CN114648390B

CN114648390B - Method and system for generating agricultural product promotion advertisement

Info

Publication number: CN114648390B
Application number: CN202210396969.XA
Authority: CN
Inventors: 孙彤; 黄桂恒
Original assignee: Brake Agricultural Big Data Technology Group Co ltd
Current assignee: Brake Agricultural Big Data Technology Group Co ltd
Priority date: 2021-06-22
Filing date: 2021-06-22
Publication date: 2022-12-02
Anticipated expiration: 2041-06-22
Also published as: CN113435967A; CN113435967B; CN114648390A

Abstract

The embodiment of the specification discloses a method and a system for generating an agricultural product promotion advertisement. The method comprises the following steps: determining candidate agricultural products based on keywords input by a user; judging whether the candidate agricultural products belong to preset agricultural products, wherein the preset agricultural products are agricultural products with quality artifacts easy to artificially manufacture; in response to not, displaying a first preset advertisement matched with the candidate agricultural product; in response, processing relevant information of the candidate agricultural product based on a machine learning model, and evaluating the truth of the candidate agricultural product; responding to the fact that the target truth meets a first preset condition, and determining respective promotion labels of growth form images and current form images of the candidate agricultural products in each growth stage; and generating and displaying a second preset advertisement based on the first preset advertisement and the promotion label.

Description

Method and system for generating agricultural product promotion advertisement

Description of the different cases

The application is a divisional application which is provided by Chinese application with the application date of 2021, 06, 22 and the application number of 2021106950258, and the invention is named as 'a method and a system for automatically determining marketing content of agricultural products'.

Technical Field

The specification relates to the technical field of agriculture, in particular to a method and a system for generating agricultural product promotion advertisements.

Background

With the development of economy and the improvement of living standard, people pay more and more attention to diet health. Therefore, people pay more and more attention to the quality of agricultural products in the process of purchasing the agricultural products. In practical situations, the quality or variety of agricultural products cannot be completely reflected by only depending on the appearance of the agricultural products, and the sales volume of the agricultural products and the shopping experience of users can be improved by generating corresponding marketing contents (e.g., advertisements) for agricultural products of different qualities or different varieties.

Therefore, there is a need to provide a method and system for automatically determining marketing content of agricultural products to present different advertisements based on different agricultural products.

Disclosure of Invention

One embodiment of the present specification provides a method for generating an agricultural product promotion advertisement. The method comprises the following steps: determining candidate agricultural products based on keywords input by a user; judging whether the candidate agricultural products belong to preset agricultural products, wherein the preset agricultural products are agricultural products easy to artificially manufacture quality artifacts; in response to not, displaying a first preset advertisement matched with the candidate agricultural product; responding to the agricultural product, processing relevant information of the candidate agricultural product based on a machine learning model, and evaluating the target truth degree of the candidate agricultural product; responding to the fact that the target truth meets a first preset condition, and determining respective promotion labels of growth form images and current form images of the candidate agricultural products in each growth stage; and generating and displaying a second preset advertisement based on the first preset advertisement and the promotion label.

One of the embodiments of the present specification provides a system for generating an agricultural product promotion advertisement, where the system includes: the first determination module is used for determining candidate agricultural products based on keywords input by a user; the judging module is used for judging whether the candidate agricultural products belong to preset agricultural products, and the preset agricultural products are agricultural products with quality artifacts easy to artificially manufacture; a first display module, configured to display a first preset advertisement matching the candidate agricultural product in response to not; the second display module is used for responding to the request, processing the relevant information of the candidate agricultural products based on the machine learning model and evaluating the target truth of the candidate agricultural products; the second determining module is used for determining respective promotion labels of the growth form images and the current form images of the candidate agricultural products in each growth stage in response to the fact that the target truth meets a first preset condition; the generating module is used for generating a second preset advertisement based on the first preset advertisement and the promotion label; and the third display module is used for displaying the second preset advertisement.

One of the embodiments of the present specification provides an apparatus for generating an agricultural product promotion advertisement, the apparatus includes a processor and a memory, the memory is configured to store instructions, and the processor is configured to execute the instructions to implement operations corresponding to the method for automatically determining marketing content of an agricultural product according to any one of the foregoing items.

One of the embodiments of the present specification provides a computer-readable storage medium, which stores computer instructions, and when the computer instructions are executed by a processor, the computer instructions implement the operations corresponding to the method for generating an agricultural product promotion advertisement according to any one of the foregoing embodiments.

Drawings

The present description will be further explained by way of exemplary embodiments, which will be described in detail by way of the accompanying drawings. These embodiments are not intended to be limiting, and in these embodiments like numerals are used to indicate like structures, wherein:

FIG. 1 is an exemplary flow diagram of a method of automatically determining marketing content of agricultural products, according to some embodiments of the present description;

FIG. 2 is an exemplary flow chart illustrating evaluation of the truth of a candidate agricultural product according to some embodiments of the present description;

FIG. 3 is a schematic diagram of a structure of a machine learning model according to some embodiments of the present description;

FIG. 4 is an exemplary flow diagram illustrating the generation of a second pre-set advertisement in accordance with some embodiments of the present description; and

FIG. 5 is a block diagram of a system for automatically determining marketing content of agricultural products, in accordance with some embodiments of the present description.

