Intelligent dynamic pushing method and device based on marketing canvas construction, equipment and medium

Abstract

The invention provides an intelligent dynamic pushing method and device based on marketing canvas construction, equipment and a medium, and the method comprises the following steps: obtaining basic data of a marketing canvas, and constructing a marketing canvas model according to the basic data; adjusting a marketing pushing strategy weight according to the real-time behavior and historical preference of the user; on the basis of the adjusted marketing pushing strategy weight, a pushing path is optimized according to user feedback and context information; based on the generation model, automatically generating personalized contents of the user and evaluating a pushing effect in real time; according to the method, a marketing canvas model based on multiple data is constructed, data such as products, customers, channels and communication strategies are combined, and real-time behaviors of users are dynamically analyzed and predicted by using a deep learning model. In this way, the accuracy of the user portrait can be improved, the pushing strategy can be updated and adjusted in real time, and therefore accurate personalized pushing is achieved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an intelligent dynamic push method, device, equipment and medium based on a marketing canvas. Background Art

[0002] As a systematic marketing tool, the marketing canvas can comprehensively consider marketing strategies and execution paths based on multiple data. The current intelligent push system mainly relies on user behavior data or a simple rule engine to push information, but cannot deeply understand user needs or adjust push content in real time.

[0003] Many current marketing push methods are based only on static user data (such as demographic characteristics, browsing history, etc.), lacking the ability to deeply understand user behavior and needs. Traditional methods cannot dynamically adjust push content and strategies, often resulting in low relevance of push content and failure to meet users' changing preferences; In existing technologies, marketing path optimization often relies on static rules or simple machine learning algorithms, lacking a dynamic feedback mechanism. The selection and adjustment of push paths often rely on empirical rules or speculation based on historical data, which makes it difficult to cope with real-time changes in user behavior and market environment; Traditional push content generation mostly relies on manual templates or simple recommendation systems, which makes it difficult to dynamically generate creative content based on user behavior, resulting in push content fatigue and a decline in user experience. In traditional push systems, the evaluation of push effects is often delayed and can only be based on relatively basic indicators (such as click-through rate CTR, conversion rate, etc.), lacking a real-time feedback mechanism to adjust push strategies in a timely manner.

[0004] To this end, an intelligent dynamic push method, device, equipment and medium based on a marketing canvas are proposed. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide an intelligent dynamic push method, device, equipment and medium based on the marketing canvas. Through dynamic optimization and real-time adjustment of push content, path and effect, the present invention can greatly improve the accuracy and efficiency of marketing push, and promote marketing technology to develop in a more efficient and intelligent direction, so as to solve or alleviate the technical problems existing in the prior art and provide at least one of the beneficial options.

[0006] In order to achieve the above-mentioned purpose, a first aspect of the embodiment of the present disclosure provides an intelligent dynamic push method based on a marketing canvas, comprising the following steps: Obtain the basic data of the marketing canvas and build a marketing canvas model based on the basic data; Adjust marketing push strategy weights based on users’ real-time behavior and historical preferences; Based on the adjusted marketing push strategy weights, optimize the push path according to user feedback and contextual information; Automatically generate user-personalized content based on the generation model and evaluate the push effect in real time.

[0007] As a further preferred embodiment of the present technical solution, constructing a marketing canvas model according to basic data specifically includes: Define basic data of the marketing canvas, including product data, customer data, channel data, and communication strategy data; Based on the marketing canvas model, product data, customer data, channel data and communication strategy data are collected to build a multidimensional matrix , the multidimensional matrix Including push strategy and weight; Integrate external data: Enhance your marketing canvas with data from social media and search engines.

[0008] As a further preferred embodiment of the present technical solution, the weight of the marketing push strategy is adjusted according to the real-time behavior and historical preference of the user, specifically including: Obtaining user behavior data, including user browsing history, click behavior, and purchase records; Modeling user behavior data based on a long short-term memory network to predict future user behavior, wherein the long short-term memory network generates a user behavior vector; Adjusting multidimensional matrix based on user behavior vector The weight of marketing push strategy in; Update the multidimensional matrix based on user preferences , make sure the strategy meets the needs of current users.

