A bC integrated marketing method based on one thing one code

Abstract

The application provides a bC integrated marketing method based on one code for one thing, and particularly relates to the field of bC integrated marketing. The Neural-Symbolic reasoning method is introduced into the bC integrated marketing system of one code for one thing, the b-end control strategy and the C-end incentive path suitable for the current attributes of goods, the sales scene and the user label are dynamically reasoned and bound, the one-to-one mapping relationship between the entity code and the digital twin code is constructed, the adaptive evolution and remote update of the strategy are realized based on the code scanning behavior in the process of goods flow, thereby solving the problems of fixed strategy, rigid response, low utilization rate of behavior data and difficulty for the B-end to realize real-time perception of C-end feedback in the existing one code for one thing system, realizing the strategy individualization of the goods granularity level, the logic dynamic update driven by the code scanning behavior and the visual feedback of the code scanning heat, and greatly improving the bC linkage efficiency and the precision of the marketing strategy.

Description

Technical Field

[0001] This invention relates to the field of bC integrated marketing, specifically a bC integrated marketing method based on one item, one code. Background Technology

[0002] With the rapid development of the Internet of Things (IoT) and digital marketing, one-item-one-code technology is widely used in areas such as product traceability, anti-counterfeiting, and user interaction. In existing B2C integrated marketing systems, assigning a unique physical code to each product enables the collection of basic data on product circulation and consumer behavior. Combined with big data analytics and user profiling, merchants can formulate specific marketing strategies, such as QR code incentives and coupon distribution, to encourage consumer participation and improve sales. Simultaneously, some systems have begun analyzing product sales channels and user feedback to adjust marketing strategies and control warehousing, achieving integrated management of both B2B and B2C ends.

[0003] However, existing technologies largely rely on static or pre-defined strategy rules, lacking a deep understanding and real-time response capability for the multi-dimensional dynamic behavior of goods during circulation. Traditional methods struggle to achieve synchronized updates between product codes and digital twins, leading to a disconnect between scanning behavior feedback and marketing strategies, resulting in delayed strategy adjustments and hindering real-time optimization and precise targeting of marketing effectiveness. Furthermore, insufficient real-time visualization support for consumer scanning heatmap data at the warehousing end makes it difficult to comprehensively control the status of goods circulation and market response, limiting the B2B's ability to manage C2C behavior with precision. To address these shortcomings, there is an urgent need for an innovative marketing method capable of dynamically iterating product code strategies and achieving real-time interaction between scanning behavior and digital twins, thereby improving the intelligence and responsiveness of integrated B2B / C marketing systems. Summary of the Invention

[0004] This invention proposes a one-item-one-code integrated bC marketing method, aiming to solve the problems of disconnect between product scanning behavior feedback and marketing strategy, lagging strategy adjustment, and lack of real-time visualization of scanning heat at the warehouse end in existing bC integrated marketing systems. By using a one-item-one-code Neural-Symbolic fusion reasoning method, it realizes dynamic binding and strategy iteration between product codes and digital twin codes, collects and analyzes scanning behavior events in real time, dynamically adjusts b-end control strategies and c-end incentive paths, and ultimately achieves intelligent marketing management with bC-end collaborative optimization.

[0005] One example of a B2C integrated marketing method based on a unique product code includes the following steps:

[0006] S1. Based on the product's attribute parameters, sales scenario, and user tags, generate a feature vector for the product, and use the Neural-Symbolic method to reason about the bC strategy combination bound to the feature vector. Make a joint decision on the b-end control strategy and C-end incentive path that should be adopted in the current stage of the product, and output the bC strategy combination structure.

[0007] Specifically, the above steps, based on the product's inherent attribute parameters (such as category, SKU, shelf life, etc.), sales scenarios (such as online promotions, offline counters, regional distribution, etc.), and user tags (such as user profiles, activity levels, consumption habits, etc.), perform structured feature modeling of the product, generating a high-dimensional feature vector. This vector is fed into the Neural-Symbolic inference engine. First, the neural network part performs pattern recognition and similarity learning on historical strategy effect data, initially obtaining multiple candidate sets of feasible strategy combinations. Then, the symbolic logic part performs rule filtering and logical consistency verification on the candidate strategies to ensure that the strategies are consistent with the current product status and marketing policies. Finally, the Neural and Symbolic parts jointly make decisions, outputting a complete bC strategy combination structure that simultaneously includes b-end control strategies and C-end incentive paths, achieving personalized strategy matching for the current product status.

[0008] S2. Based on the bC strategy combination obtained through reasoning, generate an entity code for the product, and synchronously generate a digital twin code containing response logic for each entity code. The entity code and the digital twin code are bound together through the product's unique code to form a one-to-one code pair mapping relationship.

[0009] Specifically, the above steps, based on the bC strategy combination derived in S1, generate a "physical code" for the specific product entity. This code contains static identifiers and strategy-directing information, used for the actual deployment of the product in logistics and sales. Simultaneously, a "digital twin code" is generated. Its structure includes response logic for scanning events, incentive control parameters, and status update identifiers, serving as a virtual mapping model of product behavior. The physical code and digital twin code are bound together through a unique product coding field, ensuring that any C-end scanning action can be accurately mapped to the corresponding digital twin logic, thus forming a one-to-one code pair mapping relationship. This achieves bidirectional closed-loop mapping and management of each product in the real world and the virtual system.