Detailed Description

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings used in the description of the embodiments will be briefly described below. It is obvious that the drawings in the following description are only examples or embodiments of the present description, and that for a person skilled in the art, the present description can also be applied to other similar scenarios on the basis of these drawings without inventive effort. Unless otherwise apparent from the context, or otherwise indicated, like reference numbers in the figures refer to the same structure or operation.

It should be understood that "system", "device", "unit" and/or "module" as used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, other words may be substituted by other expressions if they accomplish the same purpose.

As used in this specification and the appended claims, the terms "a," "an," "the," and/or "the" are not intended to be inclusive in the singular, but rather are intended to be inclusive in the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that steps and elements are included which are explicitly identified, that the steps and elements do not form an exclusive list, and that a method or apparatus may include other steps or elements.

Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of the present specification. It should be understood that the preceding or following operations are not necessarily performed in the exact order in which they are performed. Rather, the steps may be processed in reverse order or simultaneously. Meanwhile, other operations may be added to or removed from these processes.

FIG. 1 is an exemplary flow diagram of a method of automatically determining marketing content of agricultural products, according to some embodiments described herein. As shown in fig. 1, the process 100 may include the following steps.

Step 102, determining candidate agricultural products based on keywords input by a user. In some embodiments, step 102 may be performed by the first determination module 510.

In some embodiments, the keywords may be words that the user enters for querying the agricultural commodity. Agricultural products may include, but are not limited to, melons, fruits, and vegetables. Such as watermelon, tomato, corn, cabbage, etc. In some embodiments, the user may enter the keyword through an application of the terminal device. In some embodiments, a terminal device may include a device capable of communication and/or data transfer. In some embodiments, the terminal device may include, but is not limited to, one or more of the following: sales terminals, mobile phones, tablet computers, notebook computers, and the like. In some embodiments, the input mode of the keyword may include, but is not limited to, one or more of the following: text, audio, and images.

In some embodiments, the candidate agricultural products may be agricultural products searched for selection by the user based on keywords entered by the user. The number of candidate agricultural products may include one or more. In some embodiments, when the number of candidate agricultural products includes a plurality, the plurality of candidate agricultural products may be the same or different varieties. For example, if the keyword entered by the user is "tomato," the candidate agricultural products may include, but are not limited to, one or more of the following: native tomatoes, field tomatoes, cherry tomatoes, and the like. As another example, if the keyword entered by the user is "hainan fruit," then the candidate agricultural products may include, but are not limited to, one or more of the following: coconut, pineapple, litchi, etc.

And 104, judging whether the candidate agricultural products belong to preset agricultural products. In some embodiments, step 104 may be performed by the decision module 520.

In some embodiments, the preset agricultural products can be flexibly set according to actual requirements. In some embodiments, the preset agricultural product may include, but is not limited to: the agricultural products with quality false images are easy to artificially manufacture. It can be understood that during the growth process of the preset agricultural product, the appearance of the agricultural product can present better quality through preset human intervention (for example, human adding of ceresine and the like), so as to produce good quality false images. In some embodiments, the preset agricultural products may include, but are not limited to, tomatoes and/or watermelons and the like.

And 106, in response to the answer, displaying a first preset advertisement matched with the candidate agricultural product. This step may be performed by first presentation module 530.

In some embodiments, the first preset advertisement may be an advertisement that matches the candidate agricultural product. In some embodiments, different candidate agricultural products may match different first preset advertisements. In some embodiments, the database and/or the storage device may be pre-stored with first preset advertisements matching different agricultural products. In some embodiments, when the determining module 520 determines that the candidate agricultural product does not belong to the preset agricultural product, the first presentation module 530 may call and present a first preset advertisement matching the candidate agricultural product from a database and/or a storage device.

In some embodiments, the first preset advertisement may be presented in a manner including one or more of the following: text, audio, images, video, and the like. In some embodiments, the manner in which the first preset advertisement is presented may be determined based on user settings and/or user information. In some embodiments, the user information may include, but is not limited to, one or more combinations of the following: age, gender, occupation, preference, etc. For example, if the user sets video display, the display mode of the first preset advertisement is video. For another example, if the user sets the image display and the user preference is the video display, the display mode of the first preset advertisement may include the image and/or the video.

And 108, in response, processing the relevant information of the candidate agricultural products based on the machine learning model, evaluating the real situation of the candidate agricultural products, and displaying a second preset advertisement matched with the evaluation result. In some embodiments, step 108 may be performed by second presentation module 540.

In some embodiments, the machine learning model may be a pre-trained model. The trained machine learning model can process relevant information of the candidate agricultural products, and the real situation of the candidate agricultural products is evaluated to obtain an evaluation result. In some embodiments, the machine learning model may include one or more pre-trained models. In some embodiments, the machine learning model may include one or more combinations of the following: neural Networks (NN), convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN), among others. For specific details of the machine learning model, reference may be made to fig. 2 and fig. 3 and their associated descriptions, which are not repeated herein.

In some embodiments, the information related to the candidate agricultural product may be related to a growth process and/or a current morphology of the candidate agricultural product. In some embodiments, the information related to the candidate agricultural product may include at least one or more of: the environment data of the candidate agricultural products in each growth stage, the growth form images of the candidate agricultural products in each growth stage, the current form images of the candidate agricultural products and the like. Specific details of the relevant information about the candidate agricultural product can be seen in fig. 2 and the relevant description thereof, which are not repeated herein.

