Rule Engine-Based Marketing Intelligence Triggering and Optimization Method, Device, Equipment, and Medium

Abstract

The present invention relates to a marketing intelligent triggering and optimization method, device, equipment and medium based on a rule engine, including obtaining marketing user information of a user in a marketing activity with a first activity content put out through a first put-out information; based on the marketing user information of the user in the marketing activity put out through the first put-out information, adjusting a triggering rule in a marketing activity triggering expression with the optimal action value as the goal based on a reinforcement learning algorithm to obtain a new marketing activity triggering expression; adjusting the first activity content of the marketing activity according to the new marketing activity triggering expression to obtain a second activity content of the marketing activity, and determining a second put-out information of the marketing activity after adjusting the activity content according to the new marketing activity triggering expression and the second activity content; and when the new marketing activity triggering expression is satisfied, re-putting out the marketing activity after adjusting the activity content according to the second put-out information.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular, to a method, apparatus, device, and medium for intelligent triggering and optimization of marketing based on a rule engine. Background Art

[0002] In the field of digital marketing, precise placement and optimization of marketing activities are the keys to improving conversion rates and user engagement. Traditional marketing methods often rely on manually set triggering rules and static campaign content, which are not only inefficient but also difficult to adapt to the rapidly changing market demands and user behaviors. With the development of big data and artificial intelligence technologies, data-driven marketing strategies have gradually become the mainstream. Among them, the rule engine, as a powerful tool, can dynamically trigger and execute marketing activities according to preset rules, while machine learning algorithms, especially reinforcement learning, provide the possibility for continuous optimization of these rules.

[0003] In the prior art, although there have been some attempts to optimize marketing activities by combining a rule engine and machine learning algorithms, these solutions often have limitations. For example, they may only rely on simple machine learning models to predict user behaviors, ignoring the complexity and diversity of the marketing activities themselves; or they may lack the ability to dynamically adjust the triggering rules and content of marketing activities, resulting in limited optimization effects. Therefore, it is particularly important to develop a method that can comprehensively consider the characteristics of marketing activities, user behavior patterns, and changes in the market environment to achieve intelligent triggering and optimization. Summary of the Invention

[0004] The object of the present invention is to provide a method, apparatus, device, and medium for intelligent triggering and optimization of marketing based on a rule engine, aiming to improve the accuracy and conversion rate of marketing activities and reduce the cost of manual intervention.

[0005] To achieve the above object, in the first aspect of the embodiments of the present disclosure, a method for intelligent triggering and optimization of marketing based on a rule engine is provided, including:

[0006] Obtain marketing user information of a user in a marketing activity with first campaign content delivered through first delivery information. The marketing activity with first campaign content delivered through first delivery information is based on a random forest algorithm. After screening out triggering rules from a rule library according to the first delivery information to construct a marketing activity trigger expression, the marketing activity is delivered to the user when the marketing activity trigger expression is satisfied;

[0007] Based on the marketing user information of the user in the marketing activity carried out through the first delivery information, and based on the reinforcement learning algorithm, with the optimal action value as the goal, adjust the triggering rule in the marketing activity triggering expression to obtain the new marketing activity triggering expression;

[0008] According to the new marketing activity triggering expression, adjust the first activity content of the marketing activity to obtain the second activity content of the marketing activity, and determine the second delivery information of the marketing activity after adjusting the activity content according to the new marketing activity triggering expression and the second activity content;

[0009] When the new marketing activity triggering expression is satisfied, re-deliver the marketing activity after adjusting the activity content according to the second delivery information.

[0010] In a possible implementation manner, the adjusting the triggering rule in the marketing activity triggering expression based on the reinforcement learning algorithm with the optimal action value as the goal according to the marketing user information of the user in the marketing activity carried out through the first delivery information to obtain the new marketing activity triggering expression includes:

[0011] Construct a virtual intelligent agent corresponding to the user, and loop to execute: use the marketing user information of the user in the marketing activity carried out through the first delivery information as the basis for selecting the movement path of the virtual intelligent agent, and determine the branch point of each movement from the forest tree corresponding to the marketing activity triggering expression;

[0012] For the probability that the virtual intelligent agent selects a movement path at any of the branch points, determine the action value assignment for moving to the next branch point along the movement path each time;

[0013] Calculate the sum value of the action values of all the branch points passed by when the virtual intelligent agent reaches a treetop of the forest tree each time;

[0014] When the sum value is less than the average value of the action values of the virtual intelligent agent corresponding to the user moving to all the treetops, sequentially select the target triggering rule from the rule library to replace the original triggering rule at the branch point, and re-execute the movement of the virtual intelligent agent until the difference between two adjacent average values is less than the preset difference threshold, so as to obtain the marketing activity triggering expression corresponding to the optimal action value.

[0015] In a possible implementation manner, the method for the tree tip to sequentially select a target trigger rule from the rule library to replace the original trigger rule at the branch point, and re-execute the movement of the virtual agent until the difference between the adjacent two averages is less than a preset difference threshold to obtain the marketing activity trigger expression corresponding to the optimal action value includes:

[0016] The tree tip sequentially selects a target trigger rule from the rule library to replace the original trigger rule at the branch point, and re-executes the movement of the virtual agent until the difference between the adjacent two averages is less than a preset difference threshold, and then stops replacing the original trigger rule at the branch point;

[0017] Determine the maximum action value that the virtual agent corresponding to the user can obtain when reaching a tree tip of the forest tree when stopping replacing the original trigger rule at the branch point;

[0018] Take the branch point on the movement path corresponding to the maximum action value of the virtual agent corresponding to the user as the locked branch point. When replacing the original trigger rule at the branch point by other mobile agents corresponding to the users, do not replace the trigger rule corresponding to the locked branch point;

[0019] Perform virtual agent movement and replace the original trigger rule at the branch point for all virtual agents corresponding to the users until all users are traversed, so as to obtain the marketing activity trigger expression corresponding to the optimal action value.

[0020] In a possible implementation manner, the method for the tree tip to sequentially select a target trigger rule from the rule library to replace the original trigger rule at the branch point, and re-execute the movement of the virtual agent until the difference between the adjacent two averages is less than a preset difference threshold, and then stop replacing the original trigger rule at the branch point includes:

[0021] Use the tree tip as the target branch point to execute replacing the original trigger rule at the target branch point, and re-execute the movement of the virtual agent reaching the target branch point to obtain the action value assignment for moving to the target branch point according to the movement path;

[0022] Determine whether the difference between the averages corresponding to the previous movement and the current movement is less than the preset difference threshold;

[0023] When the difference is less than the preset difference threshold, stop replacing the original trigger rule of the target branch point, and use the parent node of the target branch point in the forest tree as the next target branch point, and re - execute the movement of the virtual agent to reach the target branch point, and so on, until the last branch point on this movement path has been used as the target branch point, re - execute the movement of the virtual agent to reach the target branch point, and stop replacing the original trigger rule of the branch point.

