Agricultural big data analysis method and system based on intelligent understanding

Abstract

The invention provides an agricultural big data analysis method and system based on intelligent understanding, and belongs to the technical field of medical equipment, and the method specifically comprises the steps: carrying out the generation processing of a marketing video of an optimization target based on a generation scheme, and obtaining a generation result, on the basis of the generation scheme of the marketing videos of the optimization targets with the incidence relation, intelligent understanding strategies of different marketing videos are determined, and on the basis of understanding results obtained through the intelligent understanding strategies, updating results of the optimization targets of the marketing schemes are determined. And the matching degree between the generation processing result of the marketing video and the user demand is improved.

Description

Technical Field

[0001] This invention belongs to the field of data analysis technology, and in particular relates to an agricultural big data analysis method and system based on intelligent understanding. Background Technology

[0002] Agricultural products often suffer from a mismatch between supply and demand. Therefore, how to combine short video platforms to generate personalized marketing videos and promote the sales of agricultural products has become an urgent technical problem to be solved. The existing technical solution CN202410973299.2, "A Short Video Intelligent Understanding Method Based on Contextual Information Enhancement", determines the reflection processing mode by using the evaluation results of short videos and video context information, and obtains modular short video intelligent understanding results based on the reflection processing results, thereby enabling targeted short video generation and processing.

[0003] Before generating marketing videos for agricultural products awaiting market launch, existing technical solutions neglect to determine the generation and processing scheme of marketing videos based on the correlation between the agricultural products for which marketing videos are to be generated and the agricultural products awaiting market launch. As a result, it is impossible to clarify the real needs of users and thus impossible to accurately determine the matching degree of the generation and processing of marketing videos for agricultural products awaiting market launch.

[0004] To address the aforementioned technical problems, this application provides an agricultural big data analysis method and system based on intelligent understanding. Summary of the Invention

[0005] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides an agricultural big data analysis method based on intelligent understanding, which includes: S1 uses agricultural product data as a basis to determine the association between the marketing labels of the agricultural products and the agricultural products to be marketed, determines the optimization target of the marketing plan for the agricultural products based on the association, and determines the generation plan of the marketing video for the optimization target based on the association between the optimization target and the marketing labels of the agricultural products to be marketed and the listing date of the agricultural products to be marketed. S2 generates marketing videos for the optimization target based on the generation scheme to obtain the generation result. Based on the generation scheme of marketing videos for the optimization target that are related to the marketing label of the agricultural product to be marketed, different intelligent understanding strategies for marketing videos are determined. Based on the understanding results obtained by the intelligent understanding strategies, the update result of the optimization target of the marketing scheme is determined.

[0006] The beneficial effects of this invention are as follows: Based on the correlation between the optimization target and the marketing label of the agricultural product to be marketed, as well as the listing date of the agricultural product to be marketed, the generation scheme of the marketing video for the optimization target is determined. This realizes the determination of the generation scheme of the marketing video for the optimization target from the perspective of the generation needs of the marketing video for the agricultural product to be marketed. By determining the differentiated marketing video generation scheme according to the marketing label, the difficulty of generating and processing the marketing video is reduced, while ensuring the efficiency and reliability of the identification and processing of user needs for the marketing label of the agricultural product to be marketed. This lays the foundation for the targeted generation of marketing videos for the agricultural product to be marketed.

[0007] Based on the understanding results obtained from the intelligent understanding strategy, the optimization goals of the marketing plan are updated to avoid the technical problem of poor matching degree of marketing video generation and processing due to the unreliability of the user demand identification and processing of marketing labels for some agricultural products to be launched, without the optimization goal update processing. This lays the foundation for further improving the matching degree of marketing video generation and processing for agricultural products to be launched.

[0008] Furthermore, the agricultural products mentioned are those for which marketing videos will be generated.

[0009] Furthermore, the agricultural products to be marketed are those that need to be marketed within a preset time period in the future. In one possible embodiment, agricultural products that need to be marketed within the next one to three months are considered as agricultural products to be marketed. By optimizing the marketing video, the agricultural products to be marketed and the issues that customers care about can be obtained, so that the marketing video can be generated in a targeted manner.

[0010] Furthermore, the association between the agricultural product and the marketing label of the agricultural product to be marketed is determined based on the same number of marketing labels on the agricultural product and the agricultural product to be marketed.

[0011] Furthermore, the method for determining the optimization objectives of the marketing plan for the agricultural products is as follows: Based on the association between the marketing labels of the agricultural products and the marketing labels of the agricultural products to be marketed, determine the marketing labels that are associated with the marketing labels of the agricultural products and the agricultural products to be marketed, and use them as associated marketing labels. Based on the associated marketing label data of different agricultural products to be marketed, the associated agricultural products are determined; Based on the associated agricultural product data, determine whether the agricultural product is the target of the marketing plan optimization.