In some embodiments, the truth of the candidate agricultural product may reflect at least one truth information of the agricultural product. For example, the presence or absence of preset human intervention, actual maturity, actual sweetness level, etc. of the agricultural product during growth can be reflected. Actual maturity refers to the current true maturity of the agricultural product. The actual maturity may be different from the maturity exhibited by the produce. The actual sweetness is the current true sweetness of the agricultural product. The actual sweetness may be different from the sweetness exhibited by the agricultural product. For example, the agricultural product is ripened or reddened, etc., so that the sweetness or ripeness exhibited does not match the reality. In some embodiments, the result of the evaluation of the truth of the candidate agricultural product may be represented by a numerical value. In some embodiments, the evaluation of the truth of the candidate agricultural product may be represented by a target truth. For specific details of the target truth, reference may be made to fig. 2 and the related description thereof, which are not repeated herein.

In some embodiments, the second preset advertisement may be an advertisement that matches the evaluation result. In some embodiments, the second pre-set advertisement may be an advertisement that matches the target liveness. In some embodiments, the processing device may adjust the first preset advertisement based on the target degree of realism and/or information related to the target degree of realism to generate the second preset advertisement. For details of generating the second preset advertisement, reference may be made to fig. 4 and the related description thereof, which are not repeated herein.

In some embodiments, the display manner of the second preset advertisement is similar to that of the first preset advertisement, which can be referred to the above step 106 and the related description, and is not repeated herein.

In some embodiments, the processing device may generate and present a second preset advertisement based on the target liveness satisfying the first preset condition. The first preset condition may be specifically set according to actual requirements. In some embodiments, the first preset condition may include, but is not limited to: the target truth is larger than a preset threshold value. In some embodiments, the preset threshold may be specifically set according to actual conditions. E.g., 0.9 and 0.95, etc.

In some embodiments, the processing device may send warning information to the target object based on the target truth meeting a second preset condition. In some embodiments, the second preset condition may be specifically set according to actual requirements. The second preset condition may include, but is not limited to: the target truth is smaller than a preset threshold value. The preset threshold may refer to the above description, and is not described herein again.

In some embodiments, the target object may be a terminal device and/or an administrator of an application of the terminal device. Such as merchants and any persons with administrative privileges, etc. In some embodiments, the alert information may include multiple forms. Such as whistling, lighting, voice announcements, text displays, and the like. In some embodiments of the present specification, the warning information may prompt relevant management personnel (for example, a merchant) that the truth degree of a corresponding candidate agricultural product is too low, so that the management personnel can determine whether to continue to sell or put on shelf the agricultural product, thereby avoiding a consumer from purchasing an agricultural product with an unqualified quality, and improving the shopping experience of the user.

In some embodiments of the specification, a candidate agricultural product belonging to a preset agricultural product is subjected to truth evaluation, and a corresponding second preset advertisement is displayed based on an evaluation result, so that a first preset advertisement corresponding to the candidate agricultural product can be adjusted in real time according to the evaluation result, and the first preset advertisement is optimized, so that the sales volume of the candidate agricultural product is increased under the condition that the quality of the candidate agricultural product is in compliance.

FIG. 2 is an exemplary flow chart illustrating evaluation of the truth of a candidate agricultural product according to some embodiments of the present description. As shown in fig. 2, the process 200 may include the following

steps

202 and 204. In some embodiments,

steps

202 and 204 may be performed by second presentation module 540.

Step 202, at least processing the environment data of the candidate agricultural products at each growth stage, the growth form images of the candidate agricultural products at each growth stage and the current form images of the candidate agricultural products by using the machine learning model to obtain the target truth of the candidate agricultural products.

In some embodiments, the various growth stages of the agricultural product may include at least a germination stage, a seedling stage, a flowering stage, a fruiting stage, and the like. In some embodiments, the environmental data may include factors that affect growth of the candidate agricultural product. In some embodiments, the environmental data may include at least light intensity, oxygen concentration, growth temperature, soil composition, and the like. In some embodiments, the intensity of the illumination, the oxygen concentration, and the growth temperature may be monitored by sensors. For example, the intensity of light is monitored by a light sensor, the concentration of oxygen is monitored by an oxygen control sensor, the growth temperature is monitored by a temperature control sensor, and the like. In some embodiments, the soil may be detected by a detection department to obtain the soil composition.

In some embodiments, the growth morphology images of the candidate agricultural product at various growth stages may refer to appearance images of the candidate agricultural product at a germination stage, a seedling stage, a flowering stage, and a fruiting stage. In some embodiments, the current morphology image of the candidate agricultural product may refer to a current appearance image of the candidate agricultural product. In some embodiments, the growth modality image and the current modality image may be acquired by a camera and/or photo-enabled device. Such as a camera, video camera, etc.