[0010] S3. Collect timestamps, spatial coordinates, and response link information from product scanning events, input them into a Neural-Symbolic structure for behavior analysis and strategy adjustment, output new strategy combinations, and update the logical content of the original digital twin code based on the new strategy combinations;

[0011] Specifically, after a product is placed on the market, users' scanning behavior generates event data, including scanning timestamps, spatial coordinates of the scanning location, and the actual response path triggered after scanning. This information is accurately collected according to the mapping relationship between the physical code and the digital twin code, forming a spatiotemporal data stream of product scanning behavior. This behavioral data is input into a Neural-Symbolic structure. The neural part is used to identify behavioral feature patterns and trend changes, while the symbolic part evaluates the actual adaptability and potential failure risk of the currently bound strategy. If the inference result shows that the current strategy combination is no longer applicable under the current behavior, the system will generate a new bC strategy combination based on the inference result and update the response logic and incentive parameters in the corresponding digital twin code through a remote interface, thereby realizing strategy evolution. If the strategy does not trigger the evolution condition, scanning behavior continues to be collected, maintaining the original strategy structure, thus forming a dynamic cyclic update mechanism.

[0012] S4. Based on the scanning events recorded in the updated digital twin code, perform coordinate clustering statistics on the scanning location field of the C-end, generate a scanning density distribution map, and call its heat distribution results on the warehouse side according to the product number to visualize the B-end warehouse.

[0013] Specifically, the updated digital twin retains structured scanning event information, especially the geographic coordinates of user scans. The system first standardizes this coordinate data and removes outliers, then uses clustering algorithms (such as DBSCAN or Mean-Shift) for spatial clustering analysis to identify C-end scanning hotspots and their density distribution. The analysis results are generated as a scan density distribution map, forming a heatmap data structure, and categorized and stored according to the product's unique identifier. B-end warehouse managers can access this heatmap to visually monitor the active scanning areas for different products within the warehouse system, assisting in subsequent business decisions such as warehousing, replenishment, and strategy adjustments, thus constructing a dynamic perception loop that reflects C-end behavior back to B-end management.

[0014] Furthermore, step S1, by inputting product attributes, sales scenarios, and user tags into the Neural-Symbolic structure for joint inference, dynamically generates the optimal bC strategy combination for each product. This breaks away from the traditional static approach of "fixed strategy + batch coding," achieving "on-demand strategy, code-driven strategy," improving the fit between products and user scenarios, and enhancing the flexibility of strategy response and targeted delivery effectiveness. Step S2, through a one-to-one binding mechanism between physical codes and digital twin codes, ensures that each product not only possesses a unique physical identification code but also logical response capabilities and strategy status tracking capabilities. Scanning behavior is no longer merely a trigger point for information display but a real-time, feedback-enabled, and updatable behavioral entry point, constructing a digital closed loop for product behavior management and improving the utilization rate of scanning events and the quality of strategy response. Step S3, by collecting real behavioral data such as scanning time, spatial location, and response chain, and inputting it into the Neural-Symbolic inference engine for strategy adaptability and failure risk assessment, allows the system to automatically identify the lifecycle status of the strategy. When the current strategy becomes ineffective, it can automatically evolve and be updated remotely, effectively avoiding "strategy stagnation" or "incentive waste," and improving strategy efficiency and lifecycle management. Step S4 involves clustering and constructing heatmaps from the scanned geographic coordinate data. The B-end warehousing system can then view the real-time scanning activity of goods in various regions, thereby assisting in refined management such as regional deployment optimization, replenishment priority adjustment, and promotional schedule arrangement. Compared to the traditional model that relies solely on sales data for warehousing decisions, this enhances the visualization value of C-end behavioral data for B-end systems, building a perception-driven warehousing response mechanism.

[0015] The beneficial effects of the invention are:

[0016] This invention introduces the Neural-Symbolic reasoning method into a one-item-one-code integrated B2C marketing system. It dynamically infers and binds B2C control strategies and C2C incentive paths adapted to the current attributes of the product, sales scenarios, and user tags. It constructs a one-to-one mapping relationship between physical codes and digital twin codes, and realizes adaptive evolution and remote updates of strategies based on scanning behavior during the product circulation process. This solves the problems of fixed strategies, rigid responses, low utilization of behavioral data, and difficulty for B2C to perceive C2C feedback in real time in existing one-item-one-code systems. It realizes product-level strategy personalization, dynamic logical updates driven by scanning behavior, and visualized feedback of scanning heatmaps, which greatly improves the efficiency of B2C linkage and the accuracy of marketing strategies. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a bC integrated marketing method based on one item, one code, as provided in an embodiment of the present invention. Detailed Implementation

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention; that is, the described embodiments are only a part of the embodiments of the invention, and not all of them. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention. It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0021] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or machine that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or machine. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or machine that includes said element.

[0022] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0023] Example 1

[0024] like Figure 1 A B2C integrated marketing method based on one item, one code includes the following steps:

[0025] S1. Based on the product's attribute parameters, sales scenario, and user tags, generate a feature vector for the product, and combine the Neural-Symbolic method to reason about the bC strategy combination bound to it. Make a joint decision on the b-end control strategy and C-end incentive path that should be adopted for the current stage of the product, and output the bC strategy combination structure.

