Advertisement material generation method and device

Abstract

The embodiment of the invention provides an advertisement material generation method and device, and relates to the technical field of advertisement generation, and the method comprises the steps: obtaining different travel scenes; performing material demand analysis based on the travel scene; according to a set label, labeling and preprocessing the collected data to obtain a data set; according to the data set, performing fine adjustment on a set model; constructing a cue word template through a material demand, and inputting the cue word template into the fine-tuned model to generate an advertisement material; advertisement materials which are highly related to travel scenes and meet personalized requirements of users can be generated, the attraction and effect of advertisements are improved, and meanwhile the labor cost and the manufacturing time are reduced.

Description

Technical Field

[0001] The present application relates to the technical field of advertisement generation, and in particular to a method and device for generating advertisement material. Background Art

[0002] With the development of artificial intelligence technology, the application of AIGC (Artificial Intelligence Generated Content) in the advertising field has gradually increased. At present, in terms of advertising material generation, the common methods are mainly the following: using traditional image editing software, designers manually create advertising pictures, which requires a lot of manpower and time costs, and it is difficult to meet large-scale, personalized advertising needs; or using existing general AIGC models for picture generation. However, these models are not specially optimized and pre-trained for the travel field, and the generated pictures cannot accurately capture the specific needs of the travel field. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a method and device for generating advertising materials, so as to solve the problem that the existing advertising generation methods cannot accurately capture the specific needs of the travel field.

[0004] In a first aspect, an embodiment of the present application provides a method for generating advertising materials, including:

[0005] Get different travel scenarios;

[0006] Conduct material demand analysis based on travel scenarios;

[0007] According to the set labels, the collected data is labeled and preprocessed to obtain a data set;

[0008] Fine-tune the set model according to the data set;

[0009] A prompt word template is constructed according to the material requirements, and the prompt word template is input into the fine-tuned model to generate advertising materials.

[0010] In the above implementation process, different travel scenarios are obtained; based on the travel scenarios, material demand analysis is performed; according to the set labels, the collected data is annotated and preprocessed to obtain a data set; according to the data set, the set model is fine-tuned; a prompt word template is constructed according to the material demand, and the prompt word template is input into the fine-tuned model to generate advertising materials; it is possible to generate advertising materials that are highly relevant to the travel scenarios and meet the personalized needs of users, thereby improving the attractiveness and effectiveness of advertisements and reducing labor costs and production time.

[0011] Furthermore, the obtaining of different travel scenarios includes:

[0012] Define the scope of travel scenarios;

[0013] For each scene, different scene factors are set, wherein the scene factors include but are not limited to time, weather and mood.

[0014] In the above implementation process, different travel scenarios are obtained to ensure that the specific needs of users in the travel field are accurately captured.

[0015] Furthermore, the material demand analysis based on the travel scenario includes:

[0016] Based on travel scenarios, determine the types of materials required for different types of advertisements, and filter materials that meet specific resolution standards; among them, material types include but are not limited to high-definition pictures, video clips and animated images.

[0017] In the above implementation process, it is possible to accurately capture the specific needs and demand types of users in the travel field.

[0018] Furthermore, the collected data is labeled and preprocessed according to the set labels to obtain a data set, including:

[0019] Label the collected pictures according to the set tags; wherein the set tags include but are not limited to scene, emotion, time and weather;

[0020] Screening the collected images to remove blurry, distorted, or irrelevant images;

[0021] Unify the size of the pictures;

[0022] Enhance the image;

[0023] Unify the format of setting labels;

[0024] Combine the processed images and labels to get the dataset.

[0025] In the above implementation process, the input images are processed to ensure that they meet the input requirements of the model. At the same time, the sample diversity is increased and the labels are processed to ensure the consistency and availability of the labels.