For some details regarding the machine learning model, see step 108 above and its associated description. In some embodiments, the machine learning model may be a model, such as a Neural Network (NN), convolutional Neural Network (CNN), or Recurrent Neural Network (RNN), among others. All relevant information of the candidate agricultural products can be processed through the model, and the target truth degree of the candidate agricultural products is obtained. In some embodiments, the machine learning model may be trained using a plurality of first training samples carrying labels. In some embodiments, the first training sample may include at least environmental data of the sample agricultural product at various growth stages, a growth morphology image of the sample agricultural product at the various growth stages, and a current morphology image of the sample agricultural product. In some embodiments, the tags may characterize a target truth of the sample agricultural product. In some embodiments, the label may be generated by manual line marking. For example, the target truth of the sample agricultural product is manually defined, and a corresponding label is generated. In some embodiments, the label may be generated by a goodness of the sample agricultural product captured online. For example, there is a preset corresponding relationship between the goodness of appreciation and the target truth, and after the goodness of appreciation of the sample agricultural product is obtained, a label of the corresponding target truth can be generated based on the preset corresponding relationship. In some embodiments, the goodness rating may include a historical user's rating of the agricultural product. Specifically, the parameters of the initial machine learning model may be iteratively updated based on a plurality of first training samples, so that the loss function of the model satisfies a preset condition, for example, the loss function converges, or the loss function value is smaller than a preset value. And completing model training when the loss function meets the preset condition to obtain a trained machine learning model.

In some embodiments, the target truth may reflect at least: whether preset human intervention, the actual maturity of the candidate agricultural product and the actual sweetness of the candidate agricultural product exist in the growth process of the candidate agricultural product. In some embodiments, the preset human intervention may be specifically set according to actual requirements. In some embodiments, the preset human intervention may be a human operation that interferes with the natural growth process of the agricultural product. For example, if the predetermined agricultural product is an agricultural product with quality artifacts easily produced by human, the predetermined human intervention at least includes adding erythrogenin by human.

In some embodiments, the target degree of truth may be inversely related to whether the preset human intervention is likely to exist in the growth process of the candidate agricultural product, i.e., the lower the target degree of truth, the more likely the preset human intervention is likely to exist in the growth process of the candidate agricultural product. In some embodiments, the target degree of truth may be positively correlated to the actual degree of maturity and the actual sweetness of the candidate agricultural product, i.e., the higher the target degree of truth, the higher the degree of maturity and the sweetness of the candidate agricultural product.

And 204, determining the target truth as the evaluation result of the candidate agricultural product.

In some embodiments, the machine learning model may be composed of a plurality of models, and the plurality of models are used to process the information related to the candidate agricultural products respectively to determine the target truth. FIG. 3 is a block diagram of a machine learning model according to some embodiments of the present description. As shown in fig. 3, the machine learning model may include a first predictive model and a second predictive model.

In some embodiments, the first predictive model and the second predictive model may be pre-trained machine learning models. The trained first prediction model can process the environment data of the candidate agricultural products in each growth stage and the current form images of the candidate agricultural products to obtain the current truth of the candidate agricultural products. The trained second prediction model can process the environmental data of the candidate agricultural products in each growth stage and the growth form images of the candidate agricultural products in each growth stage to obtain the truth of the candidate agricultural products in each growth stage. For the training of the first prediction model and the second prediction model, reference may be made to the related description below, and details are not repeated here.

In some embodiments, the first predictive model may include at least a first convolutional neural network layer, a recurrent neural network layer, and a first fully-connected layer. In some embodiments, the processing of the first predictive model may include: processing the current form image of the candidate agricultural product by utilizing the first convolution neural network layer to obtain the current form image characteristics of the agricultural product; processing the environment data of the candidate agricultural products at each growth stage by utilizing the recurrent neural network layer to obtain the growth characteristics of the agricultural products, wherein the growth characteristics of the agricultural products can represent the overall growth characteristics of the agricultural products; and processing the current form image characteristics and the growth characteristics of the agricultural products by utilizing the first full-connection layer to obtain the current truth degree. The recurrent neural network is suitable for processing time series type data, and can better process environmental data of each growth stage by adopting the recurrent neural network to obtain more accurate growth characteristics.

In some embodiments, the second predictive model may include at least a second convolutional neural network layer, a second fully-connected layer, and a third fully-connected layer. The processing of the second predictive model may include: processing the growth form images of the candidate agricultural products in each growth stage by utilizing a second convolutional neural network layer to obtain image characteristics of each growth stage; processing the environmental data of the candidate agricultural products in each growth stage by utilizing the second full-connection layer to obtain the growth characteristics of each growth stage; and processing the image characteristics of each growth stage and the growth characteristics of each growth stage by using the third full-connection layer to obtain the truth of each growth stage.

In some embodiments, the first predictive model and the second predictive model may be trained separately. Specifically, in some embodiments, the first prediction model may be obtained by training a plurality of second training samples carrying the labels. In some embodiments, the second training sample may include at least environmental data of the sample agricultural product at various growth stages and a current morphological image of the sample agricultural product, and the label may characterize a current degree of truth of the sample agricultural product. In some embodiments, the tags may be obtained by manual labeling. Specifically, the parameters of the initial first prediction model may be iteratively updated based on a plurality of second training samples, so that the loss function of the model satisfies a predetermined condition, for example, the loss function converges, or the loss function value is smaller than a predetermined value. And finishing model training when the loss function meets the preset condition to obtain a trained first prediction model. The training mode of the second prediction model is similar to that of the first prediction model, and is not repeated here.

