Intelligent agent map specific crowd message reaching method and system and readable storage medium

Abstract

The embodiment of the invention provides an agent map specific crowd message reaching method and system and a readable storage medium. The method comprises the steps of obtaining multi-source data processing of crowds in a specific area to construct an agent map, obtaining target condition crowds according to message pushing target attribute feature processing, and obtaining personalized messages according to personalized features of the target condition crowds in combination with core information, according to target condition crowd agent map characteristic parameters, individual characteristic parameters and attribute parameter processing, a group criticality value is obtained, a corresponding message propagation path level and a message pushing strategy are obtained, a reaching effect level is obtained according to message reaching data processing, an agent map is updated, and a message generation and pushing strategy is optimized; by constructing the agent map, screening the target population, generating the message in a personalized manner, making the message pushing strategy and evaluating and optimizing the message reaching effect, the precise and personalized pushing of the message is realized, the message reaching rate and the user acceptability are improved, and the multi-scene application requirements are met.

Description

Technical Field

[0001] This application relates to the field of message push technology, and more specifically, to a method, system, and readable storage medium for delivering messages to specific groups of people in an intelligent agent graph. Background Technology

[0002] In today's information explosion, there is a core need to efficiently and accurately deliver important messages to specific groups of people in many scenarios such as emergency rescue, commercial promotion, and public service notifications. Traditional message push methods generally suffer from low delivery efficiency, poor accuracy, and high user interference. Existing push technologies based on geolocation information mostly rely on simple regional divisions and do not fully consider key factors such as individual differences, behavioral habits, social relationships, and scenario adaptability of people in the region. This results in poor message push effects and makes it difficult to meet the actual needs of efficiently reaching specific groups of people in different scenarios.

[0003] Therefore, relevant technical solutions are urgently needed to address the above problems. Summary of the Invention

[0004] The purpose of this application is to provide a method, system, and readable storage medium for delivering messages to specific groups of people using an intelligent agent graph. This can achieve precise and personalized message delivery, improve message reach and user acceptance, and adapt to the needs of multiple application scenarios.

[0005] This application also provides a method for reaching specific groups of people through an intelligent agent graph, including the following steps: Acquire and process multi-source population data within a specific region to construct an intelligent agent map; The target audience is obtained by processing the attribute feature data of the message push target; Personalized messages are obtained by processing the personalized feature data of the target population in combination with core information. Based on the feature parameter data of the agent graph of the target population, the individual characteristic parameters are combined with the attribute parameter data to obtain the group criticality value, and the corresponding message propagation path level and message push strategy are obtained. The message reach data is processed to obtain the reach effect level, the agent graph is updated, and the message generation and push strategy is optimized.

[0006] Optionally, in the method for reaching specific groups of people using an intelligent agent graph as described in this application embodiment, the step of acquiring and processing multi-source data of the population within a specific area to construct an intelligent agent graph includes: Acquire multi-source data on people within a specific region, including basic user information, behavioral data, and social relationship data; The multi-source data is cleaned, fused, and standardized to obtain standardized multi-source data; A graph structure is constructed based on the node and boundary relationships of the standardized multi-source data to build an agent graph.

[0007] Optionally, in the method for reaching specific groups of people using an intelligent agent graph as described in this application embodiment, the step of processing the attribute feature data of the message push target to obtain the target group includes: Obtain attribute feature data of the message push target, including specific geographic location range data, user attribute data, and behavioral pattern data; The attribute feature data is processed by combining the intelligent agent graph with a preset algorithm model to obtain the target population.

[0008] Optionally, in the method for reaching specific groups of people using an intelligent agent graph as described in this application embodiment, the step of processing the personalized feature data of the target group in combination with core information to obtain personalized messages includes: Obtain personalized feature data of the target population and construct a feature tag library; The personalized feature data includes the user's interests, language style, and message receiving habits; Obtain the core information of the message push target, including key information, supporting instructions, and action instructions; The personalized feature data is processed in conjunction with core information to generate and present personalized messages.

[0009] Optionally, in the method for reaching specific groups of people using an agent graph as described in this application embodiment, the step of processing the feature parameter data, individual characteristic parameters, and attribute parameter data of the agent graph of the target group to obtain a group keyness value, and obtaining the corresponding message propagation path hierarchy and message push strategy, includes: Obtain the attribute parameter data of the message to be pushed, including the message timeliness weight coefficient and the message importance weight coefficient; Obtain characteristic parameter data of the target population, including the size and distribution characteristics of the target population; The attribute parameter data and feature parameter data are standardized. Individual characteristic parameters, including social reach, interaction frequency, and information forwarding rate, are extracted from the intelligent agent graph of the target population. The social influence index is obtained by weighting the individual characteristic parameters. The social influence index is aggregated and combined with standardized message timeliness weight coefficients and message importance weight coefficients, as well as scale data and distribution characteristic data, to obtain the group keyness value. Based on the group criticality value, a threshold judgment is performed to divide the message propagation path hierarchy and the corresponding message push strategy.