[0026] Furthermore, before fine-tuning the set model according to the data set, the method further includes:

[0027] Fine-tune business marketing knowledge, including:

[0028] Injecting commercial marketing knowledge into the set model; where the commercial marketing knowledge includes but is not limited to advertising strategies, user psychology and market analysis;

[0029] Collect relevant data sources, including but not limited to commercial marketing literature, advertising cases and market analysis reports;

[0030] Business marketing knowledge and data sources are labeled and classified separately to obtain the marketing key points of each data.

[0031] In the above implementation process, commercial marketing knowledge is injected into the set model to improve the model's understanding of advertisement generation.

[0032] Furthermore, the step of fine-tuning the set model according to the data set includes:

[0033] Divide the dataset into training, validation and test sets;

[0034] Load and initialize the set image generation model;

[0035] Integrate a custom fine-tuning module into the model as an additional layer to adapt to specific domains and tasks;

[0036] Setting training parameters, wherein the training parameters include learning rate, batch size and training cycle;

[0037] Pre-training using data from commercial marketing knowledge;

[0038] Use data from travel scenarios for fine-tuning;

[0039] Use the validation set to evaluate the fine-tuned model and generate evaluation results. The evaluation indicators include the quality, relevance, and attractiveness of the generated images.

[0040] Based on the evaluation results, adjust the model parameters and training strategies to optimize the model.

[0041] In the above implementation process, commercial marketing knowledge data is used for pre-training to enable the model to understand the basic concepts and strategies of commercial marketing; travel scenario data is used for further fine-tuning to enhance the model's understanding and generation capabilities for specific scenarios; the fine-tuned model is evaluated using a validation set, and based on the evaluation results, the model parameters and training strategies are adjusted to further optimize the model.

[0042] Furthermore, constructing a prompt word template according to material requirements and inputting the prompt word template into the fine-tuned model to generate advertising materials includes:

[0043] Constructing a prompt word template according to material requirements, wherein the structure of the prompt word template includes scene description, emotion description and material type;

[0044] Use the fine-tuned model and input the constructed prompt word template to generate advertising materials;

[0045] Evaluate the quality of the generated materials and make optimization adjustments.

[0046] In the above implementation process, the fine-tuned model is used and the constructed prompt word template is input to generate advertising materials that meet expectations; the quality of the generated materials is evaluated to ensure that they meet advertising requirements, and necessary optimization adjustments are made.

[0047] Furthermore, it also includes:

[0048] Through the user feedback mechanism, collect users' evaluation of the generated materials and obtain feedback data;

[0049] Adjust and optimize the model based on feedback data.

[0050] In the above implementation process, the model is further adjusted and optimized based on the feedback data to improve its generation capability and material quality.

[0051] In a second aspect, an embodiment of the present application provides an advertising material generating device, including:

[0052] Scenario definition module, used to obtain different travel scenarios;

[0053] Material analysis module, used to analyze material needs based on travel scenarios;

[0054] The data processing module is used to label and preprocess the collected data according to the set labels to obtain the data set;

[0055] Model fine-tuning module, used to fine-tune the set model according to the data set;

[0056] The advertisement generation module is used to construct a prompt word template according to the material requirements, and input the prompt word template into the fine-tuned model to generate advertisement materials.

[0057] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0058] A processor, a memory and a bus, wherein the processor is connected to the memory via the bus, and the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, they are used to implement the advertising material generation method as described above.

[0059] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a server, the advertising material generation method as described above is implemented.

[0060] In a fifth aspect, an embodiment of the present application provides a computer program product, wherein the computer program product includes instructions, and when the instructions are executed by a computer, the computer implements the advertising material generation method as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 A schematic diagram of a flow chart of a method for generating advertising materials provided in an embodiment of the present application;

[0063] Figure 2 It is a structural schematic diagram of an advertising material generating device provided in an embodiment of the present application;

[0064] Figure 3 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0065] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.

[0066] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0067] Traditional image editing software relies on manual design and is inefficient because the number of designers is limited and each designer has limited working time and creativity, so it is impossible to quickly generate a large number of advertising images. At the same time, manual production is prone to problems such as inconsistent styles and high error rates.