[0009] As a further preferred embodiment of the present technical solution, the optimization of the push path according to user feedback and context information specifically includes: Path selection is based on the objective function, and push paths are dynamically selected according to the Q-learning reinforcement learning algorithm; Based on each state Plan the optimal push path And adjust the push strategy through the Q value update formula; Based on the feedback and contextual information of the push path, the push strategy is dynamically adjusted to ensure that the path selection matches the current user's needs and behaviors.

[0010] As a further preferred embodiment of the present technical solution, the objective function is a trade-off between the relevance of the pushed content and the push cost, and the specific formula is as follows: ; in: For products and customer groups The matching degree is calculated based on user preferences and product features; For the Products in time The weight of the moment; The push cost includes channel cost and timing cost; is the balance coefficient.

[0011] As a further preferred embodiment of the present technical solution, the Q value updating formula is provided, and the specific contents are as follows: ; in: is the Q value of taking action in the state; For the status Take action Immediate returns; is the discount factor; For the next state All possible actions The maximum value of the medium Q value.

[0012] As a further preferred embodiment of the present technical solution, the automatic generation of user personalized content based on the generation model and real-time evaluation of the push effect specifically include: Generate personalized push content based on generative adversarial networks and customize content based on user behavior and preferences; Measure the attractiveness and conversion effect of content based on click-through rate calculation, and evaluate the push effect; The push content and strategy are adjusted in real time according to the results of the push effect evaluation.

[0013] Second aspect: To solve the above technical problems, another technical solution adopted by this application is: an intelligent dynamic push device based on the marketing canvas, including: The acquisition module is used to obtain the basic data of the marketing canvas and build a marketing canvas model based on the basic data; The adjustment module is used to adjust the weight of marketing push strategies based on users’ real-time behavior and historical preferences; The optimization path module is used to optimize the push path based on user feedback and context information based on the adjusted marketing push strategy weights; The evaluation push module is used to automatically generate user personalized content based on the generation model and evaluate the push effect in real time.

[0014] To solve the above technical problems, another technical solution adopted in the present application is: an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor; when the processor executes the program, the steps of the intelligent dynamic push method based on the marketing canvas as described in any one of the above are implemented.

[0015] To solve the above technical problems, another technical solution adopted in the present application is: a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the intelligent dynamic push method based on the marketing canvas as described in any one of the above are implemented.

[0016] The embodiment of the present invention has the following advantages due to the adoption of the above technical solution: 1. The present invention combines data such as products, customers, channels and communication strategies by building a marketing canvas model based on multiple data, and uses a deep learning model to dynamically analyze and predict users' real-time behaviors. This can not only improve the accuracy of user portraits, but also update and adjust push strategies in real time, thereby achieving accurate personalized push; 2. The present invention dynamically optimizes the marketing path through reinforcement learning, and can automatically adjust the push path by combining real-time user behavior feedback. Through the dynamic optimization of the objective function, it can maximize the balance between the relevance of the push content and the push cost, and achieve more efficient path selection; 3. The present invention introduces a generative adversarial network to automatically generate personalized push content, which can dynamically adjust the content according to the user's real-time behavior and preferences, improve the novelty and attractiveness of the content, and thus enhance user engagement and conversion rate; 4. The present invention evaluates the push effect (such as click-through rate, conversion rate, etc.) in real time and feeds the evaluation results back to the push path and content generation. Through the combination of reinforcement learning and deep learning, the push strategy and content can be adjusted in real time to improve the efficiency and effect of push.