In some embodiments, the first predictive model and the second predictive model may share parameters during training of the two. For example, in some embodiments, the convolutional neural network layers and/or the fully-connected layers in the first predictive model and the second predictive model may share parameters. Specifically, the first convolutional neural network layer and the second convolutional neural network layer may share parameters; the third fully connected layer and the first fully connected layer may share parameters. By sharing the parameters, when the parameters of the first convolutional neural network layer are updated, the parameters of the second convolutional neural network layer are also updated synchronously, so that the parameter sharing of the model is realized. It can be understood that the shared parameters of the first fully-connected layer and the third fully-connected layer are the same, and are not described herein again. Parameters during model training are reduced and model training efficiency is improved through the shared parameters of the first prediction model and the second prediction model.

In some embodiments, the first predictive model and the second predictive model may be trained jointly. It is understood that the machine learning model consisting of the first predictive model and the second predictive model may be obtained by end-to-end training. Similarly, in some embodiments, a machine learning model composed of the first prediction model and the second prediction model may be obtained through training of the plurality of first training samples carrying the labels. For specific details of the training of the first training sample to obtain the machine learning model, refer to step 202 and its related description, which are not repeated herein. The machine learning model consisting of the first prediction model and the second prediction model is trained end to end, namely the first prediction model and the second prediction model are jointly trained, so that the training complexity of the machine learning model can be reduced, the training efficiency of the machine learning model is improved, and meanwhile, the model can learn more characteristic conditions in the training process, so that the model can process more complex actual conditions.

In some embodiments, the current degree of truth and the degrees of truth of the various growth stages may be fused to determine the target degree of truth. In some embodiments, the fusing may include at least weighted summation, and/or weighted averaging, and the like. In some embodiments, the current degree of truth and the degrees of truth for each growth stage have their corresponding weights. In some embodiments, the weight may be set artificially, for example, the truth of the flowering phase and the truth of the fruiting phase may be weighted higher, 0.8. In some embodiments, the weights may also be determined based on other means. In some embodiments, the weight may be determined based on the type of the candidate agricultural product. For example, the candidate agricultural products of type a may be subjected to preset human intervention in the seedling stage, the degree of truth of the seedling stage is low, and the weight of the degree of truth of the seedling stage may be set to a small value, for example, 0.1. For another example, if the candidate agricultural products of type B can be subjected to a predetermined human intervention during the germination period, the truth of the germination period is low, and the weight of the truth of the germination period can be set to a small value, for example, 0.05. The weight is determined according to the type of the agricultural product, preset human intervention can be considered in the process of determining the target truth degree, the validity of the truth degree of the candidate agricultural product possibly provided with the preset human intervention is further ensured, and the accuracy of the final target truth degree is improved.

FIG. 4 is an exemplary flow diagram illustrating the generation of a second pre-set advertisement according to some embodiments of the present description. As shown in fig. 4, the process 400 may include the following steps.

Step 402, determining respective promotion labels of the growth form image of the candidate agricultural product in each growth stage and the current form image of the candidate agricultural product. In some embodiments, step 402 may be performed by the second determination module 570.

For specific details of the growth form image of the candidate agricultural product at each growth stage and the current form image of the candidate agricultural product, reference may be made to step 202 and the related description thereof, which are not repeated herein.

In some embodiments, promotional tags may be used to characterize the priority and/or strength of presentation of images, which may reflect at least whether the images are presented, the order in which the images are presented, and/or the length of time the images are presented, among other things. In some embodiments, the promotional tags can include at least a promotional index that positively correlates with the target degree of truth. It can be appreciated that the higher the promotion index, the higher the target truth.

In some embodiments of the present description, since the promotion index is positively correlated with the target truth, the higher the promotion index is, the higher the target truth of the image with the high target truth is, the priority and/or strength of the image display with the high target truth is improved, so that the consumer can more comprehensively and intuitively know the candidate agricultural products with the high target truth, and further the sales volume of the candidate agricultural products is improved.

In some embodiments, characteristic information of the user may be obtained; and processing the characteristic information of the user by using a third prediction model, and determining a promotion label. In some embodiments, the characteristic information of the user may include at least one or more of, but is not limited to: historical behavior of the user, user information, and the like. In some embodiments, the historical behavior of the user may include, but is not limited to, at least one or more of the following: agricultural products purchased by a user in history, agricultural products browsed by the user in history, agricultural products commented by the user in history, agricultural products complied with by the user in history and the like. For specific details of the user information, reference may be made to step 106 and its related description, which are not described herein again.

In some embodiments, the third predictive model may include a neural network model (NN). In some embodiments, the third prediction model may be obtained by training a plurality of third training samples carrying the label. In some embodiments, the third training sample may include at least feature information of the sample user, and the label may characterize a growth morphology image of the sample agricultural product at each growth stage and a promotion index of a current morphology image of the candidate agricultural product. Specifically, the parameters of the initial third prediction model may be iteratively updated based on a plurality of third training samples, so that the loss function of the model satisfies a preset condition, for example, the loss function converges, or the loss function value is smaller than a preset value. And finishing model training when the loss function meets the preset condition to obtain a trained third prediction model.