[0024] In a possible implementation manner, the method further includes:

[0025] When the difference is greater than or equal to the preset difference threshold, select a target trigger rule from the rule library again to replace the original trigger rule of the target branch point, and re - execute the movement of the virtual agent until the difference between two adjacent averages is less than the preset difference threshold.

[0026] In a possible implementation manner, adjusting the first activity content of the marketing activity according to the new marketing activity trigger expression to obtain the second activity content of the marketing activity, and determining the second delivery information of the marketing activity after adjusting the activity content according to the new marketing activity trigger expression and the second activity content includes:

[0027] Adjust the first activity content of the marketing activity according to the trigger conditions and corresponding trigger rules in the new marketing activity trigger expression to obtain the second activity content of the marketing activity, where adjusting the first activity content of the marketing activity includes at least one of the following: changing the preferential intensity of the activity, adjusting the time range of the activity, adding or deleting activity goods or services, modifying the promotional copy of the activity;

[0028] According to the new marketing activity trigger expression and the second activity content, re - determine the second delivery information of the adjusted marketing activity, where the second delivery information includes at least one of the following: delivery time, delivery channel, delivery frequency, and the user group for delivery.

[0029] In a possible implementation manner, the marketing user information includes at least one of the following:

[0030] The frequency of participating in the marketing activity, the duration of participating in the marketing activity, the time period of participating in the marketing activity, the basic information of the users participating in the marketing activity.

[0031] In a second aspect of the embodiments of the present disclosure, a marketing intelligent trigger and optimization device based on a rule engine is provided, and the device includes:

[0032] An acquisition module, configured to acquire marketing user information of users in a marketing activity with a first activity content launched through a first delivery message. The marketing activity with the first activity content launched through the first delivery message is based on a random forest algorithm. After screening out trigger rules from a rule base according to the first delivery message to construct a marketing activity trigger expression, it is launched to users when the marketing activity trigger expression is satisfied;

[0033] An adjustment module, configured to, based on a reinforcement learning algorithm and aiming at the optimal action value, adjust the trigger rules in the marketing activity trigger expression according to the marketing user information of users in the marketing activity launched through the first delivery message, so as to obtain a new marketing activity trigger expression;

[0034] A determination module, configured to adjust the first activity content of the marketing activity according to the new marketing activity trigger expression to obtain a second activity content of the marketing activity, and determine a second delivery message of the marketing activity after adjusting the activity content according to the new marketing activity trigger expression and the second activity content;

[0035] A delivery module, configured to, when the new marketing activity trigger expression is satisfied, redeliver the marketing activity after adjusting the activity content according to the second delivery message.

[0036] In a third aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the method according to any one of the first aspects are implemented.

[0037] In a fourth aspect of the embodiments of the present disclosure, there is provided an electronic device, including:

[0038] A memory, on which a computer program is stored;

[0039] A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of the first aspects.

[0040] The present invention provides a marketing intelligent trigger and optimization method, device, equipment and medium based on a rule engine. Compared with the prior art, the following beneficial effects are achieved:

[0041] The random forest algorithm is used to deeply analyze the first delivery message, and the trigger rules most matching the current marketing activity are intelligently screened out from the rule base to construct an initial marketing activity trigger expression. This not only improves the accuracy and pertinence of the trigger rules, but also ensures that the marketing activity can be launched for the most suitable user group.

[0042] According to the marketing user information of users in marketing activities, the triggering rules are dynamically adjusted using reinforcement learning algorithms. Through continuous trial and error and learning, with the optimal action value as the goal, the triggering rules are gradually optimized, enabling marketing activities to more precisely trigger potential interested users, thereby improving the delivery efficiency and conversion rate.

[0043] Based on the optimization of the triggering rules, the content of the marketing activity is further adjusted according to the new triggering expression to generate the second activity content, and the corresponding second delivery information is determined. This not only enhances the flexibility and personalization of the marketing activity but also enables the marketing activity to better adapt to changes in market demand and user preferences. Finally, under the condition of meeting the new triggering expression, the adjusted marketing activity is re-delivered according to the second delivery information, achieving comprehensive optimization from triggering to content.

[0044] In summary, by integrating the rule engine and reinforcement learning algorithms, an efficient and intelligent marketing triggering and optimization method is achieved. It not only improves the accuracy and conversion rate of marketing activities but also reduces the cost of manual intervention, bringing significant technological innovation and economic benefits to the digital marketing field.

[0045] Other features and advantages of the present disclosure will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are used to provide a further understanding of the present disclosure and form a part of the specification, and are used to explain the present disclosure together with the following specific implementation, but do not constitute a limitation to the present disclosure. In the drawings:

[0047] Figure 1 is a flowchart of a marketing intelligent triggering and optimization method based on a rule engine shown according to an embodiment of the specification.

[0048] Figure 2 is a block diagram of a marketing intelligent triggering and optimization device based on a rule engine shown according to an embodiment of the specification.

[0049] Figure 3 is a block diagram of another marketing intelligent triggering and optimization device based on a rule engine shown according to an embodiment of the specification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] The specific embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for the purpose of illustrating and explaining the present disclosure, and are not intended to limit the present disclosure.

[0052] The present disclosure provides a marketing intelligent triggering and optimization method based on a rule engine. Figure 1 It is a flowchart of a marketing intelligent triggering and optimization method based on a rule engine shown according to an embodiment. The method includes:

[0053] In step S11, obtain marketing user information of users in a marketing activity with a first activity content that is launched through a first delivery message. The marketing activity with the first activity content launched through the first delivery message is based on a random forest algorithm. After screening out trigger rules from a rule library according to the first delivery message to construct a marketing activity trigger expression, it is launched to users when the marketing activity trigger expression is satisfied.

[0054] Among them, the marketing user information is data generated by users during the participation in a specific marketing activity, including but not limited to user behavior data, transaction data, preference data, etc. The first delivery message is a series of parameters and settings used when initially launching a marketing activity, such as delivery time, delivery channel, target user group, etc. The marketing activity trigger expression is a logical expression constructed based on preset rules and user data, and is used to determine whether to launch a marketing activity to a specific user.

[0055] In an embodiment of the present disclosure, extract key features from the first delivery message as the input of the random forest algorithm. These features should be able to reflect the characteristics of the target user group and the requirements of the marketing activity. Perform preprocessing on the features, such as missing value filling, outlier processing, feature scaling, etc., to improve the performance and accuracy of the algorithm. Use the extracted features and the trigger rules in the rule library to train the random forest model. During the training process, the model will learn how to construct a marketing activity trigger expression based on user features and the rules in the rule library. Through multiple iterations and cross-validation, optimize the parameters and performance of the model to ensure that the model can accurately predict which users meet the trigger conditions.