[0017] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present invention will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a flow chart of the intelligent dynamic push method constructed based on the marketing canvas of the present invention; Figure 2 A flowchart for constructing a marketing canvas model for the basic data of the present invention; Figure 3 A flow chart for adjusting the weight of marketing push strategies based on the real-time behavior and historical preferences of users of the present invention; Figure 4 This is a flow chart of the present invention for automatically generating user personalized content based on a generation model and evaluating push effects in real time; Figure 5 It is a structural schematic diagram of an intelligent dynamic push device constructed based on a marketing canvas of the present invention; Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0021] It should be clear that the following embodiments of the present disclosure are described by specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other in the absence of conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0022] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0023] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The drawings only show components related to the present disclosure rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0024] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0025] Embodiment 1 Figure 1 is a flow chart of an intelligent dynamic push method based on a marketing canvas according to an embodiment of the present invention. It should be noted that if there are substantially the same results, the method of the present application is not based on Figure 1 The process sequence shown is limited.

[0026] Traditional marketing push methods mostly rely on simple behavior history or rule engines, but lack dynamic push capabilities based on deep understanding and real-time feedback. To meet this challenge, this solution proposes to dynamically optimize the marketing push path by building a multi-data marketing canvas, combining user behavior data, and using deep learning and reinforcement learning techniques.

[0027] like Figure 1-Figure 4 As shown: The intelligent dynamic push method based on the marketing canvas includes the following steps: Step 10: Obtain the basic data of the marketing canvas and build a marketing canvas model based on the basic data.

[0028] Specifically include: Define the basic data of the marketing canvas, which includes product data, customer data, channel data and communication strategy data; the customer data is divided into: retail customers, high net worth customers, private banking customers, student customers, elderly customers, etc.

[0029] Based on the marketing canvas model, product data, customer data, channel data and communication strategy data are collected to build a multidimensional matrix , the multidimensional matrix Including push strategy and weight; Integrate external data: Enhance the marketing canvas with data from social media and search engines; Specifically, define the basic data of the marketing canvas: Product data: Each product has different characteristics (such as product type, features, discounts, etc.).

[0030] Customer data: Divide customer groups by user profile data (such as gender, age, etc.) age, interests, etc.).

[0031] Channel data: online (e.g., social media, apps) and offline channels (e.g., bank branches).

[0032] Communication strategy data: including the timing of push, content personalization, language style, etc.

[0033] Build a marketing canvas matrix and construct a multidimensional matrix , each item corresponds to a specific product-customer combination, including the push strategy and weight: in: For the The feature vector of the product.

[0034] For the The feature vector of each customer group.

[0035] Each element Indicates the adaptability of marketing push strategies under specific products and customer groups; Enhance the data of the marketing canvas by integrating external data such as social media and search engines to obtain more real-time user feedback.

[0036] Step 20: Adjust the weight of marketing push strategy based on the user's real-time behavior and historical preferences; Specifically include: Obtaining user behavior data, including user browsing history, click behavior, and purchase records; Modeling user behavior data based on a long short-term memory network to predict future user behavior, wherein the long short-term memory network generates a user behavior vector; Adjusting multidimensional matrix based on user behavior vector The weight of marketing push strategy in; Update the multidimensional matrix based on user preferences , make sure the strategy meets the needs of current users.

[0037] Specifically, collect user behavior data: User behavior data includes user browsing history, click behavior, purchase history, etc. These behavior data are represented as a time series ,in Indicates time: ; in: Indicates that the user is at time A specific behavior on the website, such as click, purchase, browse, etc.

[0038] LSTM (Long Short-Term Memory Network) is used to model user behavior data to predict future user behavior. LSTM Long Short-Term Memory Network generates a user behavior vector , indicating that at a point in time When the user's preference is expressed as: ; in: For time The behavior sequence of the time.

[0039] It is the user's historical behavior preference vector, which records the user's past behavior patterns.

[0040] After obtaining the user behavior representation vector After that, adjust the strategy weights in the marketing canvas model. Assume that the initial push strategy weights are , Dynamic adjustment is performed using the following formula: ; in: is the adjusted push strategy weight.