[0026] S2. Based on the bC strategy combination obtained through reasoning, generate an entity code for the product, and synchronously generate a digital twin code containing response logic for each entity code. The entity code and the digital twin code are bound together through the product's unique code to form a one-to-one code pair mapping relationship.

[0027] S3. Collect timestamps, spatial coordinates, and response link information from product scanning events, input them into a Neural-Symbolic structure for behavior analysis and strategy adjustment, output new strategy combinations, and update the logical content of the original digital twin code based on the new strategy combinations;

[0028] S4. Based on the scanning events recorded in the updated digital twin code, perform coordinate clustering statistics on the scanning location field of the C-end, generate a scanning density distribution map, and call its heat distribution results on the warehouse side according to the product number to visualize the B-end warehouse.

[0029] In steps S1 and S3, the neural network plays a crucial role in automatically extracting deep-level features from multi-dimensional product attributes and barcode scanning behavior data. Specifically, the network architecture is built based on a multilayer perceptron (MLP) or combined with temporal models (such as LSTM or Transformer) to adapt to the nonlinear relationships of product attributes and the spatiotemporal dynamics of barcode scanning behavior. The input layer receives structured, standardized feature vectors (including product category, SKU, sales scenario, user tags, etc.), and through nonlinear transformations in the hidden layers, extracts latent feature representations closely related to the bC strategy's effectiveness, enhancing the model's adaptability to complex environments. Further, in step S1022, the neural network encoder maps the standardized features of the current product to a high-dimensional feature space, while simultaneously encoding the performance features of strategy combinations stored in the historical database. Through the weighted cosine similarity index introduced during training, the model can quickly and accurately identify the strategy combination that best matches the current product and has excellent historical performance, forming a potentially effective candidate set. In step S3023, the neural network acts as an encoder, transforming the multi-dimensional scanning behavior feature matrix (time interval, path offset, response link) into a low-dimensional feedback vector, capturing the spatiotemporal patterns and response features of the scanning behavior. This vector, in conjunction with the symbolic logic module, participates in inference, leveraging both the generalization and learning capabilities of the neural network and the logical rigor of the business rules.

[0030] In addition, in step S1, the Neural-Symbolic method is used to combine multi-dimensional product attributes with business rules to achieve joint reasoning and decision-making for product bC strategy combinations. First, the neural network module is responsible for performing multi-layer nonlinear mapping on the structured attribute parameters of the product (including category, SKU, sales scenario, user tags, etc.), automatically extracting high-dimensional latent feature representations. These representations capture the complex intrinsic relationship between product attributes and the sales environment. Subsequently, the symbolic reasoning module formalizes the strategy constraints and logical relationships defined in the business rule base into symbolic expressions, and performs compliance verification and logical screening based on the candidate strategy set extracted by the neural network. The two complement each other: the neural network is responsible for data-driven feature abstraction and similarity matching, while symbolic reasoning ensures the legality of strategy combinations and consistency with business logic. Finally, through a multi-objective optimization algorithm, joint decision-making is performed on the strategy set filtered by symbolic reasoning, outputting the optimal combination structure of b-end behavior control strategy and c-end incentive path that meets the needs of the current product stage. This Neural-Symbolic fusion mechanism combines the adaptive capabilities of deep learning with the interpretability of symbolic logic, enabling dynamic reasoning on complex, multi-dimensional product features and business rules, effectively improving the accuracy and compliance of strategy combinations. In step S3, the Neural-Symbolic structure focuses on the dynamic analysis and strategy adjustment of scanning behavior feedback. The neural network encoder receives the constructed scanning behavior feature matrix, including the time interval of scanning events, spatial path offset, and response link information. It extracts low-dimensional feedback vectors of scanning behavior through a deep neural network, capturing the spatiotemporal dynamic patterns of scanning behavior and user response characteristics. The symbolic logic module then performs semantic interpretation of the feedback vectors based on business rules, calculates symbolic logic scores, and evaluates the adaptability and potential failure risk of the current strategy combination in actual scanning behavior. The two parts combined constitute the Neural-Symbolic inference engine for feedback analysis, enabling accurate identification of scanning behavior and real-time determination of strategy effectiveness. When the symbolic logic score and neural network feedback jointly indicate that the existing policy is ineffective or mismatched, the inference engine triggers policy evolution, dynamically generating new bC policy combinations, and updating the response logic and stimulus parameters in the digital twin code in real time through a remote interface. This mechanism organically integrates deep learning driven by barcode scanning behavior data with symbolic logic inference, realizing intelligent iteration of the digital twin code and dynamic adaptive adjustment of the bC policy.

[0031] Furthermore, step S1 specifically includes the following sub-steps:

[0032] S101. Based on product category, SKU, target sales region and user profile, perform structured processing on product attribute data, construct standardized feature vectors, and input the standardized feature vectors into the Neural-Symbolic fusion inference framework.

[0033] S102. Call the trained neural network model to perform similarity matching on the performance characteristics of the product and the bC strategy combination in the historical database, and identify potential effective strategy combination candidates, wherein the bC strategy combination includes b-end control strategy and C-end incentive path;

[0034] S103. Combining the policy constraints set in the business rule base, the symbolic reasoning engine performs compliance screening and logical verification on the candidate policy combinations output by the neural network, and eliminates policy paths that do not conform to the rules.