[0056] Construct a marketing activity trigger expression. Based on the trained random forest model, construct a marketing activity trigger expression according to user features and the rules in the rule library. This expression is a logical judgment statement used to determine whether a user meets the trigger conditions of the marketing activity. The trigger expression may contain multiple conditions and sub-conditions, which are combined through logical operations (such as AND, OR, NOT) to form a complete trigger logic.

[0057] Collect user data in real time and make judgments based on the trigger expressions of marketing campaigns. When the real-time data of the user meets the trigger expression, it is considered that the user meets the trigger conditions of the marketing campaign. Under the condition of meeting the trigger conditions, according to the preset marketing strategies and delivery information (such as delivery time, delivery channels, etc.), a marketing campaign with the first campaign content is delivered to the user. During the delivery process, continuously monitor user feedback and the effectiveness of the marketing campaign to adjust the marketing strategy and optimize the trigger expression in a timely manner.

[0058] In the embodiments of the present disclosure, these information are analyzed using the random forest algorithm, and the trigger rules that are most matched with the current marketing campaign are screened out from the preset rule library. These rules may involve multiple dimensions such as user portraits, behavior patterns, geographical locations, etc. Based on the screened trigger rules, a marketing campaign trigger expression is constructed. When the real-time data of the user meets this expression, a marketing campaign with the first campaign content is delivered to the user. Furthermore, marketing user information under the first delivery information is collected. This information includes but is not limited to the user's click behavior, browsing history, purchase records, etc.

[0059] For example, assume that a certain bank plans to push a new loan activity to potential users. By using the random forest algorithm to analyze historical user data, it is found that users who have browsed relevant loans are more likely to be interested in this loan activity. Therefore, the system screens out these users as the target group and constructs a marketing campaign trigger expression, such as "the user has browsed the bank loan page in the past 30 days and has not taken a loan". When users who meet this condition go online, the system pushes this new loan activity to them. Collect the first delivery information, such as the delivery time is on weekends, and the delivery channels are social media and email.

[0060] In step S12, based on the marketing user information of the user in the marketing campaign delivered through the first delivery information, and based on the reinforcement learning algorithm, with the optimal action value as the goal, the trigger rules in the marketing campaign trigger expression are adjusted to obtain a new marketing campaign trigger expression;

[0061] Among them, the reinforcement learning algorithm can learn the optimal strategy through the interaction between the agent and the environment to maximize the cumulative reward. The action value in reinforcement learning refers to the expected return that can be obtained after executing an action.

[0062] In the embodiments of the present disclosure, based on the feedback of the user under the first delivery information (i.e., marketing user information), the reinforcement learning algorithm is used to adjust the triggering rules in the marketing activity triggering expression. The reinforcement learning algorithm aims to maximize the cumulative reward through continuous trial and error and learning, that is, to improve the conversion rate or user engagement of the marketing activity, and gradually optimize the triggering rules. By simulating the user response situations under different triggering rules and adjusting the rule weights or introducing new rules according to the actual feedback, a new marketing activity triggering expression is obtained.

[0063] Continuing with the above example, assume that the conversion rate of the pushed promotional activity is lower than expected. Using the reinforcement learning algorithm to analyze the actual feedback data of the user, it is found that although the user is interested in loans, they are not sensitive enough to the interest rate of the loan activity. Therefore, the system adjusts the triggering rule, changing "the user has browsed the bank loan-related pages in the past 30 days and has not taken out a loan" to "the user has browsed the bank loan-related pages in the past 30 days and has not taken out a loan and is highly sensitive to the interest rate". At the same time, other triggering conditions are also considered, such as the user's repayment information, overdue repayment, etc., to further optimize the triggering rule.

[0064] In step S13, according to the new marketing activity triggering expression, the first activity content of the marketing activity is adjusted to obtain the second activity content of the marketing activity, and according to the new marketing activity triggering expression and the second activity content, the second delivery information of the marketing activity after adjusting the activity content is determined;

[0065] Among them, the activity content adjustment is to modify and optimize the specific content of the marketing activity (such as bank type, loan type, etc.) according to the user portrait and needs after the triggering rule adjustment. The second delivery information is a series of parameters and settings used when the marketing activity is redelivered after adjusting the activity content.

[0066] In the embodiments of the present disclosure, the first activity content is adjusted according to the new marketing activity triggering expression. This adjustment may involve multiple aspects such as the adjustment of the preferential intensity, the optimization of product recommendations, and the modification of the activity copywriting. At the same time, the system also determines the second delivery information according to the new triggering expression and the second activity content. These information may include the adjusted delivery time, delivery channel, target user group, etc. By comprehensively considering the changes in the triggering rules and user needs, the system ensures that the adjusted marketing activity can reach the target users more accurately and improve the conversion rate.

[0067] Continuing with the above example, assume that the triggering rule is adjusted and it is found that users are highly sensitive to loan interest rates. The content of the first activity is adjusted by modifying the original bank interest rate from "annual interest rate of 3.75%" to "annual interest rate of 3.55%", and some loan repayment activities that users may be interested in are added. At the same time, the system also determines the second delivery information according to the new triggering expression, such as adjusting the delivery time to the prime time in the evening and increasing the delivery proportion on social media platforms, etc.

[0068] In step S14, when the new marketing activity triggering expression is satisfied, the marketing activity with adjusted activity content is redelivered according to the second delivery information.

[0069] In the embodiments of the present disclosure, when the new marketing activity triggering expression is satisfied, the marketing activity with adjusted activity content is redelivered. During the redelivery process, user feedback data, that is, marketing user information, is continuously collected and analyzed, and the effect of the marketing activity is evaluated. By comparing the changes in indicators such as conversion rate and user engagement before and after redelivery, the system can determine whether the adjusted marketing activity has achieved the expected effect. If the effect is still not ideal, it can return to step S12 for further adjustment of the triggering rule and content optimization.

[0070] Continuing with the above example, the adjusted marketing activity is redelivered according to the new triggering expression and the second delivery information. During the delivery process, user feedback data is continuously collected and analyzed, and indicators such as conversion rate and user engagement are monitored and evaluated. Assume that the conversion rate after redelivery has increased significantly and the user engagement has also increased, indicating that the adjusted marketing activity has achieved the expected effect. At this time, the system can consider that the process of triggering and optimizing this marketing activity has ended, and use the relevant experience and data for future marketing activity planning and optimization.

[0071] The above technical solution uses the random forest algorithm to deeply analyze the first delivery information, intelligently selects the triggering rule that best matches the current marketing activity from the rule base, and constructs an initial marketing activity triggering expression. It not only improves the accuracy and pertinence of the triggering rule, but also ensures that the marketing activity can be delivered to the most suitable user group.