[0068] The general AIGC model lacks specificity in the travel field. Due to the lack of in-depth understanding and pre-training in the travel field, the generated images may not meet the characteristics of the travel scene. For example, in the urban commuting scene, the elements of speed and convenience need to be reflected, and the general model may not be able to accurately grasp these points. When the general AIGC model generates requirements in language expression, the generated effect is very different from the effect required in the travel field. In addition, without considering user preference data, it is difficult to meet the personalized needs of different users.

[0069] Based on this, the embodiment of the present application proposes a method for generating advertising materials to build an AIGC advertising material content generation solution for travel scenarios, improve the efficiency and quality of advertising material generation, and meet the personalized needs in the travel field.

[0070] Please see Figure 1 , Figure 1 A schematic diagram of a method for generating advertising material provided in an embodiment of the present application. The method for generating advertising material includes:

[0071] 100. Obtain different travel scenarios.

[0072] Optionally, define the range of travel scenarios; for each scenario, set different scenario factors, wherein the scenario factors include but are not limited to time, weather and mood; wherein, by acquiring different travel scenarios, it is ensured that the specific needs of users in the travel field are accurately captured.

[0073] Specifically, first, we need to define the scope of travel scenarios, such as city commuting, holiday travel, business trips, returning home at night, etc. For each scenario, we consider different times (early morning, afternoon, night), weather (sunny, rainy, snowy), and emotions (nervous, relaxed, excited).

[0074] Exemplarily, the time may include specific time points (such as noon, dawn, evening, midnight, six in the morning), time periods (such as breakfast, morning, morning, evening, late at night), travel duration (such as less than 30 minutes, 1 to 2 hours, more than 3 hours, more than 5 hours) and travel dates (such as weekdays, holidays, long holidays, short holidays), etc.

[0075] Exemplarily, the weather may include temperature-related (such as hot, warm, cool, cold, freezing, and scorching), humidity-related (such as damp, dry, moist, and hot), clear and cloudy (such as clear, sunny, cloudy, overcast, foggy, and dark), precipitation phenomena (such as rain: drizzle, heavy rain, rainstorm, drizzle, and showers; snow: heavy snow, light snow, blizzard, hail, and frost), wind force and direction (such as breeze, gentle breeze, strong wind, gale, typhoon, hurricane, northeast wind, southwest wind), special weather phenomena (such as rainbows, rosy clouds, auroras, tornadoes, sandstorms, icicles, fog, lightning, thunder), and climate changes (such as clearing up, turning cloudy, warming, cooling down, and rapid changes), etc.

[0076] Exemplarily, travel scenarios may include commuting (such as commuting to get off work and school), business activities (such as shopping in malls, purchasing in markets, bank transactions, and dining in restaurants), obtaining public services (such as visiting government agencies, medical institutions (medical consultations, physical examinations), libraries, post offices, etc.), social and family affairs (attending friends' gatherings, family gatherings, such as celebrations, daily family affairs, such as picking up and dropping off children, caring for the elderly, etc.), education and cultural participation (such as learning in various educational institutions; training; cultural and artistic activities: such as visiting museums, art galleries, theater performances, etc.), leisure and entertainment (such as parks, gyms, cinemas, and tourist attractions), emergency travel (responding to emergencies, such as emergency medical treatment, responding to natural disasters, and emergency rescue), cargo transportation, travel for special needs (such as travel by the elderly, pregnant women, etc.), sightseeing, etc.

[0077] Exemplarily, emotions may include basic expressions, specifically including anger, disgust, fear, happiness, sadness, and surprise.

[0078] 200. Conduct material demand analysis based on travel scenarios.

[0079] Specifically, based on travel scenarios, determine the types of materials required for different types of advertisements, and screen materials that meet specific resolution standards; material types include but are not limited to high-definition pictures, video clips, and animated images; each type of material should meet specific resolution standards to ensure that it is attractive enough.