[0035] S104. Based on the strategy set filtered by symbolic reasoning, a multi-objective optimization method is used for joint decision-making to determine and output the optimal strategy combination of the b-end behavior control strategy and the C-end incentive path that should be bound to the product at the current stage.

[0036] Specifically, by modeling the structured attributes of products, a standardized feature vector is established as the input for inference. First, a trained neural network is used to extract the matching degree between product features and historical strategies, and candidate strategy combinations with high similarity are selected from the historical database. Then, the candidate results are subjected to constraint judgment and rule verification by a symbolic inference logic structure. Based on the preset business logic expression, strategy paths with illogical or conflicting business logic are filtered out. Finally, an optimization algorithm is used to jointly evaluate the objectives of the strategies that meet the constraints (such as maximizing ROI and balancing user activity) to output the most suitable combination of B-end control and C-end incentive paths for the current stage of the product's life cycle, realizing a dynamic strategy matching process with the joint participation of data-driven and symbolic decision-making.

[0037] Furthermore, in step S102, the B-end control strategy includes warehouse scheduling, shelf deployment, inventory rhythm, batch strategy, code lifecycle setting and product permission configuration, and the C-end incentive path includes scan reward parameters, behavior chain guidance, incentive presentation form, user verification mechanism, cumulative behavior tracking, sharing and fission path and incentive dynamic update mechanism.

[0038] Furthermore, step S102 specifically includes the following sub-steps:

[0039] S1021. Read the stored strategy combination vectors from the historical bC strategy database. Each strategy combination vector corresponds to the performance characteristics of a strategy combination.

[0040] S1022. Calculate the weighted cosine similarity between the standardized feature vector of the current product and the vector of each historical strategy combination;

[0041] S1023. Using a preset similarity threshold as the screening criterion, select the strategy combination that is most relevant to the current product and has a good historical performance to form a candidate strategy set.

[0042] Specifically, in S1021, the system first accesses the historical bC strategy database through an interface. Each record in this database contains complete strategy combinations and their performance evaluation results in historical applications. These strategy combinations are typically preprocessed into fixed-length vectors, with each vector dimension corresponding to specific strategy execution parameters or behavioral performance indicators, such as barcode response latency, user incentive acceptance, inventory turnover cycle, and C-end viral propagation range, used to characterize the comprehensive behavioral performance of the combination in real circulation scenarios. The system sequentially reads these strategy combination vectors through the data processing module and caches them in memory as a reference benchmark for subsequent matching operations. In S1022, the system uses the standardized feature vector generated for the current product in S101 as the target vector and calls the matching engine to calculate the similarity between it and the historical strategy combination vectors one by one. During the calculation process, the system does not use the general cosine formula but introduces a feature importance model. The system learns the contribution of each dimension to the matching quality through the model and applies differentiated weights to the feature dimensions when calculating similarity, so that dimensions with higher weights have a greater impact in the matching process, thereby improving the accuracy of strategy recognition and the stability of behavior prediction. In S1023, the system makes threshold judgments based on similarity results. The system sets a set of floating similarity thresholds to automatically adapt to the differences in behavioral characteristics of strategy reuse across different product categories. When the matching score of a strategy combination is higher than the threshold and its historical performance meets the system's built-in performance optimization criteria, the strategy will be included in the candidate set, along with its source and matching score as input to the subsequent rule engine. This ensures that candidate strategies are not only structurally similar but also possess traceability and adaptability in terms of actual effectiveness. This process enhances the effectiveness of strategy selection through the synergy of neural matching and historical evidence, and is a key manifestation of the "experience-driven" module in the Neural-Symbolic architecture.

[0043] Furthermore, step S103 specifically includes the following sub-steps:

[0044] S1031. Load the constraints involving B-end control and C-end incentive paths from the current business rule base;

[0045] S1032. For each candidate strategy combination, the symbolic reasoning engine is used to apply the business rules sequentially, mapping the parameters of the strategy combination to the rule expression and performing logical verification.

[0046] S1033. Mark strategy combinations that fully satisfy all rule conditions as compliant, remove non-compliant strategy combinations, and output the strategy set filtered by symbolic reasoning.

[0047] Specifically, in S1031, the system dynamically loads the rule set in effect for the current operational phase from a pre-built business rule base. This rule set covers B-end control dimensions (such as inventory upper and lower limits, allocation restrictions, shelf configuration requirements, warehouse priority, etc.) and C-end incentive path constraints (such as user participation thresholds, incentive presentation methods, guidance link length limits, platform compliance requirements, etc.). These rules are stored in the form of formal logic expressions and have callable interfaces for easy subsequent inference. In S1032, for each candidate strategy combination selected in S102, the system maps its parameter fields as input items to corresponding variables in the rule expression, and executes the rule chain verification process through the symbolic inference engine. This inference process is based on first-order logic or descriptive logic semantics to determine whether the strategy combination semantically satisfies all constraints, and dynamically triggers rule cascading on the dependency chain to ensure that the logical consistency within all rule sets is fully verified. If a conflict, dependency violation, or failure to meet preconditions occurs during the inference process, the strategy combination will be immediately judged as non-compliant. In S1033, the system will uniformly mark all strategy combinations that have passed the symbol rule chain verification as "compliant" and remove combinations that have not passed verification from the candidate set.