[0012] Furthermore, the method for determining the updated results of the optimization objectives of the marketing plan is as follows: Based on the understanding results obtained from the aforementioned intelligent understanding strategy, the browsing user data of the marketing video corresponding to the marketing label of the agricultural product to be marketed is determined; Based on the browsing user data, the number of users browsing the marketing videos associated with the marketing tags of the agricultural products to be launched is determined. A marketing deviation factor is determined based on the proportion of agricultural products whose total number of browsing users does not meet the requirement among all agricultural products to be launched. Based on the marketing matching deviation factor of the agricultural products whose total number of browsing users does not meet the requirement and the marketing deviation factor, the updated result of the optimization objective of the marketing plan is determined.

[0013] Secondly, this application provides a computer system, including: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described method for agricultural big data analysis based on intelligent understanding when running the computer program.

[0014] Other features and advantages will be set forth in the following description, and the objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0015] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart of an agricultural big data analysis method based on intelligent understanding; Figure 2 This is a flowchart illustrating the method for determining the optimization objectives of marketing plans for agricultural products. Figure 3 This is a flowchart illustrating the method for determining a marketing video generation strategy that optimizes objectives; Figure 4 This is a flowchart illustrating the method for determining the intelligent understanding strategy for marketing videos; Figure 5 It is a framework diagram of a computer system. Detailed Implementation

[0018] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0019] This application improves the matching degree of marketing video generation for agricultural products that currently require marketing video generation by associating them with the marketing tags of agricultural products awaiting market launch.

[0020] Example 1 like Figure 1 As shown, this application provides an agricultural big data analysis method based on intelligent understanding, specifically including: S1 Based on agricultural product data, determine the correlation between the agricultural product and the marketing labels of agricultural products to be marketed, determine the optimization target of the marketing plan in the agricultural product based on the correlation, and determine the generation plan of the marketing video of the optimization target based on the correlation between the optimization target and the marketing labels of agricultural products to be marketed and the listing date of agricultural products to be marketed; Furthermore, the agricultural products mentioned are those for which marketing videos will be generated.

[0021] Furthermore, the agricultural products to be marketed are those that need to be marketed within a preset time period in the future. In one possible embodiment, agricultural products that need to be marketed within the next one to three months are considered as agricultural products to be marketed. By optimizing the marketing video, the agricultural products to be marketed and the issues that customers care about can be obtained, so that the marketing video can be generated in a targeted manner.

[0022] Furthermore, the association between the agricultural product and the marketing label of the agricultural product to be marketed is determined based on the same number of marketing labels on the agricultural product and the agricultural product to be marketed.

[0023] It is understandable that the marketing labels of the agricultural products include pollution-free, green, organic, pesticide residue-free, traceable, freshly picked, place of origin, and suitable user group.

[0024] Specifically, such as Figure 2 As shown, the method for determining the optimization objective of the marketing plan for the agricultural product is as follows: Based on the association between the marketing labels of the agricultural products and the marketing labels of the agricultural products to be marketed, determine the marketing labels that are associated with the marketing labels of the agricultural products and the agricultural products to be marketed, and use them as associated marketing labels. Based on the associated marketing label data of different agricultural products to be marketed, the associated agricultural products are determined; Based on the associated agricultural product data, determine whether the agricultural product is the target of the marketing plan optimization.

[0025] It is understood that the related agricultural products are agricultural products awaiting market launch that have multiple related marketing labels.

[0026] Specifically, based on the associated agricultural product data, determining whether the agricultural product is the optimization target of the marketing plan includes: When the associated agricultural product data of the agricultural product meets the requirements, the agricultural product is determined to be the optimization target of the marketing plan.

[0027] Specifically, when the number of related agricultural products of the agricultural product is greater than a preset threshold for the number of related agricultural products, the agricultural product is determined to be the optimization target of the marketing plan.

[0028] Specifically, the core objective of this method is to identify the "key products" most worthy of in-depth optimization from the agricultural products currently requiring marketing videos. The selection criterion is whether the agricultural product has a high marketing relevance to several important agricultural products that will be launched in the future. If so, optimizing its marketing video can generate long-term value with multiple benefits.

[0029] Operation: Compare the "agricultural product" for which a video needs to be made with each "agricultural product awaiting market launch". The tags that these products share are defined as "related marketing tags".