[0041] is the adjustment coefficient, which is used to control the sensitivity of behavioral data to strategy weight adjustment.

[0042] Update the marketing canvas matrix based on user preferences , ensuring that each strategy meets the needs of current users.

[0043] Step 30: Based on the adjusted marketing push strategy weights, optimize the push path according to user feedback and context information; Specifically include: Path selection is based on the objective function, and push paths are dynamically selected according to the Q-learning reinforcement learning algorithm; Based on each state Plan the optimal push path And adjust the push strategy through the Q value update formula; Based on the feedback and contextual information of the push path, the push strategy is dynamically adjusted to ensure that the path selection matches the current user's needs and behaviors.

[0044] The Q value update formula is as follows: ; in: is the Q value of taking an action in the state, that is, an estimate of the expected benefit of taking this action in the state; For the status Take action The immediate return is usually evaluated based on indicators such as click-through rate (CTR). For example, in a marketing strategy push system, if a marketing strategy (action ), the user clicks on the marketing strategy, then the instant return calculated based on the click-through rate will be updated ; is a discount factor, ranging from 0 to 1, used to control the impact of future returns. is close to 0, the agent pays more attention to the immediate reward, and if If it is close to 1, the agent pays more attention to future rewards. For example, in a long-term recommendation system, if you want the system to pay more attention to long-term user satisfaction and retention rate, It can be set larger; if you are more concerned about short-term click effects, Can be set smaller; For the next state All possible actions The maximum Q value in the equation reflects the agent's expectation of the future state. It updates the Q value of the current state by selecting the action with the maximum Q value, thereby guiding the agent to choose a better path.

[0045] Specific: Path optimization problem modeling: Push path optimization is modeled as a dynamic optimization problem, the goal of which is to maximize the trade-off between the relevance of the pushed content and the push cost; The objective function It is a trade-off between the relevance of the pushed content and the push cost. The specific formula is as follows: ; in: For products and customer groups The matching degree is calculated based on user preferences and product features; For the Products in time The weight of the moment; The push cost includes channel cost and timing cost; is the balance coefficient, which is used to control the trade-off between matching degree and cost.

[0046] According to the user's current status (such as browsing history, interest preferences, etc.), select the appropriate content push path (such as recommended news, videos, products, etc.), and update the Q value through user feedback (such as clicks, purchases, etc.), optimize the push strategy, and improve user satisfaction and platform revenue.

[0047] Based on the feedback from the push path, the push strategy is dynamically adjusted to ensure that the path selection matches the current user needs and behaviors.

[0048] Step 40: Automatically generate user personalized content based on the generation model and evaluate the push effect in real time; Specifically include: Generate personalized push content based on generative adversarial networks and customize content based on user behavior and preferences; Measure the attractiveness and conversion effect of content based on click-through rate calculation, and evaluate the push effect; The push content and strategy are adjusted in real time according to the results of the push effect evaluation.

[0049] Specifically, a generative adversarial network is used to generate personalized push content. , and customize content based on user behavior and preferences: ; in: To generate the network.

[0050] Represents the user behavior vector.

[0051] Generate diverse content for random noise; Push effect evaluation is to measure the attractiveness and conversion effect of content by calculating indicators such as click-through rate (CTR).

[0052] The effect evaluation formula is: ; in: To generate content and products Click-through rate; According to the push effect evaluation results, adjust the push content and strategy in real time to ensure the best marketing effect.

[0053] Specifically, by dynamically updating user portraits and combining them with a marketing canvas model, the present invention can accurately generate push content based on each user's real-time behavior and preferences, thereby achieving personalized marketing push.

[0054] With the help of reinforcement learning technology, the push path can be dynamically adjusted based on real-time feedback, thereby increasing user engagement and conversion rate and maximizing the push effect.

[0055] Through dynamic optimization of objective functions and content generation, it is possible to adapt to market changes and user needs in a short period of time and significantly improve the ROI (return on investment) of marketing activities.