[0048] Furthermore, step S2 specifically includes the following sub-steps:

[0049] S201. Generate a unique product entity code based on the coding rules, lifecycle, and incentive parameters in the strategy combination; and construct the corresponding digital twin code data structure based on the data content in the strategy combination.

[0050] S202. Based on the product's unique code, perform a dual-code binding operation between the physical code and the digital twin code, and synchronously write the bound code pair information into distributed storage.

[0051] Furthermore, in step S201, the digital twin code data structure includes scanning response logic, excitation control parameters, and dynamic update identifier.

[0052] Specifically, in S201, a unique identifier code is generated for each product based on the coding rules included in the selected strategy combination. These coding rules comprehensively consider product lifecycle management, incentive parameter settings, and product attributes to ensure the uniqueness and traceability of the physical code throughout the entire supply chain. Simultaneously, based on the detailed data content in the strategy combination, a corresponding digital twin code data structure is constructed. This data structure integrates scanning response logic, incentive control parameters, and dynamically updated identifiers, enabling real-time mapping of the product's status and behavioral responses during circulation, forming a digital product mapping entity. In S202, the generated unique product code is used as an anchor point to bind the physical code and digital twin code one-to-one, establishing a robust code-to-code mapping relationship. After binding, the relevant code-to-code information is synchronously written to a distributed storage environment, ensuring efficient access and data consistency across multiple nodes.

[0053] Furthermore, step S3 specifically includes the following sub-steps:

[0054] S301. Based on the mapping relationship between the physical code and the digital twin code, collect the timestamp, user's geographical location and response path of each product scanning event to construct a scanning behavior time series;

[0055] S302. Based on the scanning time series, calculate the scanning frequency, path offset, and stimulus response rate to form a feedback vector; and based on the feedback vector, call the Neural-Symbolic inference engine to analyze the adaptability and failure risk of the current strategy combination in actual behavior, and determine whether to trigger strategy evolution.

[0056] S303. When policy evolution is triggered, a new bC policy combination is dynamically generated based on the analysis results, and the response logic and stimulus parameters in the corresponding digital twin code are updated in real time through the remote interface to complete the policy iteration; when policy evolution is not triggered, step S301 is repeated.

[0057] Specifically, in S301, based on the mapping relationship between physical codes and digital twin codes, detailed data for each product scanning event is collected in real time, including the scanning timestamp, user's geographical location information, and the response chain triggered by the scanning. A complete scanning behavior time series is constructed through continuous events, reflecting the dynamic interaction process of goods in the circulation process. In S302, this scanning time series is analyzed, calculating the scanning frequency to measure user activity, measuring scanning path offset to capture changes in geographical distribution, and simultaneously calculating the incentive response rate to evaluate the actual effect of incentive measures. These indicators are combined to form a multi-dimensional feedback vector. Subsequently, the feedback vector is input into the Neural-Symbolic inference engine. The system integrates the behavioral feature abstraction of the neural network with the rule constraints of symbolic logic to comprehensively evaluate the adaptability and potential failure risk of the current strategy combination in the real behavioral environment, determining whether the threshold for triggering strategy evolution has been reached. In S303, if the triggering condition is met, a new bC strategy combination is dynamically generated based on the reasoning analysis results. This combination includes an updated b-end control strategy and a C-end incentive path. The corresponding response logic and incentive parameters in the digital twin code are also updated synchronously through a remote interface to achieve real-time iterative optimization of the strategy. If the strategy evolution is not triggered, the scanning data continues to be collected, and S301 is executed repeatedly to ensure that the strategy is continuously adaptively adjusted based on the latest behavioral data.

[0058] Furthermore, step S302 specifically includes the following sub-steps:

[0059] S3021. Based on the QR code scanning event records, extract the timestamp, spatial coordinates, and response link information to form a QR code scanning event set: , wherein This represents the set of QR code scanning events, the... This represents the record of the i-th scanning event. These represent the timestamp, x-coordinate, y-coordinate, and response link information of the i-th scanning event, respectively. The index representing the number of QR code scanning events, the This represents the total number of QR code scanning events collected within the statistical time frame.

[0060] S3022. Construct a scanning interval vector based on the timestamp sequence corresponding to the timestamp: , wherein Represents the scanning interval vector, the This represents the time difference between the i-th scan and the (i-1)-th scan, used to reflect the time interval between scans. This represents the timestamp of the (i-1)th scan event.

[0061] Calculate the difference in scanning path vectors based on spatial coordinates: , wherein The difference in the scanning path vector reflects the change in spatial distance between consecutive scanning events. The distance between the i-th scanning event and the (i-1)-th scanning event is represented by the following. The horizontal coordinate of the spatial coordinates of the (i-1)th scanning event is represented by the following: The ordinate represents the spatial coordinates of the (i-1)th scan event;

[0062] By combining the response link strength sequence, a behavioral feature matrix is ​​formed: , wherein The behavioral feature matrix is ​​a multi-dimensional feature set composed of scanning event intervals, path differences, and response link information.

[0063] S3023. Input the behavior feature matrix into the pre-trained neural network encoder, output a feedback vector of low-dimensional feature representation, and combine it with the symbolic logic module of the policy rule base to infer whether the current scanning behavior conforms to the expected policy and calculate the symbolic logic score.