[0080] For example, for pictures, it is recommended to use a resolution of no less than 1920x1080; for videos, the resolution is at least 1080p and the frame rate is above 30fps. GPT4V generates materials: Use GPT4V to generate corresponding materials for each travel scenario. Accurate generation can be achieved by describing specific scenes, emotions, and the required material types. For example, "generate a video describing the scene of going home at night, emphasizing the warm and safe atmosphere, with a resolution of 1080p."

[0081] 300. According to the set labels, the collected data are labeled and preprocessed to obtain a data set.

[0082] 310. Label the collected pictures according to the set tags; wherein the set tags include but are not limited to scene, emotion, time and weather.

[0083] Specifically, clearly label the collected images, select a suitable image annotation tool, which can be a software platform, an online service, or a dedicated annotation application, go through each image one by one and label each image, and verify the results against the standard.

[0084] 320. Screen the collected images to remove blurry, distorted or irrelevant images.

[0085] 330. Unify the size of pictures.

[0086] Specifically, the images are standardized: the image size is unified to ensure that it meets the input requirements of the Stable Diffusion model (for example, 256x256 pixels).

[0087] 340. Enhance the image.

[0088] Optionally, perform image enhancement: perform appropriate data enhancement (such as rotation, scaling, cropping, color adjustment, etc.) to increase sample diversity.

[0089] 350. Unify the format of setting labels.

[0090] Specifically, standardize the setting labels: unify the label format to ensure the consistency and usability of the labels.

[0091] 360. Combine the processed images and labels to obtain the dataset.

[0092] As mentioned above, the input images are processed to ensure that they meet the input requirements of the model, while increasing sample diversity and processing the labels to ensure their consistency and availability.

[0093] 400. Fine-tune the set model according to the data set.

[0094] Optionally, the Stable Diffusion model is used to fine-tune the found materials and commercial marketing knowledge through LoRA, and a travel scenario prompt word template is constructed to generate advertising materials that meet expectations.

[0095] 410. Divide the dataset into training set, validation set and test set.

[0096] Specifically, a dataset of processed images and labels is used, and the dataset is divided into a training set, a validation set, and a test set.

[0097] Optionally, the training set is used to train the model; the validation set is used to adjust the model parameters and hyperparameters during the training process to find the best configuration; and the test set is used to evaluate the performance of the model after the model training is completed.

[0098] 420. Load and initialize the set image generation model.

[0099] Specifically, load the basic model, that is, initialize the Stable Diffusion basic model.

[0100] Optionally, you can develop and train deep learning models through TensorFlow or PyTorch; generate materials through GPT4V; and post-process and optimize the materials through image editing software (such as Adobe Photoshop).

[0101] Optionally, high-performance GPUs can be used to accelerate the training and reasoning process of deep learning models; large amounts of travel scenario data and user preference data can be stored using high-capacity storage devices.

[0102] 430. A set of fine-tuning modules are integrated into the model as an additional layer to adapt to specific domains and tasks.

[0103] Optionally, a LoRA module can be integrated into the model as an additional layer adapted to specific domains and tasks.

[0104] 440. Set training parameters, where the training parameters include a learning rate, a batch size, and a training cycle.

[0105] Optional, training parameter settings: Learning Rate: usually between 1e-4 and 1e-5; Batch Size: depending on the GPU memory setting, generally 16 to 32 images; Training Cycles (Epochs): depends on the amount of data and model effect, generally 10 to 20 cycles.

[0106] 450. Use data from business marketing knowledge for pre-training.

[0107] Among them, commercial marketing knowledge data is used for pre-training to enable the model to understand the basic concepts and strategies of commercial marketing.

[0108] 460. Use travel scenario data for fine-tuning.

[0109] Among them, travel scenario data is used for further fine-tuning to improve the model's understanding and generation capabilities for specific scenarios.

[0110] 470. Use the validation set to evaluate the fine-tuned model and generate evaluation results; the evaluation indicators include the quality, relevance and attractiveness of the generated images.

