Figure 2 again for further explanation.

[0064] Once the first and second search results are obtained (S400), the service server 10 integrates the first and second search results to generate the final search result for the ad request query (S500). The service server 10 may filter out any overlapping data between the first and second search results. The service server 10 can derive the final search result by merging the first and second search results based on the overlapping data between them.

[0065] The following describes how to generate final search results according to some embodiments of the present disclosure, with reference to Figure 6. Figure 6 is a detailed flowchart illustrating the detailed operation of the advertising strategy generation method according to some embodiments of the present disclosure, as described with reference to Figure 2.

[0066] Referring to Figure 6, the service server 10 preprocesses the first and second search results by attribute (S510). For example, the service server 10 can preprocess the first and second search results into standardized data by attribute according to a certain table format.

[0067] Subsequently, the service server 10 matches the pre-processed first search result and the pre-processed second search result based on common attributes, which are attributes common to both search results (S511).

[0068] If the attribute values ​​corresponding to the common attributes of the pre-processed first search result and the pre-processed second search result are the same, the service server 10 merges the first search result and the second search result (S512).

[0069] The embodiments shown in Figure 6 will be further described below with reference to Figure 7. Figure 7 is a diagram illustrating how to generate final search results for an ad request query according to some embodiments of the present disclosure.

[0070] Referring to Figure 7, the pre-processed first search result 60, pre-processed second search result 61, pre-processed third search result 62, and pre-processed fourth search result 63 are shown. The demographic data, which is the first search result 60, may include attributes such as region, age group, gender, and population ratio. The consumer behavior data, which is the second search result 61, may include attributes such as age group, purchase keywords, preferred channel, and purchase frequency. The trend data, which is the third search result 62, may include attributes such as keywords, search volume growth rate, and relevant season. The sales data, which is the fourth search result 63, may include attributes such as region, keywords, sales, and growth rate.

[0071] The service server 10 can map the first search result 60, the second search result 61, the third search result 62, and the fourth search result 63 based on common attributes. The service server 10 can map the first search result 60 and the second search result 61 based on age range, which is a common attribute of the two search results. The service server 10 can map the second search result 61 and the third search result 62 based on (purchase) keyword, which is a common attribute of the two search results. The service server 10 can map the second search result 61 and the fourth search result 63 based on keyword (=category), which is a common attribute of the two search results. The service server 10 can map the third search result 62 and the fourth search result 63 based on keyword (=category), which is a common attribute of the two search results. The service server 10 can map the first search result 60 and the fourth search result 63 based on region, which is a common attribute of the two search results.

[0072] The service server 10 can merge the first search result 60, second search result 61, third search result 62, and fourth search result 63 to generate the final search result 64 if the attribute values ​​corresponding to the common attributes of the first search result 60, second search result 61, third search result 62, and fourth search result 63 are the same. The final search result 64 may include information such as the region is Gangnam-gu, Seoul, the age range is 20s to 30s, the percentage of women is 35%, the keyword is vegan cosmetics, the preferred channel is Instagram, the purchase frequency is once a week, the search volume growth rate is 30%, the relevant season is spring, the sales are 50 million won, and the growth rate is 30%. In other words, the specific advertising strategy details for the ad request query received from the advertiser terminal 20 ("I want to efficiently deliver cosmetics advertisements in Seoul.") may be to deliver vegan cosmetics to women in their 20s to 30s in Gangnam-gu, Seoul during the spring, via Instagram.

[0073] According to this embodiment, it is possible to generate specific advertising strategy details in response to advertising request queries received from the advertiser terminal 20. Therefore, it has the advantage of maximizing advertising effectiveness by generating an optimal advertising strategy according to specific advertising strategy details.

[0074] The following describes how to generate final search results according to some embodiments of the present disclosure, with reference to Figure 8. Figure 8 is a detailed flowchart illustrating the detailed operation of the advertising strategy generation method according to some embodiments of the present disclosure, as described with reference to Figure 2.

[0075] Referring to Figure 8, the service server 10 calculates a first importance score, which is the degree of importance of the first search result, based on a first weight predefined for the first data source 31 (S520). The service server 10 calculates a second importance score, which is the degree of importance of the second search result, based on a second weight predefined for the second data source 32 (S521).

[0076] The first and second weights may be values ​​set by the user for each data source. The first and second weights can be set to differ depending on the recency of the data (how much of the latest information is reflected from the time the search results were obtained), the degree of relevance to the region identified from the ad request query received from the advertiser terminal 20, and the confidence level for each data source.

[0077] Subsequently, the service server 10 integrates the first and second search results based on the first and second importance scores (S522). For example, if the first importance score is above a certain threshold and the second importance score is below that threshold, the service server 10 may include the first search result in the final search results while excluding the second search result. Alternatively, the service server 10 may perform percentile calculations on the first and second importance scores, determine a ratio for integrating the first and second search results according to each importance score calculated as a percentile, and then integrate the first and second search results according to the determined ratio.