[0030] Agricultural product A (current): labeled as [green, freshly picked, traceable]; Agricultural product B (future): labeled as [green, organic, traceable, place of origin], the associated marketing label for A and B is: [green, traceable]; Agricultural products that have multiple (≥2) identical marketing labels to the current agricultural product are defined as "related agricultural products". If agricultural product A and product B have two identical related marketing labels (green, traceable), then product B is a "related agricultural product" of the current agricultural product A. If the number of "related agricultural products" of the current agricultural product is not less than 2, then it is selected as the target that needs to be optimized, that is, the agricultural product is the optimization target of the marketing plan.

[0031] Specifically, such as Figure 3As shown, the method for determining the generation scheme of the marketing video for the optimization target is as follows: Based on the correlation between the optimization target and the marketing labels of the agricultural products to be marketed, identify the agricultural products to be marketed that have overlapping marketing labels with the optimization target, and treat them as products with overlapping labels. Based on the listing date of the products with overlapping labels, determine the interval between the listing date of the products with overlapping labels and the current date; Based on the quantity and interval of agricultural products to be launched associated with the marketing tags, a demand generation factor is determined, and a marketing video generation scheme for the optimization target is determined based on the demand generation factor.

[0032] Optionally, when the generation demand factor of the marketing tag is greater than a preset demand factor threshold, the generation scheme of the marketing video for the optimization target is determined, and a target number of marketing videos are generated for each marketing tag.

[0033] It should be noted that the target number is greater than the second preset number.

[0034] Furthermore, when the generation demand factor of the marketing tag is not greater than a preset demand factor threshold, the proportion of the marketing tag for generating a target number of marketing videos among the marketing tags of the agricultural products is obtained and used as the generation matching factor. When the generation matching factor is greater than the preset matching factor threshold, if the generation demand factor of the marketing tag is within the preset demand factor range, or if no marketing videos are generated for the marketing tags of the agricultural products to be marketed associated with the marketing tag, then it is determined that a second preset number of marketing videos need to be generated. If the generation demand factor of the marketing tag is not within the preset demand factor range, then there is no need to generate marketing videos for the marketing tag.

[0035] Additionally, it is understood that when the generated matching factor is not greater than the preset matching factor threshold, if the marketing tag has associated agricultural products awaiting market launch, then a second preset number of marketing videos need to be generated; if the marketing tag does not have associated agricultural products awaiting market launch, then there is no need to generate marketing videos for the marketing tag.

[0036] Let's take a specific agricultural product case to fully demonstrate the decision-making process of the entire marketing video generation solution.

[0037] Suppose the agricultural product for which we need to create a video is "High Mountain" apples, which have the marketing tags [organic, traceable, freshly picked, origin]. In our system, the threshold for generating demand factors is set to 1, and the threshold for generating matching factors is set to 0.6. We also have a series of agricultural products that will be available in the future: "Sunshine" oranges (tagged [organic, traceable, freshly picked]) available in one month, "Field" strawberries (tagged [organic, pesticide-free]) available in two months, and "Clear Spring" blueberries (tagged [traceable, origin]) available in three months.

[0038] First, the system calculates a "demand generation factor" for each marketing tag for "High Mountain" apples. This factor comprehensively considers the breadth (number of associated products) and urgency (market launch time) of the tag's association with future products. For example, the "Organic" tag is associated with both "Sunshine" oranges and "Garden" strawberries, and their market launch time is approaching, so its demand generation factor is calculated to be 1.5, indicating high demand. The "Traceable" tag is also associated with "Sunshine" oranges and "Clear Spring" blueberries, but with a slightly longer time span, resulting in a factor of 0.7. The "Origin" tag is only associated with "Clear Spring" blueberries, with a factor of 0.4. While the "Freshly Picked" tag is associated with "Sunshine" oranges, its extremely short shelf life makes it meaningful only for the current product and has low reuse value for future products, hence its factor is only 0.1.

[0039] Based on these factors, the system begins to implement tiered decision-making. For the "Organic" tag, with a demand factor as high as 1.5, it directly exceeds the preset threshold of 1, and the system determines it as a high-value core tag. Therefore, the system decides to generate two marketing videos for it to ensure content richness and sufficient future material reuse capabilities. Next, the system processes the remaining tags whose demand factors do not reach the threshold. It first calculates the "generation matching factor," i.e., the proportion of high-demand tags. Currently, only the "Organic" tag is considered high-demand, with a proportion of 1 / 4 = 0.25, below the threshold of 0.6. This means that resources are not yet highly concentrated, and the system then adopts a more "inclusive" strategy to handle these low-to-medium demand tags.