[0072] According to the marketing user information of users in the marketing activity, the triggering rule is dynamically adjusted using the reinforcement learning algorithm. Through continuous trial and error and learning, with the optimal action value as the goal, the triggering rule is gradually optimized, so that the marketing activity can more accurately trigger potential interested users, thereby improving the delivery efficiency and conversion rate.

[0073] On the basis of optimizing the trigger rules, further adjust the content of the marketing activity according to the new trigger expression to generate the second activity content, and determine the corresponding second delivery information, which not only enhances the flexibility and personalization of the marketing activity, but also enables the marketing activity to better adapt to the changes in market demand and user preferences. Finally, when the new trigger expression is satisfied, re-deliver the adjusted marketing activity according to the second delivery information, achieving a comprehensive optimization from trigger to content.

[0074] In summary, by integrating the rule engine and the reinforcement learning algorithm, an efficient and intelligent marketing trigger and optimization method is realized. It not only improves the accuracy and conversion rate of marketing activities, but also reduces the cost of manual intervention, bringing significant technological innovation and economic benefits to the digital marketing field.

[0075] In a possible implementation manner, in step S12, based on the marketing user information of the user in the marketing activity delivered by the first delivery information, and based on the reinforcement learning algorithm, with the optimal action value as the goal, adjust the trigger rule in the marketing activity trigger expression to obtain the new marketing activity trigger expression, including:

[0076] In step S121, construct the virtual agent corresponding to the user, and loop to execute: use the marketing user information of the user in the marketing activity delivered by the first delivery information as the basis for selecting the movement path of the virtual agent, and determine the branch point of each movement from the decision tree corresponding to the marketing activity trigger expression;

[0077] Among them, a virtual agent (Virtual Agent) is an entity that represents performing actions and receiving environmental feedback in reinforcement learning. In this example, the virtual agent represents the user and is used to select a path in the decision tree (or forest tree, hereinafter uniformly referred to as the decision tree for simplicity of understanding) of the marketing activity trigger expression.

[0078] In the embodiment of the present disclosure, a virtual agent is constructed for each user, and this agent selects a path in the decision tree according to the marketing user information (such as browsing records, browsing duration, etc.) of the user under the first delivery information. The movement path of the agent reflects the process by which the user meets the trigger conditions.

[0079] Illustrate with an example: Suppose there are multiple branch points in the decision tree, and each branch point represents a trigger rule (such as "the user has browsed the pages of bank loans in the past 30 days, has not applied for a loan, and is highly sensitive to interest rates"). The virtual agent selects the corresponding branch to move according to the actual behavior data of the user (such as the user has indeed not applied for a loan in the past 30 days).

[0080] In step S122, for the probability that the virtual agent selects a movement path at any of the branch points, determine the action value assignment for each time of moving to the next branch point along the movement path.

[0081] Among them, the action value is, in reinforcement learning, the expected value of the cumulative reward obtained after performing a certain action. In this example, the action value is used to evaluate the effect of the trigger rule in the marketing activity.

[0082] In the embodiments of the present disclosure, for each branch point in the decision tree, calculate the probability that the virtual agent selects a path at this point, and assign an action value according to the selected path. This assignment reflects the effect of the trigger rule in the current user group.

[0083] For example: Suppose at a certain branch point, the virtual agent has an 80% probability of selecting "yes" (i.e., the user meets the trigger rule), then the action value of this path may be relatively high. On the contrary, if the agent has a 20% probability of selecting "no", then the action value of this path may be relatively low.

[0084] In step S123, calculate the sum value of the action values of all the branch points passed by the virtual agent each time it reaches a leaf node of the forest tree.

[0085] Among them, the leaf node is the leaf node in the decision tree, representing the final decision result or output value. In this example, the leaf node represents the final judgment result of the marketing activity trigger expression.

[0086] In the embodiments of the present disclosure, when the virtual agent reaches the leaf node of the decision tree, calculate the sum value of the action values of all the branch points it passes through. This sum value reflects the effect of the entire trigger expression in the current user group.

[0087] For example: Suppose the virtual agent starts from the root node, passes through multiple branch points in sequence, and finally reaches a leaf node. During this process, the action values of each branch point are accumulated to form the final sum value.

[0088] In step S124, when the sum value is less than the average value of the action values of the virtual agent corresponding to the user moving to all leaf nodes, sequentially select target trigger rules from the rule library to replace the original trigger rules at the branch points, and re - execute the movement of the virtual agent until the difference between two adjacent average values is less than the preset difference threshold, so as to obtain the marketing activity trigger expression corresponding to the optimal action value.

[0089] In the embodiments of the present disclosure, if the calculated action value sum is less than the average of the action values of the virtual agent moving to all the tree tops, it indicates that the effect of the current trigger expression is not good. At this time, starting from the tree top, target trigger rules are sequentially selected from the rule base to replace the original trigger rules, and the movement of the virtual agent is re-executed. This process will be iterated until the difference between two adjacent averages is less than the preset difference threshold, and the obtained trigger expression is the optimal one at this time.

[0090] For example: Assume that in the first iteration, the calculated action value sum is low. At this time, starting from the tree top, it is possible to check whether the trigger rules at each branch point are appropriate. If it is found that the effect of a certain trigger rule is not good (such as the rule "the user has browsed pages related to bank loans in the past 30 days, has not taken out a loan, and is highly sensitive to interest rates" is not common among the user group), then a more appropriate rule can be selected from the rule base to replace it (such as "the user has browsed pages related to bank loans in the past 30 days, has not taken out a loan, and has a low requirement for the repayment period"). Then the movement process of the virtual agent is re-executed, and the new action value sum is calculated. This process will be repeated continuously until the optimal trigger expression is found.

[0091] In this way, the trigger rules in the marketing activity trigger expression can be adjusted and optimized by using the reinforcement learning algorithm, so as to improve the conversion rate and user engagement of the marketing activity.

[0092] In a possible implementation manner, in step S124, the process of sequentially selecting target trigger rules from the rule base to replace the original trigger rules at the branch point, and re-executing the movement of the virtual agent until the difference between two adjacent averages is less than the preset difference threshold to obtain the marketing activity trigger expression corresponding to the optimal action value includes:

[0093] In step S1241, target trigger rules are sequentially selected from the rule base to replace the original trigger rules at the branch point, and the movement of the virtual agent is re-executed until the difference between two adjacent averages is less than the preset difference threshold, and then stop replacing the original trigger rules at the branch point;

[0094] In the embodiments of the present disclosure, starting from the tree top of the decision tree, the trigger rules at each branch point are sequentially checked, and target trigger rules are selected from the rule base for replacement. Then the movement of the virtual agent is re-executed, and the new action value and average are calculated. If the difference between two adjacent averages is less than the preset difference threshold, then stop replacing the trigger rules at this branch point.