[0111] 480. Based on the evaluation results, adjust the model parameters and training strategies to optimize the model.

[0112] Optionally, analyze the evaluation results (which may include analyzing performance indicators and viewing samples of model prediction errors to understand the types and causes of errors), adjust model parameters, modify training strategies, repeat iterations, etc. based on the evaluation results. Thus, through continuous iteration and adjustment, the performance of the model can be gradually optimized to better adapt it to the task and data set.

[0113] On the basis of the above embodiment, the advertising material generation method provided in the embodiment of the present application can also be embodied as follows: before fine-tuning the set model according to the data set, it also includes:

[0114] Fine-tune business marketing knowledge, including:

[0115] Inject commercial marketing knowledge into the set model; wherein the commercial marketing knowledge includes but is not limited to advertising strategies, user psychology and market analysis; thereby, the model's understanding of advertising generation is enhanced.

[0116] Collect relevant data sources, including but not limited to commercial marketing literature, advertising cases and market analysis reports.

[0117] Business marketing knowledge and data sources are labeled and classified separately to obtain the marketing key points of each data.

[0118] Exemplarily, the model is fine-tuned through business marketing knowledge, including: learning business knowledge, adjusting the model's account structure, system ID and delivery device according to services / products, keywords / titles, videos / pictures, conversations / historical information; thereby, introducing the LLM (Large Language Model) of business knowledge, understanding advertising content and advertising delivery information.

[0119] 500. Construct a prompt word template according to the material requirements, and input the prompt word template into the fine-tuned model to generate advertising materials.

[0120] 510. Construct a prompt word template according to material requirements, wherein the structure of the prompt word template includes scene description, emotion description and material type.

[0121] Optionally, construct a Prompt word template, where the template structure includes: scenario description: specific travel scenario (such as "going home at night"); emotion description: the user's emotion in the scenario (such as "warmth and safety"); material type: generated material type and quality requirements (such as "video with a resolution of 1080p").

[0122] For example, a video is generated that describes a scene of returning home at night, emphasizing a warm and safe atmosphere, with a resolution of 1080p.

[0123] 520. Use the fine-tuned model and input the constructed prompt word template to generate advertising materials.

[0124] Optionally, use the fine-tuned Stable Diffusion model and input the constructed Prompt template to generate advertising creatives that meet expectations.

[0125] Exemplarily, the Prompt word template includes face + tone emotion recognition, user group labels, travel itinerary data, weather conditions, supply and demand heat conditions, etc., to generate advertisements / videos / pictures and texts that meet the requirements.

[0126] For example, the model can also be fine-tuned through advertising creative LLM, creative materials, company brands, industry categories, marketing selling points, etc.

[0127] 530. Evaluate the quality of the generated materials and make optimization adjustments.

[0128] Optionally, perform a quality assessment on the generated materials to ensure that they meet advertising requirements and make necessary optimization adjustments.

[0129] As described above, the embodiments of the present application obtain different travel scenarios; perform material demand analysis based on the travel scenarios; annotate and preprocess the collected data according to the set labels to obtain a data set; fine-tune the set model according to the data set; construct a prompt word template according to the material demand, and input the prompt word template into the fine-tuned model to generate advertising materials; it is possible to generate advertising materials that are highly relevant to the travel scenarios and meet the personalized needs of users, thereby improving the attractiveness and effectiveness of advertisements and reducing labor costs and production time.

[0130] On the basis of the above embodiments, the advertising material generation method provided in the embodiments of the present application can also be embodied as follows:

[0131] Through the user feedback mechanism, collect users' evaluations of the generated materials to obtain feedback data; adjust and optimize the model based on the feedback data; thus, further adjust and optimize the model based on the feedback data to improve its generation capability and material quality.

[0132] For example, feedback learning is performed based on user preference data to optimize advertising duration, sequence, and run a data flywheel.