[0040] Under this strategy, the system checks whether each tag is associated with future products. For the "Traceable" tag (factor 0.7), it is associated with "Sunshine" oranges and "Clear Spring" blueberries, so the system decides to generate a basic video for it. The "Origin" tag (factor 0.4) is also associated with "Clear Spring" blueberries, and similarly receives the decision to generate a video for it. However, the "Freshly Picked" tag (factor 0.1) is not associated with any future products, and its own demand factor is extremely low, so the system ultimately decides not to generate a dedicated video for it to avoid wasting resources.

[0041] Ultimately, the video production plan for "High Mountain" apples was as follows: two videos were created for the "organic" selling point, one video each for the "traceable" and "origin" selling points, and no separate video was created for the "freshly picked" selling point. This plan precisely allocated resources to the "organic" attribute, which has the greatest long-term reuse value, while also taking into account other labels with future synergistic effects, ensuring the foresight and efficiency of the marketing investment.

[0042] 1. Definition and Purpose: The generation demand factor is a quantitative metric for a single marketing hashtag, used to measure the necessity and urgency of creating a video for that hashtag. Its core idea is: the more future products a hashtag is associated with, and the more urgent their market launch, the greater the demand for creating a video for it now. Generation demand factor = Σ (1 / interval duration weight). Σ (summation symbol): Represents the summation of the contribution values ​​of all "products with overlapping tags" associated with this tag. Interval weight: This is a coefficient set based on the launch interval, reflecting the logic that "the closer the time, the higher the weight (i.e., the smaller the denominator, the larger the value)." For example, it can be set as follows: Launch within the next month: Interval weight = 1 (very urgent); Launch within the next 1-3 months: Interval weight = 2 (relatively urgent); Launch more than 3 months in the future: Interval weight = 3 (not urgent, can be postponed); 3. Calculation Example: Take the "organic" label of "high-altitude" apples as an example: Related product 1: "Sunshine" oranges (available in 1 month), interval weight = 1, contribution value = 1 / 1 = 1; Related product 2: "Garden" strawberries (available in 2 months, falling within 1-3 months), interval weight = 2, contribution value = 1 / 2 = 0.5. Therefore, the demand factor for generating the "organic" label = 1 + 0.5 = 1.5.

[0043] 1. Definition and Purpose: The matching factor is a macro-level indicator for the entire target agricultural product being optimized. It measures the extent to which the current video generation scheme covers high-demand tags. This factor determines whether the system adopts a "focused supplementation" strategy or a "broad coverage" strategy in subsequent decisions.

[0044] 2. Calculation formula: The generation matching factor = the number of tags that are determined to "require the generation of the target number of videos" / the total number of marketing tags for the agricultural product. This refers to the number of tags that "require the generation demand factor > the preset demand factor threshold" (>1 in the previous example), thus requiring the generation of 2 videos.

[0045] Total number of marketing tags: 4 (

Organic

Traceable

Origin

Freshly Picked

Organic

Traceable

[0046] This factor value of 0.5 (less than the threshold of 0.6) will guide the system into the "inclusive" decision branch, which means generating at least one basic video for tags that have future relevance but low demand factors (such as "origin").

[0047] Summarize Through these two formulas, the system transforms a complex marketing decision-making problem into a quantifiable and automated calculation process. The generated demand factor accurately captures the vertical value depth of each tag, while the generated matching factor intelligently assesses the horizontal coverage breadth of the current resource plan. Together, they form an intelligent marketing video generation solution decision-making system that focuses on key breakthroughs while also taking into account overall balance.

[0048] S2 generates marketing videos for the optimization target based on the generation scheme to obtain the generation result. Based on the generation scheme of marketing videos for the optimization target that are related to the marketing label of the agricultural product to be marketed, different intelligent understanding strategies for marketing videos are determined. Based on the understanding results obtained by the intelligent understanding strategies, the update result of the optimization target of the marketing scheme is determined.

[0049] Specifically, such as Figure 4 As shown, the method for determining the intelligent understanding strategy of the marketing video is as follows: Based on the marketing video generation scheme that optimizes the marketing target and is related to the marketing label of the agricultural product to be marketed, the associated marketing video of the agricultural product to be marketed is determined. Based on the marketing tags associated with the marketing videos of the agricultural products to be marketed, it is determined that the agricultural products to be marketed do not have marketing tags associated with the marketing videos. Based on the agricultural products to be launched associated with the marketing video, and the absence of marketing tags associated with the marketing video, an intelligent understanding strategy for the marketing video is determined.