[0095] For example: Suppose at a certain tip of the decision tree, the action value and average value of the corresponding user group under the original trigger rule are relatively low. At this time, a new trigger rule can be selected from the rule library to replace the original rule, and the action value and average value are recalculated. If the difference between the new average value and the old average value is less than the difference threshold (e.g., 0.01), it is considered that the trigger rule at this branch point has been optimized enough and no further replacement is needed.

[0096] In step S1242, it is determined that when the virtual intelligent agent corresponding to the user stops replacing the original trigger rule at the branch point, the maximum action value that the virtual intelligent agent corresponding to the user can obtain when reaching a tip of the forest tree.

[0097] In the embodiment of the present disclosure, after stopping replacing the trigger rule, the maximum action value that the virtual intelligent agent can obtain when reaching the tip of the decision tree is determined. This maximum action value represents the optimal effect of the current trigger expression in the user group.

[0098] For example: Suppose after stopping replacing the trigger rule, the virtual intelligent agent obtains a relatively high action value (e.g., 0.8) when reaching a certain tip. This value may be gradually optimized during multiple iterations and represents the optimal effect of the current trigger expression in the user group.

[0099] In step S1243, the branch points on the movement path corresponding to the maximum action value of the virtual intelligent agent corresponding to the user are used as locked branch points. When the movement intelligent agents corresponding to other users execute to replace the original trigger rule at the branch point, the trigger rule corresponding to the locked branch point is not replaced.

[0100] In the embodiment of the present disclosure, the branch points on the movement path corresponding to the maximum action value of the virtual intelligent agent are used as locked branch points. In the subsequent optimization process, the trigger rules of these locked branch points will no longer be replaced.

[0101] For example: Suppose the virtual intelligent agent obtains the maximum action value when reaching a certain tip, and the branch points A, B, and C on this path all contribute significantly to the final action value. At this time, A, B, and C can be used as locked branch points, and their trigger rules will not be replaced in the subsequent optimization process.

[0102] In step S1244, virtual intelligent agent movement is performed on the virtual intelligent agents corresponding to all users, and the original trigger rule at the branch point is replaced until all users are traversed, and the marketing activity trigger expression corresponding to the optimal action value is obtained.

[0103] In the embodiments of the present disclosure, the virtual agents of all users are traversed, and operations of moving and replacing triggering rules are performed. During the traversal, the rule of locking branch points is followed, and the triggering rules of the locked branch points are not replaced. Until all users have completed the traversal and the marketing activity triggering expression corresponding to the optimal action value is obtained.

[0104] For example: Suppose there are 1000 users, and each user has a corresponding virtual agent. During the traversal, operations of moving and replacing triggering rules are performed on the virtual agents of each user. At the same time, the rule of locking branch points is followed, and the triggering rules of the already locked branch points are not replaced. After multiple iterations and traversals, a marketing activity triggering expression that performs well in all user groups is finally obtained.

[0105] Through the above steps, the marketing activity triggering expression can be optimized based on the reinforcement learning algorithm to obtain the optimal combination of triggering rules in the user group. This method not only improves the conversion rate and user participation of marketing activities, but also enhances the pertinence and personalization of marketing activities.

[0106] In a possible implementation manner, in step S1241, the process of successively selecting a target triggering rule from the rule base to replace the original triggering rule of the branch point and re-executing the movement of the virtual agent until the difference between the two adjacent averages is less than a preset difference threshold and then stopping replacing the original triggering rule of the branch point includes:

[0107] In step S12411, using the tree tip as the target branch point, execute the replacement of the original triggering rule of the target branch point, and re-execute the movement of the virtual agent reaching the target branch point to obtain the action value assignment of moving to this target branch point according to the movement path.

[0108] In the embodiments of the present disclosure, starting from the tree tip of the decision tree, the currently considered branch point is used as the target branch point. A new triggering rule is selected from the rule base to replace the original triggering rule of the target branch point. Then, the movement of the virtual agent is re-executed, and the action value assignment of moving to this target branch point according to the new movement path is calculated.

[0109] For example: Suppose the action value of a user group corresponding to a certain tree tip of the decision tree is low under the original triggering rule. At this time, select the branch point corresponding to this tree tip as the target branch point, and select a new triggering rule from the rule base for replacement. Then, re-execute the movement of the virtual agent and calculate the new action value assignment.

[0110] In step S12412, determine whether the difference between the average value corresponding to the previous move and the current move is less than the preset difference threshold;

[0111] In the embodiments of the present disclosure, after replacing the trigger rule and re - executing the movement of the virtual agent, calculate the difference between the average values corresponding to the previous move and the current move. Determine whether this difference is less than the preset difference threshold.

[0112] For example: Suppose that after replacing the trigger rule, the difference between the average value corresponding to the new action value assignment and the old average value is 0.02, while the preset difference threshold is 0.01. At this time, it is determined that the difference is greater than the preset difference threshold, and iterative optimization needs to continue.

[0113] In step S12413, when it is less than the preset difference threshold, stop replacing the original trigger rule of the target branch point, and use the parent node of the target branch point in the forest tree as the next target branch point, and re - execute the movement of the virtual agent to reach the target branch point, and so on, until the last branch point on this movement path has been used as the target branch point, re - execute the movement of the virtual agent to reach the target branch point, and stop replacing the original trigger rule of the branch point.

[0114] In the embodiments of the present disclosure, if the average value difference is less than the preset difference threshold, stop replacing the trigger rule of the current target branch point. Use the parent node of the current target branch point as the next target branch point, and repeat steps S12411 and S12412 until the last branch point (i.e., the root node) on this movement path has been used as the target branch point, and stop replacing the trigger rule.

[0115] For example: Suppose that during the iteration process, the average value difference of a certain branch point is less than the preset difference threshold, then stop replacing the trigger rule of this branch point. Then, use the parent node of this branch point as the new target branch point and continue the iteration process until the root node is also considered and the replacement of the trigger rule stops.

[0116] In a possible implementation manner, the method further includes:

[0117] In step S12414, when it is greater than or equal to the preset difference threshold, select a target trigger rule from the rule library again to replace the original trigger rule of the target branch point, and re - execute the movement of the virtual agent until the difference between the two adjacent average values is less than the preset difference threshold.

[0118] In the embodiments of the present disclosure, if the average value difference is greater than or equal to the preset difference threshold, then a target trigger rule is selected again from the rule library to replace the original trigger rule at the target branch point, and the movement of the virtual agent is executed again. This process will be continuously repeated until the difference between two adjacent average values is less than the preset difference threshold.

[0119] For example: Assume that after the first iteration, the average value difference is still greater than the preset difference threshold. At this time, a new trigger rule is selected from the rule library to replace the original trigger rule at the target branch point, and the movement of the virtual agent is executed again. Then, the average value difference is calculated again, and it is determined whether to continue the iteration. This process will be continuously repeated until the stop condition is met.