[0056] Generative adversarial networks (GANs) can dynamically generate push content that meets the needs of each user. This generative method can increase the novelty of push content and reduce users' push fatigue.

[0057] The present invention can achieve real-time evaluation of push effects, and optimize push strategies in real time in combination with user behavior feedback, ensuring that push content and paths are always dynamically matched with user needs and behaviors.

[0058] Embodiment 2 How can traditional intelligent dynamic push based on marketing canvas extract real and effective preferences from massive user data and generate personalized and accurate push content in a timely manner based on their dynamic changes? Traditional methods mostly rely on simple behavior history or rule engines, and lack dynamic push capabilities based on deep understanding and real-time feedback.

[0059] In order to solve the above technical problems, based on the first embodiment, another technical solution adopted by the present application is: The intelligent dynamic push method based on the marketing canvas includes the following steps: Step 1: Build a marketing canvas model; Specific: Determine the key data of the marketing canvas: Products: involves product categories, features, discounts, etc.

[0060] Customers: By analyzing user portrait data, multiple customer groups (such as loyal customers, potential customers, etc.) can be divided.

[0061] Channels: Different push channels such as online, offline, and social platforms.

[0062] Communication strategy: personalized message push content, language style, push time, etc.

[0063] Modeling the Marketing Canvas via a multidimensional matrix: ; in: For the The feature matrix of a product.

[0064] For the The characteristic matrix of each customer group.

[0065] Each element Indicates the adaptability of marketing push strategies for specific products and customer groups.

[0066] Use external data (such as social media feedback, public opinion analysis, etc.) to enrich the data of the marketing canvas and enhance the real-time and personalization of the push strategy.

[0067] Specific: Establish a multi-data strategy model based on the marketing canvas, covering elements such as products, customers, channels, and communication strategies to support the generation of dynamic push notifications.

[0068] Step 2: Dynamically adjust the push strategy weight based on user behavior; Specific: Use deep learning models (such as GRU, LSTM) to analyze the user's behavior sequence and obtain the behavior representation vector of each user at a specific time point : ; in: : User behavior sequence, including click, purchase, browse and other behaviors.

[0069] : User historical preference vector, which represents the user's behavior trend in the past period of time.

[0070] The matrix corresponding to the marketing canvas dynamically adjusts the push strategy through the weight update mechanism.

[0071] Assume that the initial push strategy weight is , through real-time behavioral data To adjust the weights: ; in: : Initial strategy weight.

[0072] : In the The strategy weight at the moment.

[0073] : Adjustment coefficient, used to control the sensitivity of weight adjustment.

[0074] According to the adjusted strategy weights, the recommended push content is calculated: ; in: Push content for the final marketing generated.

[0075] For the Push content for a product or service.

[0076] Specifically, the weight of marketing push strategies is dynamically adjusted based on users’ real-time behaviors and historical preferences to ensure the personalization and accuracy of pushed content.

[0077] Step 3: Real-time push path optimization and dynamic adaptation; Specific: Model the push path optimization problem as a dynamic optimization problem and set the objective function : ; in: Relevance For products and customer groups The matching function of .

[0078] For the Products in The push weight at the moment.

[0079] For the cost of push channels and time, is the balance coefficient.

[0080] Optimize the push path in real time by adjusting the timing, frequency, content, and channel of push notifications. Use reinforcement learning (such as Q-learning) to dynamically adjust push notification strategies.

[0081] Setting the state space , indicating the user’s current status (for example, the user’s most recent purchase behavior, browsing time, etc.).

[0082] Use value function To calculate the expected return of the push path.

[0083] ; in: For the status Next, take action Immediate return after.

[0084] is the discount factor.

[0085] Dynamically adjust push strategies based on real-time feedback and context, and adopt a combination of multiple strategies to enhance push effects, such as multi-channel collaboration and cross-platform push.

[0086] Specifically, by combining real-time user feedback and contextual information, the push path is optimized so that the push content can be adjusted in a timely manner and differentiated push can be performed for different user groups.