[0133] Optionally, the embodiments of the present application can be combined with other data sources, such as social media data, user evaluation data, etc., to enrich the sources of user preference data and improve the personalization of advertising images; develop special advertising image generation software, integrate travel scenario data and user preference data, and provide users with more convenient advertising image generation services.

[0134] As mentioned above, the pre-training of personalized scenarios in the travel field in the embodiment of the present application: the Stable Diffusion model is pre-trained for business in the travel field, so that it can better understand the characteristics of travel scenarios and user needs. Material generation and optimization: Use large models to generate high-quality materials, and perform screening and optimization to improve the quality of advertising images. Application of user preference data: Combine user preference data to generate personalized advertising images to improve user satisfaction and advertising effectiveness.

[0135] The above steps are not to be performed in a strict order as described in the numbers, but should be understood as an overall solution.

[0136] In the second aspect, based on the above embodiments, Figure 2 This is a schematic diagram of the structure of an advertising material generation device provided in an embodiment of the present application. Figure 2 The advertising material generation device provided in this embodiment specifically includes: a scene definition module 201, a material analysis module 202, a data processing module 203, a model fine-tuning module 204 and an advertisement generation module 205.

[0137] Among them, the scenario definition module 201 is used to obtain different travel scenarios; the material analysis module 202 is used to perform material demand analysis based on travel scenarios; the data processing module 203 is used to label and preprocess the collected data according to the set labels to obtain a data set; the model fine-tuning module 204 is used to fine-tune the set model according to the data set; the advertisement generation module 205 is used to construct a prompt word template according to the material demand, and input the prompt word template into the fine-tuned model to generate advertising materials.

[0138] As described above, the embodiments of the present application obtain different travel scenarios; perform material demand analysis based on the travel scenarios; annotate and preprocess the collected data according to the set labels to obtain a data set; fine-tune the set model according to the data set; construct a prompt word template according to the material demand, and input the prompt word template into the fine-tuned model to generate advertising materials; it is possible to generate advertising materials that are highly relevant to the travel scenarios and meet the personalized needs of users, thereby improving the attractiveness and effectiveness of advertisements and reducing labor costs and production time.

[0139] The advertising material generation device provided in the embodiment of the present application can be used to execute the advertising material generation method provided in the above embodiment, and has corresponding functions and beneficial effects.

[0140] In a third aspect, an embodiment of the present application further provides an electronic device, which can integrate the advertising material generating device provided in an embodiment of the present application. Figure 3 Schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Figure 3The electronic device includes: an input device 33, an output device 34, a memory 32 and one or more processors 31; the memory 32 is used to store one or more programs; when the one or more programs are executed by the one or more processors 31, the one or more processors 31 implement the advertising material generation method provided in the above embodiment. The input device 33, the output device 34, the memory 32 and the processor 31 can be connected by a bus or other means. Figure 3 The example of connecting through bus is taken in the following.

[0141] The processor 31 executes the software programs, instructions and modules stored in the memory 32 to execute various functional applications and data processing of the device, that is, to implement the above-mentioned advertising material generation method.

[0142] The electronic device provided above can be used to execute the advertising material generation method provided in the above embodiment, and has corresponding functions and beneficial effects.

[0143] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, the computer-readable storage medium including a stored computer program; wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the advertising material generation method as described above, and can achieve the same beneficial effects as described above.

[0144] Of course, the storage medium containing computer executable instructions provided in the embodiments of the present application is not limited to the advertising material generation method described above, and the computer executable instructions can also execute related operations in the advertising material generation method provided in any embodiment of the present application.

[0145] In a fifth aspect, the embodiments of the present application also provide a computer program product, and the methods described in the various embodiments of the present application can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instruction is loaded and executed on a computer, the processes or functions described in the various embodiments of the present application are executed in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer grid, a grid device, a user device, a core network device, an OAM (Open Application Model) or other programmable device.