[0120] Through the above steps, the trigger rules in the marketing activity trigger expression can be iteratively optimized based on the reinforcement learning algorithm. This method not only improves the conversion rate and user participation of the marketing activity, but also enhances the pertinence and personalization of the marketing activity by continuously iterating to approach the optimal solution.

[0121] In a possible implementation manner, in step S13, adjusting the first activity content of the marketing activity according to the new marketing activity trigger expression to obtain the second activity content of the marketing activity, and determining the second delivery information of the marketing activity after adjusting the activity content according to the new marketing activity trigger expression and the second activity content includes:

[0122] In step S131, adjusting the first activity content of the marketing activity according to the trigger conditions and corresponding trigger rules in the new marketing activity trigger expression to obtain the second activity content of the marketing activity, where adjusting the first activity content of the marketing activity includes at least one of the following: changing the preferential intensity of the activity, adjusting the time range of the activity, adding or deleting activity goods or services, modifying the promotional copy of the activity;

[0123] In the embodiments of the present disclosure, first, the new marketing activity trigger expression is parsed to identify the trigger conditions and corresponding trigger rules therein. Then, according to these conditions and rules, the original first activity content is adjusted. The adjustment may involve the following aspects:

[0124] Changing the preferential intensity: According to the trigger conditions, such as the user's historical purchase behavior, user level, or specific festivals, etc., adjust the preferential amplitude of the activity to better attract target users.

[0125] Adjusting the activity time range: According to the trigger rules, such as the higher user activity within a specific time period, adjust the start and end times of the activity to maximize user participation.

[0126] Adding or deleting active products or services: Adjust the list of products or services involved in the activity according to user preferences, market demand, or inventory status.

[0127] Modifying promotional copy: Adjust the style, content, and attractiveness of the promotional copy according to the trigger conditions and the characteristics of the target user group to increase the exposure rate and participation rate of the activity.

[0128] Example: Suppose the trigger expression for the new loan activity defines that when the user's historical loan frequency reaches a certain number of times and all loans have been repaid, an additional interest rate reduction offer is provided. After the system recognizes this trigger condition, it automatically reduces the interest rate in the first activity content from 3.55% to 3.25% to attract these users with high loan frequencies.

[0129] In step S132, according to the new marketing activity trigger expression and the second activity content, re-determine the second delivery information of the adjusted marketing activity, where the second delivery information includes at least one of the following: delivery time, delivery channel, delivery frequency, and the user group to be delivered.

[0130] In the embodiments of the present disclosure, according to the new marketing activity trigger expression and the second activity content, re-determine the second delivery information of the adjusted marketing activity. This step aims to ensure that the activity can reach the most likely participating user group at the most appropriate time, through the most appropriate channel, and with the most appropriate frequency.

[0131] Delivery time: Select the best delivery time point or time period according to the user activity and the time distribution of the trigger conditions.

[0132] Delivery channel: Select the appropriate delivery platform or medium according to the preferences and habits of the target user group.

[0133] Delivery frequency: Adjust the number of deliveries according to user feedback, activity effects, and market changes to ensure the continuous exposure and participation rate of the activity.

[0134] User group to be delivered: A set of target users screened according to user attributes, behavior data, or trigger conditions to ensure the precise delivery of the activity.

[0135] Example: Suppose the trigger expression for the new marketing activity defines that when a user has visited a specific product page in the past month, send an activity notification via email. After the system recognizes this trigger condition, it automatically screens out the user group that meets this condition and sends the activity notification via the email channel at an appropriate time point. At the same time, according to user feedback and activity effects, the system can dynamically adjust the delivery frequency and delivery time to optimize the activity effects.

[0136] In a possible implementation manner, the marketing user information includes at least one of the following:

[0137] The activity participation frequency of the marketing activity, the activity participation duration of the marketing activity, the participation time period of the marketing activity, and the basic user information of the user participating in the marketing activity.

[0138] In the embodiments of the present disclosure, the activity participation frequency is the number of times a user participates in a specific marketing activity during a specific period. This indicator reflects the user's interest and activity level in the activity and is an important basis for evaluating the activity's attractiveness and user stickiness.

[0139] The activity participation duration is the time spent by a user in a single participation in a marketing activity. This data helps to understand the user's degree of investment in the activity and the content attractiveness, and has guiding significance for optimizing the activity design and improving the user experience.

[0140] The participation time period is the time distribution of a user's participation in a marketing activity, including specific dates, times, or time periods. By analyzing the participation time period, the peak and trough periods of user activity can be identified, providing data support for precise advertising placement and activity rhythm arrangement.

[0141] The basic user information is the personal basic information provided by a user during registration, participation in activities, or interaction, such as age, gender, region, occupation, hobbies, etc. These information helps to construct a user portrait, achieve personalized marketing, and improve the targeting and conversion rate of activities. The basic user information is indispensable basic data for formulating marketing strategies and evaluating activity effects.

[0142] The embodiments of the present disclosure further provide a marketing intelligent triggering and optimization device based on a rule engine. Refer to Figure 2 as shown, the device includes:

[0143] An acquisition module 210, configured to acquire marketing user information of a user in a marketing activity with a first activity content that is launched through a first placement information. The marketing activity with the first activity content launched through the first placement information is based on a random forest algorithm. After screening out a trigger rule from a rule library according to the first placement information to construct a marketing activity trigger expression, it is launched to the user when the marketing activity trigger expression is satisfied;

[0144] An adjustment module 220, configured to adjust the trigger rule in the marketing activity trigger expression based on a reinforcement learning algorithm with the optimal action value as the goal according to the marketing user information of the user in the marketing activity launched through the first placement information, so as to obtain a new marketing activity trigger expression;

[0145] A determination module 230, configured to adjust the first activity content of the marketing activity according to the new marketing activity trigger expression to obtain the second activity content of the marketing activity, and determine the second delivery information of the marketing activity after adjusting the activity content according to the new marketing activity trigger expression and the second activity content;

[0146] A delivery module 240, configured to re-deliver the marketing activity after adjusting the activity content according to the second delivery information when the new marketing activity trigger expression is satisfied.

[0147] In a possible implementation manner, the adjustment module 220 is configured to:

[0148] Construct a virtual intelligent agent corresponding to the user, and loop to execute: use the marketing user information of the user in the marketing activity delivered by the first delivery information as the basis for selecting the movement path of the virtual intelligent agent, and determine the branch point of each movement from the forest tree corresponding to the marketing activity trigger expression;

[0149] For the probability that the virtual intelligent agent selects a movement path at any of the branch points, determine the action value assignment for moving to the next branch point according to the movement path each time;

[0150] Calculate the sum value of the action values of all the branch points passed by the virtual intelligent agent when it reaches a treetop of the forest tree each time;

[0151] When the sum value is less than the average value of the action values of the virtual intelligent agent corresponding to the user moving to all the treetops, sequentially select target trigger rules from the rule library at the treetops to replace the original trigger rules at the branch points, and re-execute the movement of the virtual intelligent agent until the difference between the adjacent two average values is less than a preset difference threshold, so as to obtain the marketing activity trigger expression corresponding to the optimal action value.