[0078] However, the scope of this disclosure is not limited thereto, and the method of integrating search results according to each importance score may be set to vary depending on the circumstances.

[0079] According to this embodiment, there is an advantage in that the effectiveness of the advertising strategy generated based on the final search results can be maximized by adjusting the degree of importance of the search results obtained from each data source and generating the final search results based on that.

[0080] Let's refer to Figure 2 again for further explanation.

[0081] When the final search results for the ad request query received from the advertiser terminal 20 are generated (S500), the service server 10 inputs the final search results into the generative AI model 11 and generates the optimal ad strategy for the ad request query (S600). The generative AI model 11 can summarize the generated final search results in a user-friendly manner, create an ad delivery strategy and related reports, and send them to the advertiser terminal 20 and the ad execution system (not shown).

[0082] The service server 10 can transmit the generated advertising strategy to an advertising implementation system (not shown). Based on the generated advertising strategy, the advertising implementation system (not shown) can automatically generate an advertising schedule that allows for the delivery of advertising content provided by the advertiser terminal 20. The advertising implementation system can automatically deliver the advertising content according to the generated advertising schedule. That is, when the advertising implementation system transmits the advertising content to an advertising media device (e.g., an outdoor billboard), the advertising media device can display the advertising content according to the advertising schedule.

[0083] For example, the service server 10 can input the final search results into the generative AI model 11 to generate a first advertising strategy as shown below. However, the first advertising strategy shown below is just one example, and the scope of this disclosure is not limited thereto.

[0084] "Advertising area: Gangnam Station, located in Gangnam-gu, Seoul" Advertising media: Outdoor billboards (e.g., digital signage in subway stations), online advertising (e.g., advertising through social media platforms), Advertising time slots: (Weekdays) Commuting hours, (Weekends) Peak shopping hours. Advertising budget: 60% for outdoor advertising, 40% for online advertising. The service server 10 can transmit the first advertising strategy to an advertising implementation system (not shown). Based on the first advertising strategy, the advertising implementation system (not shown) can automatically generate a first advertising schedule for delivering advertising content provided by the advertiser terminal 20.

[0085] Subsequently, the advertising implementation system (not shown) can automatically distribute the advertising content according to the first advertising schedule. That is, when the advertising implementation system transmits the advertising content provided by the advertiser terminal 20 to an advertising media device such as an outdoor billboard or an online advertising platform, the advertising media device can display the advertising content according to the first advertising schedule.

[0086] According to this embodiment, even if the advertiser does not specify concrete advertising requests, an advertising strategy can be generated according to the advertiser's needs using only the minimum information provided by the advertiser. In other words, an optimal advertising strategy can be generated for advertising request queries composed in natural language. Therefore, this has the effect of improving the user experience for advertisers using the advertising strategy generation system 1.

[0087] Furthermore, according to this embodiment, advertisements can be automatically delivered based on the generated optimal advertising strategy. Therefore, by automatically generating advertising strategies and automatically delivering advertisements based on the generated strategies, convenience for advertisers can be enhanced and advertising effectiveness can be maximized.

[0088] Figure 9 is a hardware configuration diagram of a computing device according to some embodiments of the present disclosure. The computing device 1000 in Figure 9 may include one or more processors 1100, a system bus 1600, a communication interface 1200, memory 1400 for loading computer programs 1500 executed by the processors 1100, and storage 1300 for storing the computer programs 1500.

[0089] The computing device 1000 in Figure 9 shows, for example, the hardware structure of one or more computing devices that constitute the service server 10 described with reference to Figure 1.

[0090] The processor 1100 controls the overall operation of each configuration of the computing device 1000. The processor 1100 can perform calculations on at least one application or program to perform methods / operations according to various embodiments of the disclosure. The memory 1400 stores various data, instructions and / or information. The memory 1400 can load one or more computer programs 1500 from the storage 1300 to perform methods / operations according to various embodiments of the disclosure. The storage 1300 can non-temporarily store one or more computer programs 1500.

[0091] The computer program 1500 may include one or more instructions that implement the methods / operations according to various embodiments of the disclosure. When the computer program 1500 is loaded into memory 1400, the processor 1100 can execute the methods / operations according to various embodiments of the disclosure by having the one or more instructions execute.

[0092] In one embodiment, the computer program 1500 may include instructions to perform the following actions: receiving an advertising request query from an advertiser terminal; inputting the advertising request query into a generative AI model and generating a first detailed query and a second detailed query to achieve the advertising request query; determining a first data source for storing data necessary to achieve the first detailed query and a second data source for storing data necessary to achieve the second detailed query; obtaining a first search result corresponding to the first detailed query from the first data source and a second search result corresponding to the second detailed query from the second data source; integrating the first and second search results to generate a final search result for the advertising request query; and inputting the final search result into the generative AI model to generate an optimal advertising strategy for the advertising request query.