[0087] Step 4: Adaptive content generation and push effect evaluation; Specific content: Using Generative Adversarial Networks (GANs) to generate personalized push content : ; in: To generate the network.

[0088] is the user behavior vector.

[0089] is random noise, used to generate diverse content.

[0090] Use reinforcement learning and AB testing to monitor push results in real time and adjust strategies: ; in: To generate click-through rates for content and products.

[0091] Specifically, in the process of intelligent push, generative models (such as GANs) are used to automatically generate personalized content and evaluate the push effect in real time.

[0092] Specifically, the solution integrates user behavior, contextual information, and real-time feedback to dynamically generate personalized push paths and adjust strategies in real time based on changes in user behavior.

[0093] By establishing a marketing canvas model, data from multiple sources such as products, customers, and channels are comprehensively considered to provide multi-angle support for the generation of marketing strategies.

[0094] Reinforcement learning and generative adversarial networks (GANs) are used to achieve adaptive content generation and dynamic optimization, significantly improving push accuracy and user engagement.

[0095] Through path optimization and multi-strategy collaboration, we ensure the optimal configuration of push paths based on multiple data, reduce marketing costs and improve conversion rates.

[0096] Embodiment 3 Analysis of existing technical problems: 1. The user portrait is single and difficult to dynamically adapt to market demand: The current intelligent push methods are usually based on static user portraits and cannot dynamically reflect real-time changes in user behavior and interests.

[0097] 2. The marketing canvas lacks intelligent optimization: As the core of the push strategy, the marketing canvas lacks intelligent adjustment capabilities and cannot be optimized in real time based on user feedback and changes in the external environment.

[0098] 3. The push content has a low match with user needs: Existing push systems mostly use simple rules or historical behavior analysis, which makes it difficult to accurately capture users' potential needs, resulting in poor relevance of push content.

[0099] 4. The algorithm model is opaque and lacks explainability: Most push algorithms are based on black box models, which makes it difficult for companies and users to understand push decisions, affecting user experience and trust.

[0100] 5. Lack of multi-channel collaborative push capabilities: The lack of a collaborative mechanism between multi-channel pushes (such as email, SMS, social media, etc.) may lead to content duplication or conflict.

[0101] In order to solve the above technical problems, based on the first embodiment, another technical solution adopted by the present application is: The intelligent dynamic push method based on the marketing canvas includes the following steps: Step 1: Build a dynamic multi-dimensional user portrait; Specific: 1.1. Collect user behavior data , including visit records, click streams, purchase history, social interactions, etc.

[0102] ; in, Indicates user In the A collection of characteristics of a behavior.

[0103] 1.2. Feature extraction and normalization: Features Including timestamp, visit frequency, length of stay, etc.

[0104] Use the normalization formula: ; 1.3. Multi-dimensional user portrait representation: through matrix Represents user portrait: ; in, is the feature vector, is the characteristic data.

[0105] 1.4. Real-time update mechanism: using sliding windows Dynamically update user profiles, with weights decaying over time: ; in, is the time decay coefficient.

[0106] Specifically, it dynamically captures changes in user interests, behavioral preferences, and needs, providing a data basis for precise push notifications.

[0107] Step 2: Build an intelligent marketing canvas Specifically, build a dynamically adjustable marketing canvas to adapt to changes in user behavior and external environment.

[0108] Specific: 2.1. Marketing goal breakdown: The overall goal Decompose into several sub-goals .

[0109] 2.2. Canvas structure design: Nodes represent user behavior points (such as browsing, purchasing).

[0110] The edge represents the behavior conversion path, and the weight is the conversion probability .

[0111] 2.3. Optimization model: Define the conversion revenue function : ; in, For conversion value.

[0112] 2.4. Reinforcement Learning Optimization: state : Current canvas status.

[0113] action : Adjust path or node weights.

[0114] award : Dynamic evaluation based on user conversion rate.