[0152] In a possible implementation manner, the adjustment module 220 is configured to:

[0153] Sequentially select target trigger rules from the rule library at the treetops to replace the original trigger rules at the branch points, and re-execute the movement of the virtual intelligent agent until the difference between the adjacent two average values is less than a preset difference threshold, and stop replacing the original trigger rules at the branch points;

[0154] Determine the maximum action value that the virtual agent corresponding to the user can obtain when reaching a treetop of the forest tree in the case of stopping replacing the original trigger rule of the branch point.

[0155] Take the branch point on the movement path corresponding to the maximum action value of the virtual agent corresponding to the user as the locked branch point. When the movement agent corresponding to other users executes the replacement of the original trigger rule of the branch point, do not replace the trigger rule corresponding to the locked branch point.

[0156] Perform virtual agent movement for the virtual agents corresponding to all users and replace the original trigger rule of the branch point until all users have completed traversal, and obtain the marketing activity trigger expression corresponding to the optimal action value.

[0157] In a possible implementation manner, the adjustment module 220 is configured as follows:

[0158] Take the treetop as the target branch point and execute the replacement of the original trigger rule of the target branch point, and re - execute the virtual agent movement to reach the target branch point to obtain the action value assignment for moving to this target branch point along the movement path.

[0159] Determine whether the difference between the average values corresponding to the previous movement and the current movement is less than the preset difference threshold.

[0160] In the case of being less than the preset difference threshold, stop replacing the original trigger rule of the target branch point, and take the parent node of the target branch point in the forest tree as the next target branch point, and re - execute the virtual agent movement to reach the target branch point, and so on, until the last branch point on this movement path has been used as the target branch point, re - execute the virtual agent movement to reach the target branch point, and stop replacing the original trigger rule of the branch point.

[0161] In a possible implementation manner, the adjustment module 220 is configured as follows:

[0162] In the case of being greater than or equal to the preset difference threshold, select a target trigger rule from the rule library again to replace the original trigger rule of the target branch point, and re - execute the movement of the virtual agent until the difference between two adjacent average values is less than the preset difference threshold.

[0163] In a possible implementation manner, the determination module 230 is configured as follows:

[0164] According to the triggering conditions and corresponding triggering rules in the new marketing campaign triggering expression, adjust the first campaign content of the marketing campaign to obtain the second campaign content of the marketing campaign. Among them, adjusting the first campaign content of the marketing campaign includes at least one of the following: changing the preferential intensity of the campaign, adjusting the time range of the campaign, adding or deleting campaign products or services, and modifying the promotional copy of the campaign;

[0165] According to the new marketing campaign triggering expression and the second campaign content, re-determine the second delivery information of the adjusted marketing campaign. The second delivery information includes at least one of the following: delivery time, delivery channel, delivery frequency, and the user group to be delivered to.

[0166] In a possible implementation manner, the marketing user information includes at least one of the following:

[0167] The number of times of participating in the marketing campaign, the duration of participating in the marketing campaign, the time period of participating in the marketing campaign, and the basic information of the users participating in the marketing campaign.

[0168] The embodiments of the present disclosure also provide a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the steps of the method described in any one of the foregoing embodiments are implemented.

[0169] The embodiments of the present disclosure also provide an electronic device, including:

[0170] A memory, on which a computer program is stored;

[0171] A processor, configured to execute the computer program in the memory to implement the steps of the method described in any one of the foregoing embodiments.

[0172] Figure 3 The marketing intelligent trigger and optimization device 100 based on a rule engine shown includes: a processor 1001 and a memory 1003. Among them, the processor 1001 and the memory 1003 are connected, such as connected through a bus 1002. Optionally, the marketing intelligent trigger and optimization device 100 based on a rule engine may further include a communication component 1004. The communication component 1004 may be used for data interaction between the device 100 and other devices, such as sending and / or receiving data, etc. It should be noted that in actual scheduling, the communication component 1004 is not limited to one, and the structure of the marketing intelligent trigger and optimization device 100 based on a rule engine does not constitute a limitation to the embodiments of the present application.

[0173] The processor 1001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of this application. The processor 1001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0174] The bus 1002 may include a path for transmitting information between the above components. The bus 1002 may be a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The bus 1002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 only a thick line is shown herein, but it does not mean that there is only one bus or one type of bus.

[0175] The memory 1003 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, or it may also be an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store program code and can be read by a computer, which is not limited herein.

[0176] The memory 1003 is used to store program codes for executing the embodiments of the present disclosure, and the execution is controlled by the processor 1001. The processor 1001 is used to execute the program codes stored in the memory 1003 to implement the steps shown in the above-mentioned marketing intelligent triggering and optimization method embodiment based on the rule engine.

[0177] The disclosed embodiment also provides a computer-readable storage medium having program code stored thereon. When the program code is executed by a processor, the steps and corresponding contents of the aforementioned marketing intelligent triggering and optimization method embodiment based on a rule engine can be implemented.

[0178] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments; within the technical concept of the present disclosure, various changes, modifications, substitutions and variations may be made to these embodiments, and these changes, modifications, substitutions and variations all fall within the protection scope of the present disclosure.

[0179] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any appropriate manner without contradiction, and they should also be regarded as the contents disclosed in this disclosure. In order to avoid unnecessary repetition, this disclosure will not further describe various possible combinations. The technical scope of this application is not limited to the contents in the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A marketing intelligent triggering and optimization method based on rule engine, characterized in that: include: Acquiring marketing user information of a user in a marketing activity with first activity content delivered through first delivery information, wherein the marketing activity with first activity content delivered through the first delivery information is based on a random forest algorithm, and after selecting a trigger rule from a rule library according to the first delivery information to construct a marketing activity trigger expression, the marketing activity is delivered to the user when the marketing activity trigger expression is satisfied; According to the marketing user information of the user in the marketing activity delivered by the first delivery information, based on the reinforcement learning algorithm, with the optimal action value as the goal, the trigger rule in the marketing activity trigger expression is adjusted to obtain a new marketing activity trigger expression; According to the new marketing activity trigger expression, the first activity content of the marketing activity is adjusted to obtain the second activity content of the marketing activity, and according to the new marketing activity trigger expression and the second activity content, second delivery information of the marketing activity after the activity content is adjusted is determined; When the new marketing activity trigger expression is satisfied, re-delivering the marketing activity after adjusting the activity content according to the second delivery information; Wherein, the triggering rule in the marketing activity triggering expression is adjusted based on the marketing user information in the marketing activity delivered by the first delivery information, based on the reinforcement learning algorithm and taking the optimal action value as the goal, to obtain a new marketing activity triggering expression, including: Constructing a virtual agent corresponding to the user, and executing the steps in a loop: using the marketing user information of the user in the marketing activity delivered by the first delivery information as a basis for selecting a movement path of the virtual agent, and determining a branch point for each movement from the forest tree corresponding to the marketing activity trigger expression; For the probability of the virtual agent selecting a movement path at any branch point, determine the action value assignment for each movement to the next branch point according to the movement path; Calculating the sum of the action values ​​of all the branch points along the path each time the virtual agent reaches a treetop of the forest tree; When the sum is smaller than the average value of the action value of the virtual agent corresponding to the user moving to all treetops, target trigger rules are selected from the rule library from the treetops in turn to replace the original trigger rules of the branch points, and the movement of the virtual agent is re-executed until the difference between the two adjacent average values ​​is smaller than the preset difference threshold, so as to obtain the marketing activity trigger expression corresponding to the optimal action value.