[0010] Optionally, in the method for reaching specific groups of people using an intelligent agent graph as described in this application embodiment, the step of processing the message reaching data to obtain the reaching effect level, updating the intelligent agent graph, and optimizing the message generation and push strategy includes: Acquire message delivery data, including core performance data, user feedback data, and behavioral trajectory data, and perform standardized processing. The standardized message delivery data is processed to obtain a comprehensive score for delivery effectiveness, and the delivery effectiveness is classified into levels. The reach effect level is processed by a preset analysis model to obtain core influencing factors. The agent graph is dynamically updated based on the core influencing factors, and the message generation and push strategies are optimized.

[0011] Secondly, embodiments of this application provide a message delivery system for a specific group of people based on an intelligent agent graph. This system includes a memory and a processor. The memory includes a program for a message delivery method for a specific group of people based on an intelligent agent graph. When the program for the message delivery method for a specific group of people based on an intelligent agent graph is executed by the processor, it implements the following steps: Acquire and process multi-source population data within a specific region to construct an intelligent agent map; The target audience is obtained by processing the attribute feature data of the message push target; Personalized messages are obtained by processing the personalized feature data of the target population in combination with core information. Based on the feature parameter data of the agent graph of the target population, the individual characteristic parameters are combined with the attribute parameter data to obtain the group criticality value, and the corresponding message propagation path level and message push strategy are obtained. The message reach data is processed to obtain the reach effect level, the agent graph is updated, and the message generation and push strategy is optimized.

[0012] Optionally, in the intelligent agent graph-based message delivery system for specific groups described in this application embodiment, the step of acquiring and processing multi-source data of the population within a specific area to construct an intelligent agent graph includes: Acquire multi-source data on people within a specific region, including basic user information, behavioral data, and social relationship data; The multi-source data is cleaned, fused, and standardized to obtain standardized multi-source data; A graph structure is constructed based on the node and boundary relationships of the standardized multi-source data to build an agent graph.

[0013] Optionally, in the intelligent agent graph-based message delivery system for specific groups described in this application embodiment, the step of processing the attribute feature data of the message push target to obtain the target group includes: Obtain attribute feature data of the message push target, including specific geographic location range data, user attribute data, and behavioral pattern data; The attribute feature data is processed by combining the intelligent agent graph with a preset algorithm model to obtain the target population.

[0014] Thirdly, embodiments of this application also provide a computer-readable storage medium, which includes a method program for reaching specific groups of people in an intelligent agent graph. When the method program is executed by a processor, it implements the steps of the method program for reaching specific groups of people in an intelligent agent graph as described in any of the above claims.

[0015] As can be seen from the above, the intelligent agent graph-based method, system, and readable storage medium provided in this application construct an intelligent agent graph by processing multi-source data of a population in a specific region, obtains the target population based on the target attribute features of the message push, obtains personalized messages based on the personalized features of the target population combined with core information, obtains the group keyness value based on the intelligent agent graph feature parameters, individual characteristic parameters, and attribute parameters of the target population, obtains the corresponding message propagation path level and message push strategy, obtains the reach effect level based on message reach data processing, updates the intelligent agent graph, and optimizes the message generation and push strategy. By constructing an intelligent agent graph, selecting the target population, generating personalized messages, formulating message push strategies, and evaluating and optimizing message reach effects, the method achieves accurate and personalized message push, improves message reach rate and user acceptance, and adapts to the needs of multiple application scenarios.

[0016] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the method for delivering messages to specific groups of people using an intelligent agent graph, as provided in this application embodiment.

[0019] Figure 2 The flowchart illustrates the construction of the intelligent agent graph for the method of reaching specific groups of people using the intelligent agent graph provided in this application embodiment.

[0020] Figure 3 A high-level flowchart of the method for reaching specific groups of people using an intelligent agent graph, as provided in the embodiments of this application. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0022] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0023] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating message delivery to specific groups within an intelligent agent graph, as described in some embodiments of this application. This method for delivering messages to specific groups within an intelligent agent graph is used in terminal devices, such as mobile phones and computers. The method includes the following steps: S11. Acquire and process multi-source data of the population in a specific area to construct an intelligent agent map; S12. Process the attribute feature data of the message push target to obtain the target audience; S13. Process the personalized feature data of the target population in combination with core information to obtain personalized messages; S14. Based on the feature parameter data of the intelligent agent graph of the target population, the individual characteristic parameters and attribute parameter data are processed to obtain the group criticality value, and the corresponding message propagation path level and message push strategy are obtained. S15. Process the message reach data to obtain the reach effect level, update the agent graph, and optimize the message generation and push strategy.

[0024] This process involves acquiring and processing multi-source data on the population within a specific region to construct an intelligent agent graph. It then processes the attribute characteristics of the message push target to identify the target population. Next, it combines personalized characteristic data of the target population with core information to generate personalized messages. Finally, it processes feature parameters, individual characteristic parameters, and attribute parameters from the intelligent agent graph of the target population to obtain a group keyness value, leading to the corresponding message propagation path hierarchy and message push strategy. Finally, it processes message reach data to determine the reach effectiveness level, updates the intelligent agent graph, and optimizes message generation and push strategies. Through constructing an intelligent agent graph, selecting target populations, generating personalized messages, formulating message push strategies, and evaluating and optimizing message reach effectiveness, the process achieves precise and personalized message push, improving message reach and user acceptance, and adapting to the needs of multiple application scenarios.