[0064] S3024. Based on the neural network output and symbolic logic score, define an evolution trigger function, and determine whether to trigger policy evolution based on the evolution trigger function.

[0065] Furthermore, in step S3024, the evolution trigger function is specifically represented as follows: , wherein The strategy evolution trigger flag is a boolean variable indicating whether strategy evolution is currently triggered; 1 indicates triggering, and 0 indicates not triggering. The weight coefficients representing the offset portion of the behavior feedback vector, the Represents the feedback vector, the Represents the reference behavior feature vector, the The weights representing the score s of the symbolic logic module, the This represents the decision threshold that triggers evolution, i.e., the sum of the maximum tolerance for deviation and mismatch in the system; when When =1, trigger policy evolution and perform policy combination adjustment; otherwise, maintain the current policy.

[0066] Furthermore, step S4 specifically includes the following sub-steps:

[0067] S401. Based on the geographic coordinate data of the scanning events recorded in the updated digital twin, identify scanning hotspots using a spatial clustering algorithm, and calculate the scanning density and activity level of each area;

[0068] S402. Generate a scan heat map of the corresponding product by combining scan density and activity level;

[0069] S403. B-end staff can access the barcode scanning heatmap through the warehouse management terminal, and analyze the barcode scanning activity area by combining time and batch dimensions to monitor the circulation of goods and marketing strategies.

[0070] Specifically, in S401, using the real-time updated and recorded geographic coordinate data of scanning events in the digital twin, the coordinate data is first preprocessed, including outlier removal and coordinate standardization. Then, spatial clustering algorithms (such as DBSCAN or K-means) are used to group the scanning points, identifying densely scanned hotspots. Based on the number and time distribution of scanning events within each region, scanning density and activity indicators are calculated, quantifying the scanning activity and user interaction intensity in different areas. In S402, based on the scanning density and activity data of each hotspot area, and combined with geographic information system (GIS) technology, the data is transformed into a visualized scanning heatmap. This map intuitively reflects the user attention and scanning behavior intensity of products in different geographical locations, facilitating in-depth insights into regional market activity and consumer habits. In S403, B-end staff can access the barcode scanning heatmap through the warehouse management terminal. By combining time and batch information, they can conduct multi-dimensional analysis of active barcode scanning areas, identify sales trends and potential market opportunities, and thus scientifically adjust and optimize commodity circulation paths, inventory allocation and marketing strategies, achieving precise management and decision support based on real-time data.

[0071] Example 2

[0072] Furthermore, as a preferred embodiment of the above embodiments, a detailed calculation process example of the symbolic logic module described in step S3024 of the above embodiments is given. First, define:

[0073] The current product's bC strategy combination for: Each strategy element Includes policy thresholds, execution conditions, and triggering logic structure;

[0074] Anti-expensive vector of QR code scanning behavior Represented as: The scanning frequency is respectively Frequency fluctuation mean path offset Path offset fluctuation Incentive response rate ;

[0075] Each rule in the logic rule base is a formal logic expression: This indicates that a certain behavioral feature should be mapped to a certain logical decision in the strategy dimension. This represents the k-th symbolic logic rule, the... The precondition for this rule is a logical expression about the feedback vector F, used to determine whether the scanning behavior feedback meets a certain condition. The conclusion of this rule refers to the j-th policy element in the current policy combination. The logical proposition indicates whether the strategy element meets the rule requirements.

[0076] Secondly, construct the feedback logic mapping function. Each rule constructs a specific condition based on the feedback vector, for example:

[0077] High-frequency scanning area: ;

[0078] High path offset fluctuations indicate dispersed user activity.

[0079] A low incentive response rate indicates weak attractiveness of the strategy. ;

[0080] These logical conditions all return a Boolean value. ;

[0081] Among them, the , , These represent the scanning frequency threshold, path offset threshold, and stimulus response rate threshold in the feedback logic, respectively, used to determine whether the feedback condition is triggered.

[0082] Next, construct the strategy proposition mapping function. For the current strategy combination A strategy element in The symbol module extracts its logical propositions, for example:

[0083] Strategy Proposition 1: Currently, the "weak stimulus" parameter is set for the "low-frequency scanning area":

[0084] Strategy Proposition 2: Users with currently "dispersed paths" should not be repeatedly targeted with the same type of incentive. ;

[0085] Among them, the This represents the threshold of the excitation intensity.

[0086] Secondly, logical rule matching function Combining feedback premises and strategy propositions into implication logic: ;

[0087] For example:

[0088] like and If the rule does not match, then the rule is not applicable.

[0089] like Logical implication is always true (does not violate the rules);

[0090] final .

[0091] Finally, summarize the symbolic logic scores. The final symbolic logic fit score is obtained by weighted summation of all rules:

[0092] ;

[0093] Among them, the The index representing the number of all symbolic logic rules in the rule base, the This represents the number of all symbolic logic rules in the rule base. This represents the weight of the k-th symbolic logic rule, i.e., the importance of this rule in the overall judgment.

[0094] Example 3

[0095] Furthermore, as a preferred implementation of the above embodiments, an application scenario for a one-item-one-code integrated marketing method based on bC is proposed. Specifically, a large FMCG company wants to increase the market penetration and user engagement of its newly launched beverage products. It adopts the one-item-one-code integrated marketing method based on bC of this invention, combined with Neural-Symbolic inference technology, to realize dynamic strategy adjustment and precise incentives throughout the entire commodity circulation chain.