2. The marketing intelligent triggering and optimization method based on rule engine according to claim 1, characterized in that: The target trigger rule is selected from the rule base in turn from the treetop to replace the original trigger rule of the branch point, and the movement of the virtual agent is re-executed until the difference between the two adjacent average values ​​is less than the preset difference threshold, and the marketing activity trigger expression corresponding to the optimal action value is obtained, including: Selecting target trigger rules from the rule base from the treetop in turn to replace the original trigger rules of the branch point, and re-executing the movement of the virtual agent until the difference between the two adjacent average values ​​is less than a preset difference threshold, and then stopping replacing the original trigger rules of the branch point; Determine the maximum action value that the virtual agent corresponding to the user can obtain when reaching a treetop of the forest tree under the condition that the virtual agent corresponding to the user stops changing the original trigger rule of the branch point; The branch point on the moving path corresponding to the maximum action value of the virtual agent corresponding to the user is used as a locked branch point, and when the original trigger rule of the branch point is replaced by executing the mobile agent corresponding to other users, the trigger rule corresponding to the locked branch point is not replaced; The virtual agents corresponding to all users are moved and the original triggering rules of the branch points are replaced until all users have completed the traversal and the marketing activity triggering expression corresponding to the optimal action value is obtained.

3. The marketing intelligent triggering and optimization method based on rule engine according to claim 2, characterized in that: The step of selecting a target trigger rule from the rule base in turn from the treetop to replace the original trigger rule of the branch point, and re-executing the movement of the virtual agent until the difference between two adjacent average values ​​is less than a preset difference threshold, and stopping replacing the original trigger rule of the branch point, includes: When the treetop is used as the target branch point, the original trigger rule of the target branch point is replaced, and the movement of the virtual agent to reach the target branch point is re-executed to obtain the action value assignment of moving to the target branch point according to the movement path; Determine whether the difference between the average values ​​corresponding to the previous movement and the current movement is less than the preset difference threshold; When the difference is less than the preset difference threshold, stop replacing the original trigger rule of the target branch point, and take the parent node of the target branch point in the forest tree as the next target branch point, re-execute the movement of the virtual agent to reach the target branch point, and so on, until the last branch point on the movement path has been taken as the target branch point, re-execute the movement of the virtual agent to reach the target branch point, and stop replacing the original trigger rule of the branch point.

4. The marketing intelligent triggering and optimization method based on rule engine according to claim 3, characterized in that: The method further comprises: When it is greater than or equal to the preset difference threshold, the target trigger rule is selected from the rule library again to replace the original trigger rule of the target branch point, and the movement of the virtual agent is re-executed until the difference between the two adjacent average values ​​is less than the preset difference threshold.

5. The marketing intelligent triggering and optimization method based on rule engine according to claim 1, characterized in that: The step of adjusting the first activity content of the marketing activity according to the new marketing activity trigger expression to obtain the second activity content of the marketing activity, and determining the second delivery information of the marketing activity after adjusting the activity content according to the new marketing activity trigger expression and the second activity content, includes: According to the trigger condition and the corresponding trigger rule in the new marketing activity trigger expression, the first activity content of the marketing activity is adjusted to obtain the second activity content of the marketing activity, wherein the adjustment of the first activity content of the marketing activity includes at least one of the following: changing the preferential strength of the activity, adjusting the time range of the activity, adding or deleting activity goods or services, and modifying the promotional copy of the activity; According to the new marketing activity trigger expression and the second activity content, the second delivery information of the adjusted marketing activity is re-determined, and the second delivery information includes at least one of the following: delivery time, delivery channel, delivery frequency and delivery user group.

6. The marketing intelligent triggering and optimization method based on rule engine according to claim 1, characterized in that: The marketing user information includes at least one of the following: The frequency of participation in the marketing activity, the duration of participation in the marketing activity, the time period of participation in the marketing activity, and basic information of users participating in the marketing activity.

7. A marketing intelligent triggering and optimization device based on a rule engine, characterized in that: The device comprises: an acquisition module, configured to acquire marketing user information of a user in a marketing activity with first activity content delivered through first delivery information, wherein the marketing activity with first activity content delivered through the first delivery information is based on a random forest algorithm, after selecting a trigger rule from a rule library according to the first delivery information to construct a marketing activity trigger expression, and delivering to the user when the marketing activity trigger expression is satisfied; an adjustment module configured to adjust the trigger rule in the marketing activity trigger expression based on the marketing user information in the marketing activity delivered by the first delivery information, based on the reinforcement learning algorithm and with the optimal action value as the goal, to obtain a new marketing activity trigger expression; a determination module configured to adjust the first activity content of the marketing activity according to the new marketing activity trigger expression to obtain the second activity content of the marketing activity, and determine the second delivery information of the marketing activity after the activity content is adjusted according to the new marketing activity trigger expression and the second activity content; a delivery module configured to re-deliver the marketing activity after adjusting the activity content according to the second delivery information when the new marketing activity trigger expression is satisfied; Wherein, the adjustment module is configured as follows: Constructing a virtual agent corresponding to the user, and executing the steps in a loop: using the marketing user information of the user in the marketing activity delivered by the first delivery information as a basis for selecting a movement path of the virtual agent, and determining a branch point for each movement from the forest tree corresponding to the marketing activity trigger expression; For the probability of the virtual agent selecting a movement path at any branch point, determine the action value assignment for each movement to the next branch point according to the movement path; Calculating the sum of the action values ​​of all the branch points along the path each time the virtual agent reaches a treetop of the forest tree; When the sum is smaller than the average value of the action value of the virtual agent corresponding to the user moving to all treetops, target trigger rules are selected from the rule library from the treetops in turn to replace the original trigger rules of the branch points, and the movement of the virtual agent is re-executed until the difference between the two adjacent average values ​​is smaller than the preset difference threshold, so as to obtain the marketing activity trigger expression corresponding to the optimal action value.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are implemented.

9. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 6.

Images