[0050] It should be noted that, based on the agricultural products to be marketed associated with the marketing video, there are no marketing tags associated with the marketing video. The intelligent understanding strategy for determining the marketing video specifically includes: Based on the proportion of agricultural products awaiting market launch that are associated with the marketing video but do not have a marketing tag associated with the marketing video, the marketing matching deviation factor of the associated agricultural products awaiting market launch is determined. The intelligent understanding strategy for the marketing video is determined based on the sum of the marketing matching deviation factors of the agricultural products to be marketed associated with the marketing video.

[0051] It is understandable that when the sum of the marketing matching deviation factors of the agricultural products to be marketed associated with the marketing video is greater than the preset value of the deviation factor, the intelligent understanding strategy of the marketing video is determined to be to characterize the user profiles of all users who view the marketing video, thereby determining the content of interest of users who view the marketing tags corresponding to the marketing video, and then using the content of interest to generate marketing videos for agricultural products to be marketed.

[0052] Specifically, user profiles are created for all users who view the marketing videos, including: By analyzing the browsing data and comment data of the users who viewed the marketing videos, the content that the users of the marketing videos are interested in can be determined, such as the shipping method, timeliness, and test reports.

[0053] Furthermore, when the sum of the marketing matching deviation factors of the agricultural products to be marketed associated with the marketing video is not greater than the preset value of the deviation factor, it is only necessary to use the comment data of the marketing video to determine the user's attention content of the marketing tag, such as transportation method, timeliness and test report.

[0054] In a possible specific implementation, let's illustrate how this strategy works with a concrete example. Suppose we've already developed a video generation scheme for "highland" honey: generating two videos for its "natural and wild" selling point, and one video each for its "traceable" and "additive-free" selling points. Now, the system needs to evaluate how well these videos support two future products—"deep mountain" royal jelly (labeled as [natural and wild, traceable, active ingredients]) and "pasture" cheese (labeled as [traceable, additive-free, fermentation process]).

[0055] The system first performs a coverage audit to identify marketing gaps. By scanning the video library for "highland" honey, the system found that there were no readily available videos for the "active substances" tag on "deep mountain" royal jelly and the "fermentation process" tag on "pasture" cheese. These two missing tags represent the marketing gaps in the current video asset library.

[0056] Next, the system will initiate a quantitative assessment, calculating the marketing match deviation factor. This factor measures the size of the gap, and the calculation formula is: Deviation factor for a single product = Number of tags in the product's missing video / Total number of tags for the product. In this case, the deviation factor for "Deep Mountain" royal jelly is 1 / 3 ≈ 0.33, and the deviation factor for "Pasture" cheese is also 1 / 3 ≈ 0.33. Since the video for "Highland" honey is associated with both of these future products, the system calculates their total deviation factor sum to be 0.66.

[0057] Based on this sum of 0.66 (not exceeding the preset threshold of 0.7), the system will execute a more cost-effective basic analysis strategy. The core of this strategy is to deeply mine existing video comment data. The system will use natural language processing technology to analyze user comments under videos about "highland" honey, extracting high-frequency keywords and potential concerns. For example, the system might find that under "natural wild" videos, users frequently ask, "How can I verify its wild nature? Is there an authoritative testing report?"; while under "additive-free" videos, users are particularly concerned about "What is the specific shelf life? What transportation method ensures freshness?". These concerns that naturally emerge from user feedback, such as "testing reports" and "transportation timeliness," provide precise direction for video creation that fills the gaps in labels like "active substances" and "fermentation process." The marketing team can then create content to directly address these concerns.

[0058] If the "deep mountain" royal jelly also lacks video coverage due to its "rare origin" label, its bias factor will become 2 / 3 ≈ 0.67, and the total bias factor will jump to 1.34 (0.67 + 0.67). At this point, since the sum exceeds the threshold of 0.7, the system will activate a more in-depth advanced user profiling strategy. This means that the system is no longer limited to analyzing comments, but will conduct a comprehensive profiling of all users who have viewed "highland" honey videos. It will collect anonymous behavioral data, such as viewing completion rate, likes, shares, and even interest in other videos, thereby constructing a three-dimensional user profile and gaining insights into the deep needs that users have not explicitly stated but are highly concerned about, such as a preference for "limited edition" models or "celebrity endorsements," thus providing richer decision-making basis for the creativity of subsequent marketing videos.

[0059] Ultimately, the intelligent understanding strategy forms a self-optimizing loop system. It identifies content gaps through quantitative assessment, and then, based on the severity of the gaps, initiates tiered data insight methods ranging from "comment analysis" to "panoramic user profiling," ensuring that every video playback is transformed into valuable intelligence to guide future marketing, making the use of marketing resources increasingly precise and efficient.