[0093] The various embodiments of this disclosure and the effects thereof have been described above with reference to Figures 1 to 9. The effects of the technical concept of this disclosure are not limited to those described above, and other effects not mentioned will be clearly understood by a person of the ordinary skill from the following description.

[0094] Furthermore, while the embodiments described above describe multiple components being combined into one or operating in a combined state, the technical concept of this disclosure is not necessarily limited to such embodiments. That is, within the scope of the technical concept of this disclosure, all components can also be selectively combined into one or more units and operate.

[0095] The technical concepts of this disclosure described above can be implemented in computer-readable code on a computer-readable medium. A computer program recorded on a computer-readable recording medium can be transferred to another computing device via a network such as the Internet, installed on the other computing device, and thus used on the other computing device.

Claims

1. In the manner in which it is performed by a computing system, The stage where an ad request query is received from the advertiser's device, The steps include inputting the aforementioned ad request query into a generative AI model and generating a first detailed query and a second detailed query to achieve the aforementioned ad request query, The steps include determining a first data source for storing data necessary to achieve the first detailed query and a second data source for storing data necessary to achieve the second detailed query, The steps include obtaining a first search result corresponding to the first detailed query from the first data source and a second search result corresponding to the second detailed query from the second data source, The steps include: integrating the first search result and the second search result to generate the final search result for the ad request query; An advertising strategy generation method, comprising the step of inputting the final search results into the generative AI model to generate the optimal advertising strategy for the advertising request query.

2. The step of generating the first detailed query and the second detailed query is: The step of extracting keywords from the aforementioned ad request query, The advertising strategy generation method according to claim 1, further comprising the step of generating the first detailed query and the second detailed query using the aforementioned keywords.

3. The step of determining the first data source and the second data source is: The steps include comparing the first detailed query and the second detailed query with the index of each data source, The advertising strategy generation method according to claim 1, further comprising the step of determining the first data source and the second data source using the results of the comparison.

4. The step of obtaining the first search result and the second search result is: The advertising strategy generation method according to claim 1, further comprising the steps of performing a search for the first detailed query against the first data source and performing a search for the second detailed query against the second data source in parallel.

5. The step of generating the final search results is as follows: The advertising strategy generation method according to claim 1, further comprising the step of filtering out duplicate data in the first search result and the second search result.

6. The filtering step described above is The steps include preprocessing the first search result and the second search result by attribute, The steps include mapping the pre-processed first search result and the pre-processed second search result based on common attributes, The advertising strategy generation method according to claim 5, further comprising the step of merging the first search result and the second search result when the attribute values ​​corresponding to the common attribute are the same.

7. The step of generating the final search results is as follows: A step of calculating a first importance score, which is the degree of importance of the first search result, based on a first weight predefined for the first data source, A step of calculating a second importance score, which is the degree of importance of the second search result, based on a second weight that is predefined for the second data source, The advertising strategy generation method according to claim 1, further comprising the step of integrating the first search result and the second search result based on the first importance score and the second importance score.

8. Communication interface, The memory into which a computer program is loaded, The computer program includes one or more processors on which the computer program is executed. The aforementioned computer program, The system receives ad request queries from advertiser terminals and performs the necessary actions. The operation involves inputting the aforementioned ad request query into a generative AI model and generating a first detailed query and a second detailed query to achieve the aforementioned ad request query, An operation to determine a first data source for storing data necessary to achieve the first detailed query and a second data source for storing data necessary to achieve the second detailed query, The operation of obtaining a first search result corresponding to the first detailed query from the first data source and a second search result corresponding to the second detailed query from the second data source, The operation of integrating the first search result and the second search result to generate the final search result for the ad request query, An advertising strategy generation system, including instructions for inputting the final search results into the generative AI model and causing it to perform the operation of generating the optimal advertising strategy for the advertising request query.

9. Combined with computing devices, The stage where an ad request query is received from the advertiser's device, The steps include inputting the aforementioned ad request query into a generative AI model and generating a first detailed query and a second detailed query to achieve the aforementioned ad request query, The steps include determining a first data source for storing data necessary to achieve the first detailed query and a second data source for storing data necessary to achieve the second detailed query, The steps include obtaining a first search result corresponding to the first detailed query from the first data source and a second search result corresponding to the second detailed query from the second data source, The steps include: integrating the first search result and the second search result to generate the final search result for the ad request query; The final search results are input into the generative AI model, and the model is made to perform the step of generating the optimal advertising strategy for the advertising request query. A computer program stored on a computer-readable storage medium.

Images