[0096] The company first collects data such as the attribute parameters of beverage products, target sales regions, and user tags. Standardized feature vectors are then constructed through data structuring and input into a Neural-Symbolic inference framework. The neural network model matches historically successful bC strategy combinations as candidates, such as "warehouse allocation scheduling - QR code coupon offering" and "shelf control - behavioral chain guidance." The symbolic inference engine filters compliant strategies based on company-defined rules (inventory limits, activity compliance) and combines multi-objective optimization to determine the optimal strategy combination: for example, prioritizing warehouse allocation scheduling in East China, and using QR code points + sharing-based referral incentives for end-users.

[0097] Based on a defined strategy combination, the system automatically generates a unique entity code, along with a batch number and lifecycle information. This entity code is printed on the beverage bottle. Simultaneously, a digital twin code is generated, containing reward rules, point amounts, validity period, and dynamic update indicators. This digital twin is then bound to the entity code via a unique product code and written to a distributed database.

[0098] After product launch, the system collects real-time scanning event data, including scanning timestamps, geographic locations, and the user's response path after scanning. Based on the scanning time series, the system calculates scanning frequency, path offset, and incentive response rate to form a feedback vector. This vector is input into the Neural-Symbolic inference engine to calculate a symbolic logic score and evaluate the adaptability of the current strategy. When scanning behavior deviates from expectations—for example, a decrease in scanning rate at a store, insufficient sharing and viral marketing, and a symbolic logic score below a threshold—strategy evolution is triggered. The system automatically generates new bC strategy combinations, such as increasing the amount of scanning reward points or adjusting the sharing incentive method, and updates the digital twin's response logic in real-time via a remote interface to ensure a rapid closed loop between strategy and market feedback.

[0099] Based on the geographic coordinates of scanning events recorded by digital twin codes, spatial clustering algorithms such as DBSCAN are used to identify scanning hotspots and statistically analyze scanning density and activity. A scanning heatmap is generated, marking active stores and inactive areas. B2B warehouse managers can view the heatmap through a management interface and analyze product circulation in conjunction with batch and time dimensions. Through data-driven adjustments, warehouse allocation and marketing strategies are refined, increasing promotional resources in low-scanning-density areas shown on the heatmap.

[0100] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A bC integrated marketing method based on one thing one code, characterized in that, The method comprises the following steps: S1. According to the attribute parameters of the commodity, the sales scene and the user label, a feature vector of the commodity is generated, and the Neural-Symbolic method is combined to reason the bC strategy combination bound to the feature vector, to jointly decide the b-end control strategy and the C-end incentive path to be adopted by the commodity in the current stage, and output the bC strategy combination structure; S2. According to the bC strategy combination obtained by reasoning, an entity code of the commodity is generated, and a digital twin code containing response logic is synchronously generated for each entity code, the entity code and the digital twin code are bound through the unique code of the commodity, forming a one-to-one code pair mapping relationship; S3. The time stamp, spatial coordinates and response link information in the commodity scanning code behavior event are collected and input into the Neural-Symbolic structure for behavior analysis and strategy adjustment, and the new strategy combination is output, and the logic content of the original digital twin code is updated according to the new strategy combination; S4. According to the scanning code events recorded in the updated digital twin code, the coordinate clustering statistics of the C-end scanning code position field are carried out, the scanning code density distribution map is generated, and the heat distribution result of the commodity number is called according to the warehouse end, and the b-end warehouse is visually displayed; The b-end control strategy includes warehouse scheduling, shelf control, inventory rhythm, batch strategy, code life cycle setting and commodity permission configuration, and the C-end incentive path includes scanning code reward parameters, behavior chain guidance, incentive presentation form, user verification mechanism, cumulative behavior tracking, sharing fission path and incentive dynamic updating mechanism.

2. The bC integrated marketing method based on one product one code according to claim 1, wherein, The step S1 specifically comprises the following sub-steps: S101. According to the commodity category, SKU, target sales area and user portrait, the attribute data of the commodity is structured and processed, a standardized feature vector is constructed, and the standardized feature vector is input into the Neural-Symbolic fusion reasoning framework; S102. The trained neural network model is called to perform similarity matching on the performance characteristics of the bC strategy combination in the commodity and the historical database, to identify potential effective strategy combination candidates, and the bC strategy combination includes the b-end control strategy and the C-end incentive path; S103. Combined with the strategy constraint conditions set in the business rule library, the candidate strategy combination output by the neural network is screened for compliance and logically verified through the symbolic reasoning engine, and the strategy path that does not comply with the rules is eliminated; S104. According to the strategy set screened by the symbolic reasoning, the optimal strategy combination of the b-end behavior control strategy and the C-end incentive path bound to the commodity in the current stage is determined and output by adopting a multi-objective optimization method for joint decision.

3. The bC integrated marketing method based on one product one code according to claim 2, wherein, The step S102 specifically further comprises the following sub-steps: S1021. The stored strategy combination vectors are read from the historical bC strategy database, and each strategy combination vector corresponds to the performance characteristics of a strategy combination; S1022. The weighted cosine similarity is calculated for the standardized feature vector of the current commodity and each historical strategy combination vector; S1023. Select the strategy combination most relevant to the current commodity and with excellent historical performance as the screening standard with a preset similarity threshold to form a candidate strategy set.