[0146] The computer program or instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program or instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; it may also be an optical medium, such as a digital video disk; it may also be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both volatile and non-volatile types of storage media.

[0147] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0148] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0149] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for enabling an electronic device (which can be a personal computer, a server, or a grid device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0150] The above description is only an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0151] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0152] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

Claims

1. A method for generating advertising material, characterized in that: include: Get different travel scenarios; Conduct material demand analysis based on travel scenarios; According to the set labels, the collected data is labeled and preprocessed to obtain a data set; Fine-tune the set model according to the data set; A prompt word template is constructed according to the material requirements, and the prompt word template is input into the fine-tuned model to generate advertising materials.

2. The method for generating advertising material according to claim 1, characterized in that: The obtaining of different travel scenarios includes: Define the scope of travel scenarios; For each scene, different scene factors are set, wherein the scene factors include but are not limited to time, weather and mood.

3. The method for generating advertising material according to claim 1, characterized in that: The material demand analysis based on the travel scenario includes: Based on travel scenarios, determine the types of materials required for different types of advertisements, and filter materials that meet specific resolution standards; among them, material types include but are not limited to high-definition pictures, video clips and animated images.

4. The method for generating advertising material according to claim 1, characterized in that: The collected data is labeled and preprocessed according to the set labels to obtain a data set, including: Label the collected pictures according to the set tags; wherein the set tags include but are not limited to scene, emotion, time and weather; Screening the collected images to remove blurry, distorted, or irrelevant images; Unify the size of the pictures; Enhance the image; Unify the format of setting labels; Combine the processed images and labels to get the dataset.

5. The method for generating advertising material according to claim 1, characterized in that: Before fine-tuning the set model according to the data set, the method further includes: Fine-tune business marketing knowledge, including: Injecting commercial marketing knowledge into the set model; where the commercial marketing knowledge includes but is not limited to advertising strategies, user psychology and market analysis; Collect relevant data sources, including but not limited to commercial marketing literature, advertising cases and market analysis reports; Business marketing knowledge and data sources are labeled and classified separately to obtain the marketing key points of each data.

6. The method for generating advertising material according to claim 1, characterized in that: The set model is fine-tuned according to the data set, including: Divide the dataset into training, validation and test sets; Load and initialize the set image generation model; Integrate a custom fine-tuning module into the model as an additional layer to adapt to specific domains and tasks; Setting training parameters, wherein the training parameters include learning rate, batch size and training cycle; Pre-training using data from commercial marketing knowledge; Use data from travel scenarios for fine-tuning; Use the validation set to evaluate the fine-tuned model and generate evaluation results. The evaluation indicators include the quality, relevance, and attractiveness of the generated images. Based on the evaluation results, adjust the model parameters and training strategies to optimize the model.

7. The method for generating advertising material according to claim 1, characterized in that: The step of constructing a prompt word template according to the material requirements and inputting the prompt word template into the fine-tuned model to generate the advertising material includes: Constructing a prompt word template according to material requirements, wherein the structure of the prompt word template includes scene description, emotion description and material type; Use the fine-tuned model and input the constructed prompt word template to generate advertising materials; Evaluate the quality of the generated materials and make optimization adjustments.

8. The method for generating advertising material according to claim 1, characterized in that: Also includes: Through the user feedback mechanism, collect users' evaluation of the generated materials and obtain feedback data; Adjust and optimize the model based on feedback data.

9. An advertising material generating device, characterized in that: include: Scenario definition module, used to obtain different travel scenarios; Material analysis module, used to analyze material needs based on travel scenarios; The data processing module is used to label and preprocess the collected data according to the set labels to obtain the data set; Model fine-tuning module, used to fine-tune the set model according to the data set; The advertisement generation module is used to construct a prompt word template according to the material requirements, and input the prompt word template into the fine-tuned model to generate advertisement materials.

10. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the processor is connected to the memory via the bus, and the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, they are used to implement the advertising material generation method according to any one of claims 1 to 8.

Images