[0115] Optimizing policies using Deep Q Networks (DQN): ; in, is the discount factor.

[0116] Step 3: Push content generation and matching; Specifically, based on user portraits and marketing canvases, highly relevant push content is generated.

[0117] 3.1. Content generation: Generate personalized content using Generative Adversarial Networks (GANs): ; in, For the generator, is random noise.

[0118] 3.2 Content Matching: Calculate the relevance score between content and user profile : ; Step 4: Multi-channel intelligent collaborative push; Specifically, it realizes intelligent collaborative push across channels to avoid conflicts and redundancies.

[0119] Specific: 4.1. Channel Priority Sorting: Calculate priorities based on user preferences and historical data : ; in, is the weight.

[0120] 4.2 Push time optimization: Using Predictive Models Determine the best time to push: ; Step 5: Feedback mechanism and adaptive optimization; Specifically, we continuously optimize push strategies based on user feedback.

[0121] Specific: 5.1. Feedback collection: Collect data such as clicks, bounce rates, etc. .

[0122] 5.2 Model update: Use feedback to update the marketing canvas and push model parameters.

[0123] Specifically: Through the sliding window mechanism and reinforcement learning, user portraits and marketing paths are optimized in real time, and user behavior changes are dynamically adapted.

[0124] By combining generative models and relevance calculations, highly personalized push content can be achieved.

[0125] Resolve multi-channel conflicts through intelligent sequencing and time optimization.

[0126] Through the feedback mechanism, the push strategy is continuously optimized to gradually improve user experience and conversion rate.

[0127] Embodiment 4 Figure 5 is a structural diagram of an intelligent dynamic push device based on a marketing canvas according to an embodiment of the present application. Figure 5 As shown in the figure, the intelligent dynamic push device built based on the marketing canvas includes: The acquisition module is used to obtain the basic data of the marketing canvas and build a marketing canvas model based on the basic data; The adjustment module is used to adjust the weight of marketing push strategies based on users’ real-time behavior and historical preferences; The optimization path module is used to optimize the push path based on user feedback and context information based on the adjusted marketing push strategy weights; The evaluation push module is used to automatically generate user personalized content based on the generation model and evaluate the push effect in real time.

[0128] For other details about the technical solutions implemented by each module in the device of the above embodiment, please refer to the description of the intelligent dynamic push method based on the marketing canvas in the above embodiment 1, which will not be repeated here.

[0129] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For this method embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0130] Embodiment 5 Figure 6 is a schematic diagram of the structure of an electronic device according to an embodiment of the present application, such as Figure 6As shown, an electronic device includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus. The processor can call the logic instructions in the memory to execute the intelligent dynamic push method based on the marketing canvas, and the method includes: controlling the intelligent dynamic push device based on the marketing canvas to perform a test and receive the test results; displaying the test results on the display screen, such as the cold start current value, temperature data, etc.; providing a user interface to allow the user to set test parameters, view historical data, etc.; supporting data transmission with a computer-readable storage medium so as to export the test results for further analysis.

[0131] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0132] Embodiment 6 An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the intelligent dynamic push method based on the marketing canvas provided by the above-mentioned methods, and the method includes: controlling the intelligent dynamic push device based on the marketing canvas to perform a test, and receiving the test results; displaying the test results on a display screen, such as a cold start current value, temperature data, etc.; providing a user interface to allow the user to set test parameters, view historical data, etc.; supporting data transmission with a computer-readable storage medium to export the test results for further analysis.

[0133] The above-mentioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or mobile hard disk), media with built-in rewritable non-volatile memory (e.g., memory card) and media with built-in ROM (e.g., ROM box).

[0134] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0135] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0136] In the present disclosure, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. The block diagrams of the devices, apparatuses, equipment, and the method involved in the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and the method can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The word "such as" used here refers to the phrase "such as but not limited to", and can be used interchangeably with it.

[0137] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.