In some embodiments, after the feature information of the user is processed by using the third prediction model to determine the promotion index (i.e., the promotion label), the promotion index may be adjusted by the degree of truth. In some embodiments, weighting operation may be performed on the current form image of the candidate agricultural product and/or the promotion label of each growth form image of the candidate agricultural product in each growth stage according to the current truth and/or the truth of each growth stage, so as to obtain the promotion index after each adjustment. For example, the greater the degree of realism, the greater the weight set. It can be seen that the promotion index is adjusted based on the degree of truth (e.g., the current degree of truth and/or the degree of truth of each growth stage) so that the degree of truth is positively correlated with the promotion index (i.e., the promotion label). Because the target truth is determined based on the truth of each growth stage and the current truth, the promotion index (namely the promotion label) is positively correlated with the target truth.

In some embodiments, the third prediction model may be further obtained by training a plurality of fourth training samples carrying labels, where the fourth training samples may include at least growth morphology images of the sample agricultural products at various growth stages and current morphology images of the sample agricultural products, and the labels may represent target trueness of the sample agricultural products. Specifically, the training mode of the third prediction model is the same as that described above, and is not described herein again. In some embodiments of the present specification, the third prediction model may be obtained by training by using a difference between the target truth and the promotion index output by the third prediction model as a loss function, so that the promotion index (i.e., the promotion label) is positively correlated with the target truth.

In some embodiments, in determining the promotional index using the third predictive model, the inputs to the model may include target truth in addition to the user's historical behavior and user information. In some embodiments of the specification, the first predictive model, the second predictive model, and the third predictive model may be obtained by end-to-end training. For specific details of the end-to-end training, reference may be made to step 204 and its related description, which are not described herein again.

In some embodiments of the present description, the promotion label is determined by the feature information of the user, so that the promotion label of the image can better meet the preference of the user, for example, if the user occupation is an agricultural researcher, the promotion index of the image with the growth form is higher, thereby improving the display priority and the display strength of the image more meeting the user requirement, and improving the sales volume of the candidate agricultural products and the shopping experience of the user.

Step 404, generating the second preset advertisement based on the first preset advertisement and the promotion label. In some embodiments, step 404 may be performed by the generation module 580.

For specific details of the first preset advertisement, reference may be made to step 106 and the related description thereof, which are not described herein again. In some embodiments, the first preset advertisement may include at least a growth morphology image of the candidate agricultural product at each growth stage and a current morphology image of the candidate agricultural product. In some embodiments, the content of the first preset advertisement may be adjusted based on the promotion index, and the second preset advertisement may be generated. In some embodiments, the content of the first preset advertisement may be adjusted in a plurality of ways based on the promotion index. For example, in some embodiments, images in the first preset advertisement having a promotion index less than a preset threshold may be deleted or hidden. In some embodiments, the images in the first preset advertisement with the promotion index smaller than the preset threshold may be played last. In some embodiments, the preset threshold may be specifically set according to actual requirements, for example, 0.5, 0.6, and the like.

In some embodiments of the present description, by deleting an image of the first preset advertisement whose promotion index is smaller than a preset threshold, the user may be directly prevented from watching the image, and thus, dissatisfaction of the user on the candidate agricultural product is avoided. And finally, displaying the image of which the promotion index is smaller than the preset threshold value in the first preset advertisement, so that the user can be prevented from paying attention to the image to a certain extent, and the user is indirectly prevented from being unsatisfied with the candidate agricultural products.

In some embodiments, a target image for which the promotion index meets a preset requirement may be determined; and adjusting the second preset advertisement based on the target image, and displaying the adjusted second preset advertisement. In some embodiments, the second preset advertisement is generated from the first preset advertisement. The second predetermined advertisement may include at least one or more of the following images: the growth form image of the candidate agricultural product at each growth stage and the current form image of the candidate agricultural product. In some embodiments, the target image may be an image of the second preset advertisement, where the promotion index meets a preset requirement, and it may be understood that the target image may belong to a growth form image of the candidate agricultural product at each growth stage and/or a current form image of the candidate agricultural product. In some embodiments, the preset requirement may be specifically set according to actual requirements. For example, the preset requirement may be that the promotion index is greater than a preset threshold. For the preset threshold, reference may be made to the foregoing description, and details are not repeated here.

In some embodiments, the second pre-set advertisement may be adjusted a plurality of times based on the target image. For example, in some embodiments, new material may be added using a preset addition rule based on a preset order of target images. In some embodiments, the new material may be other images that are related or unrelated to the target image.

In some embodiments, the preset order of the target images may be determined according to a growth process of the candidate agricultural products. In some embodiments, the preset order of the target images may be from the growth morphology image to the current morphology image of each growth phase, i.e. the preset order of the target images may be: a germination stage image, a seedling stage image, a flowering stage image, a fruiting stage image and a current form image. The preset increasing rule may be manually set in advance in some embodiments. In some embodiments, the preset increase rule may be to increase a contrast map associated with the target image, for example, if the target image is a germination stage image, the preset increase rule may be to increase a growth contrast map of the germination stage, for example, determining a corresponding growth contrast map based on at least two germination stage images acquired every five days. The reliability of the authenticity of the target image can be increased by adding new materials, and the recognition degree of the user on the authenticity of the image is improved.