4. The bC integrated marketing method based on one product one code according to claim 2, wherein, The step S103 specifically further includes the following sub-steps: S1031. Load the constraint conditions related to b-end control and C-end incentive path in the current business rule library; S1032. For each candidate strategy combination, use the symbolic reasoning engine to apply business rules one by one, map the parameters of the strategy combination to the rule expression and perform logical verification; S1033. Mark the strategy combination that fully meets all rule conditions as compliant, and exclude the non-compliant strategy combination, and output the strategy set after symbolic reasoning screening.

5. The bC integrated marketing method based on one product one code according to claim 1, wherein, The step S2 specifically includes the following sub-steps: S201. According to the encoding rules, life cycle and incentive parameters in the strategy combination, generate a unique commodity entity code; and according to the data content in the strategy combination, build the corresponding digital twin code data structure; S202. According to the unique code of the commodity, perform double-code binding operation on the entity code and the digital twin code, and synchronize the bound code pair information to the distributed storage.

6. The bC integrated marketing method based on one product one code according to claim 5, wherein, In the step S201, the digital twin code data structure includes scan response logic, incentive control parameters and dynamic update identifier.

7. The bC-integrated marketing method based on one product one code according to claim 1, wherein, The step S3 specifically includes the following sub-steps: S301. According to the mapping relationship between the entity code and the digital twin code, collect the timestamp, user geographic location and response path of each scan event of the commodity, and build a scan behavior time sequence; S302. According to the scan time sequence, calculate the scan frequency, path offset and incentive response rate to form a feedback vector; and according to the feedback vector, call the Neural-Symbolic reasoning engine to analyze the adaptability and failure risk of the current strategy combination in actual behavior, and judge whether to trigger strategy evolution; S303. When the strategy evolution is triggered, a new bC strategy combination is dynamically generated according to the analysis result, and the response logic and incentive parameters in the corresponding digital twin code are updated in real time through the remote interface to complete the strategy iteration; when the strategy evolution is not triggered, repeat step S301.

8. The bC integrated marketing method based on one product one code according to claim 7, wherein, The step S302 specifically includes the following sub-steps: S3021. According to the code scanning event record, the timestamp, spatial coordinates and response link information are extracted to form a code scanning event set: wherein the code scanning event set is represented as represents the i-th code scanning event record, and respectively represent the timestamp, horizontal coordinate of spatial coordinates, vertical coordinate of spatial coordinates and response link information of the i-th code scanning event, and represents the number index of the code scanning event, and represents the total number of code scanning events collected within the statistical time range.​ S3022. According to the time stamp corresponding to the time stamp sequence, the code scanning interval vector is constructed: wherein the indicates the code scanning interval vector, and the indicates the time difference between the i th code scanning and the i−1 th code scanning, which is used to reflect the time interval of code scanning, and the indicates the time stamp of the i−1 th code scanning event; Based on the spatial coordinates, a scanning code path vector difference is calculated: wherein the represents the scanning code path vector difference, used to reflect the spatial distance change between the continuous scanning code events, the represents the spatial distance between the i th scanning code event and the i−1 th scanning code event, the represents the horizontal coordinate of the spatial coordinates of the i−1 th scanning code event, the represents the vertical coordinate of the spatial coordinates of the i−1 th scanning code event. Combine the response link strength sequence to form a behavior feature matrix: Wherein, the The behavior feature matrix is composed of a multi-dimensional feature set of the scanning code event interval, path difference, and response link information. S3023. Input the behavior feature matrix into the pre-trained neural network encoder to output a low-dimensional feature representation feedback vector, and combine the symbolic logic module of the strategy rule library to infer whether the current scan behavior meets the expected strategy and calculate the symbolic logic score; S3024. According to the neural network output and the symbolic logic score, define an evolution trigger function to determine whether to trigger strategy evolution according to the evolution trigger function.

9. The bC integrated marketing method based on one product one code according to claim 8, wherein, The evolution trigger function is specifically represented as: , wherein the is a policy evolution trigger identifier, which is a Boolean variable indicating whether the policy evolution is triggered or not, 1 indicating triggering and 0 indicating not triggering, the is a weight coefficient of the behavior feedback vector offset part, the is a feedback vector, the is a reference behavior feature vector, the is a weight of the symbolic logic module score s, the is a decision threshold value of triggering evolution, i.e., the maximum sum of the offset and the mismatch degree tolerated by the system; when = 1, the policy evolution is triggered, and the policy combination adjustment is performed; otherwise, the current policy is maintained.

10. The bC-integrated marketing method based on one product one code according to claim 1, wherein, The step S4 specifically includes the following sub-steps: S401. According to the geographic coordinate data recorded in the scan event in the updated digital twin code, identify the scan hotspot area through a spatial clustering algorithm, and calculate the scan density and activity of each area; S402. Generate a scan heat distribution map of the corresponding commodity in combination with the scan density and activity; S403. The b terminal staff calls the code scanning heat map through the warehouse management terminal, analyzes the code scanning active area combined with time and batch dimension, and monitors the commodity circulation and marketing strategy.

Images