[0060] Specifically, the method for determining the updated results of the optimization objectives of the marketing plan is as follows: Based on the understanding results obtained from the aforementioned intelligent understanding strategy, the browsing user data of the marketing video corresponding to the marketing label of the agricultural product to be marketed is determined; Based on the browsing user data, the number of users who viewed the marketing videos associated with the marketing tags of the agricultural products to be launched is determined. The marketing deviation factor is determined based on the proportion of agricultural products to be launched whose total number of browsing users does not meet the requirements among all agricultural products to be launched. Based on the marketing matching deviation factor of agricultural products awaiting market launch whose total number of browsing users does not meet the requirements, and the marketing deviation factor, the updated result of the optimization objective of the marketing plan is determined.

[0061] Specifically, if the marketing deviation factor is greater than a preset deviation factor threshold, for example, greater than 0.5, then all agricultural products that have the same marketing label as the agricultural products to be marketed are determined to be optimization targets.

[0062] Furthermore, if the marketing deviation factor is not greater than a preset deviation factor threshold, the marketing matching deviation factor of agricultural products awaiting market launch is determined based on the total number of browsing users not meeting the requirements. When the marketing matching deviation factor of agricultural products awaiting market launch with the total number of browsing users not meeting the requirements is greater than a preset deviation factor value, for example, greater than 0.3, agricultural products with the same marketing labels as agricultural products awaiting market launch with the total number of browsing users not meeting the requirements are all considered optimization targets.

[0063] It should also be noted that when the marketing matching deviation factor of agricultural products awaiting market launch that do not meet the requirements in terms of the total number of browsing users is not greater than the preset value of the deviation factor, it is determined that there is no need to determine the optimization target based on agricultural products awaiting market launch that do not meet the requirements in terms of the total number of browsing users.

[0064] In the final stage of dynamic optimization of the marketing plan, the system intelligently updates the "optimization target" list based on the actual performance of the published marketing videos. The core of this process lies in using real user browsing data to evaluate the effectiveness of current video content in driving future product traffic, and accordingly decide whether to expand the scope of marketing resource investment. The following specific example illustrates the implementation process of this method.

[0065] The purpose of this step is to dynamically adjust and update the list of "optimization targets" based on the actual performance (viewer data) of the previously generated marketing videos, thereby realizing an intelligent marketing system that can learn and correct itself.

[0066] The following will illustrate the operation process of this method with a specific implementation method.

[0067] Example: Dynamic target update based on video traffic generation effect Background: The system has generated marketing videos for the first batch of "optimized targets" (such as "plateau" honey and "green garden" vegetables) and run an intelligent understanding strategy, collecting user browsing data over a period of time.

[0068] Existing marketing video library: "Highland" honey video: tagged [natural and wild, traceable]; "Green Garden" vegetable video: tagged [pollution-free, fresh]; Agricultural products awaiting market launch: "Deep Mountain" Royal Jelly (labels: [Natural Wild, Traceable, Active Ingredients]), "Pasture" Cheese (labels: [High Calcium, Fresh]), "Valley" Walnuts (labels: [Brain Health, Origin]); Preset threshold: Marketing deviation factor threshold: 0.5, marketing match deviation factor threshold: 0.3, minimum number of users browsing: 1000; Detailed Explanation of Methods and Steps Step 1: Collect browsing user data and identify problematic products; The system tracks user behavior and evaluates the effectiveness of existing videos in driving traffic to future products.

[0069] Aggregated browsing user count: For each product to be launched, the system calculates the total number of unique browsing users for all videos that share tags with it.

[0070] "Deep Mountain" Royal Jelly: Associated with natural wild and traceable videos. Total viewers = 1150 -> Meets requirements. "Pasture" Cheese: Associated with traceable videos. Total viewers = 850 -> Does not meet requirements. "Valley" Walnuts: No associated videos. Total viewers = 0 -> Does not meet requirements.

[0071] Problematic products identified: Based on this, the system determined that the "agricultural products awaiting market launch whose total number of browsing users does not meet the requirement" are: "Pasture" cheese and "Valley" walnuts.

[0072] Step 2: Calculate the marketing deviation factor; This factor measures the suitability of current agricultural product marketing videos. The calculation formula is: Marketing Deviation Factor = Number of agricultural products that do not meet the requirements / Total number of agricultural products to be marketed. Calculation: Number of products that do not meet the requirements = 2, Total number = 3, Marketing Deviation Factor = 2 / 3 ≈ 0.67.

[0073] Step 3: Determine the updated result of the optimization objective based on the deviation factor; Now, the system makes decisions based on the calculated factors, and the current case data (marketing deviation factor = 0.67 > 0.5) will guide the system to execute a "full expansion strategy".