In some embodiments, other images of the second preset advertisement, which have a similarity to the target image that meets a preset threshold, may be determined based on the target image, and the other images may be processed to implement the adjustment of the second preset advertisement. In some embodiments, the processing of the other images may include one or more combinations of the following: adding music, magnifying images, and extending presentation time. In some embodiments, the added new material may also be processed by one or more of the following: adding music, magnifying images, and extending presentation time.

In some embodiments, the target image may be pre-processed to effect adjustment of the second pre-set advertisement. In some embodiments, the preset process may include, but is not limited to: adding music, enlarging images, extending presentation time, etc. For example, if the target image is a result period image, when the second preset advertisement is played to the result period image, one or more of the following preset processes may be performed: adding music, enlarging the resulting period image, and extending the presentation time of the resulting period image (e.g., to 15 s). Through the preset processing of the target image, the attention of a user can be attracted, so that the user can know the image with high reality degree more, and the sales volume of the candidate agricultural products is further improved.

As shown in fig. 5, the system 500 may include a first determining module 510, a determining module 520, a first presenting module 530, a second presenting module 540, a third presenting module 550, a sending module 560, a second determining module 570, and a generating module 580.

The first determination module 510 may be configured to determine candidate agricultural products based on keywords entered by a user.

The determining module 520 may be configured to determine whether the candidate agricultural product belongs to a preset agricultural product.

The first presentation module 530 may be configured to present a first preset advertisement matching the candidate agricultural product in response to not.

The second presentation module 540 may be configured to, in response to this, process the relevant information of the candidate agricultural product based on the machine learning model, evaluate the truth of the candidate agricultural product, and present a second preset advertisement matching the evaluation result. In some embodiments, the second presentation module 540 may be further configured to process, by using the machine learning model, at least the environment data of the candidate agricultural product at each growth stage, the growth form image of the candidate agricultural product at each growth stage, and the current form image of the candidate agricultural product, so as to obtain a target truth of the candidate agricultural product; determining the target truth as the evaluation result of the candidate agricultural product.

The third display module 550 may be configured to generate and display the second preset advertisement based on that the target degree of reality satisfies the first preset condition.

The sending module 560 may be configured to send warning information to the target object based on that the target truth meets a second preset condition.

The second determining module 570 may be configured to determine respective promotion labels for the growth configuration image of the candidate agricultural product at the respective growth stages and the current configuration image of the candidate agricultural product.

The generating module 580 may be configured to generate the second preset advertisement based on the first preset advertisement and the promotion label.

It should be appreciated that the system and its modules illustrated in FIG. 5 may be implemented in a variety of ways. For example, in some embodiments, the system and its modules may be implemented in hardware, software, or a combination of software and hardware. Wherein the hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory for execution by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the methods and systems described above may be implemented using computer executable instructions and/or embodied in processor control code, for example such code provided on a carrier medium such as a diskette, CD-or DVD-ROM, programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules in this specification may be implemented not only by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but also by software executed by various types of processors, for example, or by a combination of the above hardware circuits and software (e.g., firmware).

It should be noted that the above description of the system and its modules for automatically determining marketing content of agricultural products is for convenience of description only and should not limit the present disclosure within the scope of the illustrated embodiments. It will be appreciated by those skilled in the art that, given the teachings of the present system, any combination of modules or sub-system configurations may be used to connect to other modules without departing from such teachings. For example, in some embodiments, for example, the first determining module 510, the determining module 520, the first presenting module 530, the second presenting module 540, the third presenting module 550, the sending module 560, the second determining module 570 and the generating module 580 disclosed in fig. 5 may be different modules in one system, or may be one module to implement the functions of two or more modules described above. For example, the judging module 520 and the first displaying module 530 may be two modules, or one module may have both the judging function and the displaying function. For example, each module may share one memory module, and each module may have its own memory module. Such variations are within the scope of the present disclosure.

Having thus described the basic concept, it will be apparent to those skilled in the art that the foregoing detailed disclosure is to be considered as illustrative only and not limiting, of the present invention. Various modifications, improvements and adaptations to the present description may occur to those skilled in the art, although not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present specification and thus fall within the spirit and scope of the exemplary embodiments of the present specification.

Also, the description uses specific words to describe embodiments of the description. Reference throughout this specification to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the specification is included. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, certain features, structures, or characteristics may be combined as suitable in one or more embodiments of the specification.

Moreover, those skilled in the art will appreciate that aspects of the present description may be illustrated and described in terms of several patentable species or situations, including any new and useful combination of processes, machines, manufacture, or materials, or any new and useful improvement thereof. Accordingly, aspects of this description may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the present description may be represented as a computer product, including computer readable program code, embodied in one or more computer readable media.

The computer storage medium may comprise a propagated data signal with the computer program code embodied therewith, for example, on a baseband or as part of a carrier wave. The propagated signal may take any of a variety of forms, including electromagnetic, optical, etc., or any suitable combination. A computer storage medium may be any computer-readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code located on a computer storage medium may be propagated over any suitable medium, including radio, cable, fiber optic cable, RF, or the like, or any combination of the preceding.