[0138] It should also be noted that in the method and process of the present disclosure, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present disclosure.

[0139] Various changes, substitutions, and modifications of the techniques described herein may be made without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and actions described above. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects described herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.

[0140] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0141] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. An intelligent dynamic push method based on the marketing canvas is characterized by: The method comprises the following steps: Obtain the basic data of the marketing canvas and build a marketing canvas model based on the basic data; Adjust marketing push strategy weights based on users’ real-time behavior and historical preferences; Based on the adjusted marketing push strategy weights, optimize the push path according to user feedback and contextual information; Automatically generate user-personalized content based on the generation model and evaluate the push effect in real time.

2. The intelligent dynamic push method based on the marketing canvas according to claim 1 is characterized in that: The marketing canvas model is constructed based on basic data, specifically including: Define basic data of the marketing canvas, including product data, customer data, channel data, and communication strategy data; Based on the marketing canvas model, product data, customer data, channel data and communication strategy data are collected to build a multidimensional matrix , the multidimensional matrix Including push strategy and weight; Integrate external data: Enhance your marketing canvas with data from social media and search engines.

3. The intelligent dynamic push method based on the marketing canvas according to claim 2 is characterized in that: The adjustment of marketing push strategy weights based on the user's real-time behavior and historical preferences specifically includes: Obtaining user behavior data, including user browsing history, click behavior, and purchase records; Modeling user behavior data based on a long short-term memory network to predict future user behavior, wherein the long short-term memory network generates a user behavior vector; Adjusting multidimensional matrix based on user behavior vector The weight of marketing push strategy in; Update the multidimensional matrix based on user preferences , make sure the strategy meets the needs of current users.

4. The intelligent dynamic push method based on the marketing canvas according to claim 1 is characterized in that: The optimization of the push path according to user feedback and context information specifically includes: Path selection is based on the objective function, and push paths are dynamically selected according to the Q-learning reinforcement learning algorithm; Based on each state Plan the optimal push path And adjust the push strategy through the Q value update formula; Based on the feedback and contextual information of the push path, the push strategy is dynamically adjusted to ensure that the path selection matches the current user's needs and behaviors.

5. The intelligent dynamic push method based on the marketing canvas according to claim 4 is characterized in that: The objective function is a trade-off between the relevance of the pushed content and the push cost. The specific formula is as follows: ； in: For products and customer groups The matching degree is calculated based on user preferences and product features; For the Products in time The weight of the moment; The push cost includes channel cost and timing cost; is the balance coefficient.

6. The intelligent dynamic push method based on the marketing canvas according to claim 4 is characterized in that: The Q value update formula is as follows: ; in: is the Q value of taking action in the state; For the status Take action Immediate returns; is the discount factor; For the next state All possible actions The maximum value of the medium Q value.

7. The intelligent dynamic push method based on marketing canvas construction according to claim 1 is characterized in that: The automatic generation of user personalized content based on the generation model and real-time evaluation of the push effect specifically include: Generate personalized push content based on generative adversarial networks and customize content based on user behavior and preferences; Measure the attractiveness and conversion effect of content based on click-through rate calculation, and evaluate the push effect; The push content and strategy are adjusted in real time according to the results of the push effect evaluation.

8. An intelligent dynamic push device based on a marketing canvas is applied to an intelligent dynamic push method based on a marketing canvas as claimed in any one of claims 1 to 7, characterized in that: include: The acquisition module is used to obtain the basic data of the marketing canvas and build a marketing canvas model based on the basic data; The adjustment module is used to adjust the weight of marketing push strategies based on users’ real-time behavior and historical preferences; The optimization path module is used to optimize the push path based on user feedback and context information based on the adjusted marketing push strategy weights; The evaluation push module is used to automatically generate user personalized content based on the generation model and evaluate the push effect in real time.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the intelligent dynamic push method based on the marketing canvas as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent dynamic push method based on the marketing canvas as described in any one of claims 1 to 7 are implemented.

Images