[0074] Decision-making and implementation (corresponding path E): Decision: Since the marketing deviation factor (0.67) > threshold (0.5), it indicates that the failure of the existing video matrix to drive traffic is not an isolated case, but rather a systemic problem of insufficient coverage or poor performance. Therefore, the system adopts the most thorough comprehensive expansion strategy. The system identifies all current agricultural products that share any marketing labels with the two problematic products, "Pasture" cheese and "Valley" walnuts. Among those sharing the "fresh" or "high calcium" label with "Pasture" cheese, there may also be "Pure" milk; and among those sharing the "brain health" or "origin" label with "Valley" walnuts, there may also be "Deep Sea" fish oil and "Highland" barley.

[0075] Update results: The system has automatically added "pure" milk, "deep-sea" fish oil, and "highland" barley as new "optimization targets." In the next round of planning, the system will prioritize generating marketing videos for these new products, aiming to build a more robust and comprehensive video asset library and address the current issue of insufficient systemic viewership.

[0076] Example 2 Secondly, such as Figure 5 As shown, this application provides a computer system, including: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor. When the processor runs the computer program, it executes the above-described method for agricultural big data analysis based on intelligent understanding.

[0077] Optionally, the method for determining the optimization objective of the marketing plan for the agricultural products is as follows: Based on the association between the marketing labels of the agricultural products and the marketing labels of the agricultural products to be marketed, determine the marketing labels that are associated with the marketing labels of the agricultural products and the agricultural products to be marketed, and use them as associated marketing labels. Based on the marketing tags belonging to the associated marketing tags of agricultural products awaiting market launch, determine the agricultural product data associated with different marketing tags of the agricultural products awaiting market launch. Based on the data of agricultural products awaiting market launch associated with different marketing labels of the agricultural products, determine whether the agricultural products are the optimization targets of the marketing plan.

[0078] Specifically, the agricultural products to be marketed associated with the marketing label are those agricultural products to be marketed that belong to the associated marketing label in the marketing label.

[0079] It should be noted that, based on the data of agricultural products awaiting market launch associated with different marketing labels of the aforementioned agricultural products, it is determined whether the aforementioned agricultural products are the optimization targets of the marketing plan, specifically including: The association factors for different marketing labels are determined by the number of agricultural products awaiting market launch associated with different marketing labels. Based on the sum of the association factors of different marketing tags, determine whether the agricultural product is the optimization target of the marketing plan.

[0080] It should be noted that the association factor of the marketing label is determined by multiplying the number of agricultural products to be marketed associated with the marketing label by a preset ratio factor.

[0081] It is understandable that when the sum of the association factors of different marketing tags is greater than the preset association factor threshold, the agricultural product is determined to be the optimization target of the marketing plan.

[0082] Example 3 Furthermore, the method for determining the updated results of the optimization objectives of the marketing plan is as follows: Based on the understanding results obtained from the aforementioned intelligent understanding strategy, the browsing user data of the marketing video corresponding to the marketing label of the agricultural product to be marketed is determined; Based on the browsing user data, determine the number of users who viewed the marketing videos associated with the marketing tags of the agricultural products to be marketed; Based on the marketing tags of the agricultural products to be marketed and the number of users who viewed the associated marketing videos, the updated results of the optimization goals of the marketing plan are determined.

[0083] Furthermore, based on the number of users who viewed the marketing videos corresponding to the marketing tags of the agricultural products to be marketed, the updated results of the optimization objectives of the marketing plan are determined, specifically including: The total number of users who viewed the marketing videos associated with the marketing tags of the agricultural products to be marketed was determined by associating them with the marketing tags of the agricultural products to be marketed. When the total number of users browsing different marketing tags for all agricultural products awaiting market launch meets the requirements, the generation of marketing videos for all agricultural products awaiting market launch can effectively determine the content that users are interested in. Therefore, the update result of the optimization goal of the marketing plan is determined to be that no update processing is required.

[0084] Additionally, it should be noted that if there are agricultural products awaiting market launch whose total number of browsing users does not meet the requirement, the marketing deviation factor is determined based on the proportion of agricultural products awaiting market launch whose total number of browsing users does not meet the requirement among all agricultural products awaiting market launch. If the marketing deviation factor is greater than the preset deviation factor threshold, then all agricultural products that have the same marketing label as the agricultural products awaiting market launch are determined to be optimization targets.