Computer program code required for the operation of various portions of this specification may be written in any one or more programming languages, including an object oriented programming language such as Java, scala, smalltalk, eiffel, JADE, emerald, C + +, C #, VB.NET, python, and the like, a conventional programming language such as C, visual Basic, fortran 2003, perl, COBOL 2002, PHP, ABAP, a dynamic programming language such as Python, ruby, and Groovy, or other programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any form of network, such as a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet), or in a cloud computing environment, or as a service using, for example, software as a service (SaaS).

Additionally, the order in which the elements and sequences of the process are recited in the specification, the use of alphanumeric characters, or other designations, is not intended to limit the order in which the processes and methods of the specification occur, unless otherwise specified in the claims. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the present specification, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the embodiments. This method of disclosure, however, is not intended to imply that more features are required than are expressly recited in the claims. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.

Numerals describing the number of components, attributes, etc. are used in some embodiments, it being understood that such numerals used in the description of the embodiments are modified in some instances by the use of the modifier "about", "approximately" or "substantially". Unless otherwise indicated, "about", "approximately" or "substantially" indicates that the number allows a variation of ± 20%. Accordingly, in some embodiments, the numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by a particular embodiment. In some embodiments, the numerical parameter should take into account the specified significant digits and employ a general digit preserving approach. Notwithstanding that the numerical ranges and parameters setting forth the broad scope of the range in some embodiments of the specification are approximations, in specific embodiments, such numerical values are set forth as precisely as possible within the practical range.

For each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, etc., cited in this specification, the entire contents of each are hereby incorporated by reference into this specification. Except where the application history document does not conform to or conflict with the contents of the present specification, it is to be understood that the application history document, as used herein in the present specification or appended claims, is intended to define the broadest scope of the present specification (whether presently or later in the specification) rather than the broadest scope of the present specification. It is to be understood that the descriptions, definitions and/or uses of terms in the accompanying materials of the present specification shall control if they are inconsistent or inconsistent with the statements and/or uses of the present specification.

Finally, it should be understood that the embodiments described herein are merely illustrative of the principles of the embodiments described herein. Other variations are also possible within the scope of the present description. Thus, by way of example, and not limitation, alternative configurations of the embodiments of the present specification can be seen as consistent with the teachings of the present specification. Accordingly, the embodiments of the present description are not limited to only those embodiments explicitly described and depicted herein.

Claims

1. A method of generating an agricultural product promotional advertisement, the method comprising:

determining candidate agricultural products based on keywords input by a user;

judging whether the candidate agricultural products belong to preset agricultural products, wherein the preset agricultural products are agricultural products with quality artifacts easy to artificially manufacture;

responding to the condition that the agricultural product is not in the growth state, displaying a first preset advertisement matched with the candidate agricultural product, wherein the candidate agricultural product is in one-to-one correspondence with the first preset advertisement, and the first preset advertisement at least comprises a growth form image of the candidate agricultural product in each growth stage and a current form image of the candidate agricultural product;

responding to the information, and processing at least the environment data of the candidate agricultural products in each growth stage, the growth form images of the candidate agricultural products in each growth stage and the current form images of the candidate agricultural products by using a machine learning model to obtain the target truth of the candidate agricultural products;

responding to that the target truth meets a first preset condition, and determining respective promotion labels of the growth form image and the current form image of the candidate agricultural product in each growth stage, wherein the promotion labels at least comprise promotion indexes which are positively correlated with the target truth;

and adjusting the first preset advertisement of the candidate agricultural product based on the promotion label, and generating and displaying a second preset advertisement of the candidate agricultural product.

2. The method of claim 1, further comprising:

and sending warning information to the target object based on the fact that the target truth meets a second preset condition.

3. A system for generating an agricultural product promotional advertisement, the system comprising:

the first determination module is used for determining candidate agricultural products based on keywords input by a user;

the judging module is used for judging whether the candidate agricultural products belong to preset agricultural products, and the preset agricultural products are agricultural products with quality artifacts easy to artificially manufacture;

the first display module is used for displaying a first preset advertisement matched with the candidate agricultural product in response to the answer that the answer is not yes, the candidate agricultural product is in one-to-one correspondence with the first preset advertisement, and the first preset advertisement at least comprises a growth form image of the candidate agricultural product in each growth stage and a current form image of the candidate agricultural product;

a second display module, configured to, in response to the request, utilize a machine learning model to process at least the environmental data of the candidate agricultural product at each growth stage, the growth form image of the candidate agricultural product at each growth stage, and the current form image of the candidate agricultural product, so as to obtain a target truth of the candidate agricultural product;

a second determining module, configured to determine respective promotion labels of the growth form image and the current form image of the candidate agricultural product at each growth stage in response to that the target truth meets a first preset condition, where the promotion labels at least include a promotion index, and the promotion index is positively correlated with the target truth;

the generating module is used for adjusting the first preset advertisement of the candidate agricultural product based on the promotion label and generating a second preset advertisement of the candidate agricultural product;

and the third display module is used for displaying the second preset advertisement.

4. The system of claim 3, further comprising:

and the sending module is used for sending warning information to the target object based on the fact that the target truth meets a second preset condition.

5. An agricultural product promotion advertisement generation apparatus, the apparatus comprising a processor and a memory, the memory for storing instructions, the processor for executing the instructions to implement operations corresponding to the method of any one of claims 1 to 2.

6. A computer-readable storage medium storing computer instructions which, when executed by a processor, perform operations corresponding to the method of any one of claims 1 to 2.