[0085] Furthermore, if the marketing deviation factor is not greater than a preset deviation factor threshold, the marketing matching deviation factor of agricultural products awaiting market launch is determined based on the total number of browsing users not meeting the requirements. When the marketing matching deviation factor of agricultural products awaiting market launch with the total number of browsing users not meeting the requirements is greater than the preset deviation factor value, agricultural products with the same marketing labels as agricultural products awaiting market launch with the total number of browsing users not meeting the requirements are determined to be optimization targets.

[0086] It should also be noted that when the marketing matching deviation factor of agricultural products awaiting market launch that do not meet the requirements in terms of the total number of browsing users is not greater than the preset value of the deviation factor, it is determined that there is no need to determine the optimization target based on agricultural products awaiting market launch that do not meet the requirements in terms of the total number of browsing users.

[0087] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0088] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0089] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. An agricultural big data analysis method based on intelligent understanding, characterized in that, Specifically, it includes: Based on agricultural product data, the association between the agricultural products and the marketing labels of agricultural products to be marketed is determined. Based on the association, the optimization target of the marketing plan for the agricultural products is determined. Based on the association between the optimization target and the marketing labels of agricultural products to be marketed, as well as the listing date of agricultural products to be marketed, the generation plan of the marketing video for the optimization target is determined. Based on the aforementioned generation scheme, a marketing video for the optimization objective is generated. Using a marketing video generation scheme that is related to the marketing label of the agricultural product to be marketed as a basis, different intelligent understanding strategies for marketing videos are determined. Based on the understanding results obtained from these intelligent understanding strategies, the updated results for the optimization objective of the marketing scheme are determined.

2. The agricultural big data analysis method based on intelligent understanding as described in claim 1, characterized in that, The agricultural products mentioned are those for which marketing videos will be generated.

3. The agricultural big data analysis method based on intelligent understanding as described in claim 1, characterized in that, The agricultural products to be marketed are those that need to be marketed within a predetermined time period in the future.

4. The agricultural big data analysis method based on intelligent understanding as described in claim 1, characterized in that, The association between the marketing labels of the agricultural products and the agricultural products to be marketed is determined based on the same number of marketing labels on the agricultural products and the agricultural products to be marketed.

5. The agricultural big data analysis method based on intelligent understanding as described in claim 4, characterized in that, The marketing labels for the agricultural products include pollution-free, green, organic, pesticide-free, traceable, freshly picked, place of origin, and suitable user group.

6. The agricultural big data analysis method based on intelligent understanding as described in claim 1, characterized in that, The method for determining the optimization objectives of the marketing plan for the aforementioned agricultural products is as follows: Based on the association between the marketing labels of the agricultural products and the marketing labels of the agricultural products to be marketed, determine the marketing labels that are associated with the marketing labels of the agricultural products and the agricultural products to be marketed, and use them as associated marketing labels. Based on the associated marketing label data of different agricultural products to be marketed, the associated agricultural products are determined; Based on the associated agricultural product data, determine whether the agricultural product is the target of the marketing plan optimization.

7. The agricultural big data analysis method based on intelligent understanding as described in claim 6, characterized in that, The related agricultural products are agricultural products awaiting market launch that have multiple related marketing labels.

8. The agricultural big data analysis method based on intelligent understanding as described in claim 1, characterized in that, The method for determining the updated results of the optimization objectives of the marketing plan is as follows: Based on the understanding results obtained from the aforementioned intelligent understanding strategy, the browsing user data of the marketing video corresponding to the marketing label of the agricultural product to be marketed is determined; Based on the browsing user data, determine the number of users who viewed the marketing videos associated with the marketing tags of the agricultural products to be marketed; Based on the marketing tags of the agricultural products to be marketed and the number of users who viewed the associated marketing videos, the updated results of the optimization goals of the marketing plan are determined.

9. The agricultural big data analysis method based on intelligent understanding as described in claim 8, characterized in that, Based on the number of users who viewed the marketing videos corresponding to the marketing tags of the agricultural products to be marketed, the updated results of the optimization objectives of the marketing plan are determined, specifically including: The total number of users who viewed the marketing videos associated with the marketing tags of the agricultural products to be marketed was determined by associating them with the marketing tags of the agricultural products to be marketed. When the total number of users browsing different marketing tags for all agricultural products awaiting market launch meets the requirements, the generation of marketing videos for all agricultural products awaiting market launch can effectively determine the content that users are interested in. Therefore, the update result of the optimization goal of the marketing plan is determined to be that no update processing is required.

10. A computer system, comprising: A memory and processor connected by communication, and a computer program stored in the memory and capable of running on the processor, characterized in that, when the processor runs the computer program, it executes an agricultural big data analysis method based on intelligent understanding as described in any one of claims 1-9.

Images