[0025] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating the construction of an agent graph for reaching specific groups of people in some embodiments of this application. According to embodiments of the present invention, the step of acquiring and processing multi-source data of people within a specific region to construct an agent graph specifically involves: S21. Obtain multi-source data on the population within a specific area, including basic user information data, behavioral data, and social relationship data; S22. The multi-source data is cleaned, fused, and standardized to obtain standardized multi-source data; S23. Construct a graph structure based on the node and boundary relationships of the standardized multi-source data to build an intelligent agent graph.

[0026] This process involves acquiring multi-source data on people within a specific region, including basic user information such as age, gender, and occupation; behavioral data such as location trajectories, consumption records, and online behavior; and social relationship data such as friend relationships and group relationships in social networks. The multi-source data is then cleaned, fused, and standardized to obtain standardized multi-source data. A graph structure is constructed based on the nodes and boundary relationships of the standardized multi-source data, representing each user as a node. Relationships between users and between users and various entities such as locations, events, and points of interest are represented as edges. The weights of the edges are set according to the strength of the association, such as interaction frequency and visit frequency. This constructs an intelligent agent graph, forming a complex network structure that characterizes the individual characteristics, relationships, and scene adaptability of people within the region.

[0027] According to an embodiment of the present invention, the step of processing the attribute feature data of the message push target to obtain the target audience specifically involves: Obtain attribute feature data of the message push target, including specific geographic location range data, user attribute data, and behavioral pattern data; The attribute feature data is processed by combining the intelligent agent graph with a preset algorithm model to obtain the target population.

[0028] The process involves acquiring attribute feature data of the message push target, including specific geographical location range data, specific user attribute data such as age range and occupation category, and specific behavioral pattern data such as frequent visits to certain types of places and recent consumption behaviors. Further, to determine the screening criteria for the target audience, a pre-set algorithm model is used, employing graph traversal and node matching models. This model is based on graph neural networks such as the GAT algorithm, which extracts features from a large number of sample agent graphs to train classification or prediction models such as logistic regression. This allows for the classification or prediction of the target audience. The attribute feature data of the message push target is then input into the trained model, which outputs the probability or classification result of each individual belonging to the target audience. Based on a set threshold, the target audience is selected, thus leveraging the structured characteristics of the graph to quickly locate the user group that meets the target criteria, serving as the initial target audience for message pushes.

[0029] According to an embodiment of the present invention, the step of processing the personalized feature data of the target population in combination with core information to obtain personalized messages specifically includes: Obtain personalized feature data of the target population and construct a feature tag library; The personalized feature data includes the user's interests, language style, and message receiving habits; Obtain the core information of the message push target, including key information, supporting instructions, and action instructions; The personalized feature data is processed in conjunction with core information to generate and present personalized messages.

[0030] Specifically, for the selected target audience, personalized feature data is extracted for each user, including user interests and preferences such as technical fields, content depth, language style such as concise professionalism or colloquialism, message receiving habits such as reading time and preferred media. A structured feature tag library for the user group is established. This data is then processed in conjunction with the core information of the message push target, including key information, auxiliary instructions, and action instructions, to generate personalized messages that match the user. This includes adjusting the expression, emphasis, and presentation format such as text, image and text combination, short video, and voice, making the message more in line with the user's focus and receiving habits, and improving the message's attractiveness and readability. For example, for users who prefer concise and clear information, a short and concise message text is generated; for users who are more sensitive to visual elements, appropriate images or video content is added. For the same message topic, differentiated versions are generated for users with different characteristics to ensure that the message matches the user's focus and receiving habits.

[0031] According to an embodiment of the present invention, the step of processing the feature parameter data, individual characteristic parameters, and attribute parameter data of the agent graph of the target population to obtain the group criticality value, and obtaining the corresponding message propagation path hierarchy and message push strategy, specifically includes: Obtain the attribute parameter data of the message to be pushed, including the message timeliness weight coefficient and the message importance weight coefficient; Obtain characteristic parameter data of the target population, including the size and distribution characteristics of the target population; The attribute parameter data and feature parameter data are standardized. Individual characteristic parameters, including social reach, interaction frequency, and information forwarding rate, are extracted from the intelligent agent graph of the target population. The social influence index is obtained by weighting the individual characteristic parameters. The social influence index is aggregated and combined with standardized message timeliness weight coefficients and message importance weight coefficients, as well as scale data and distribution characteristic data, to obtain the group keyness value. Based on the group criticality value, a threshold judgment is performed to divide the message propagation path hierarchy and the corresponding message push strategy.

[0032] This process involves acquiring attribute parameter data for the message to be pushed and feature parameter data for the target audience. Attribute parameter data includes message timeliness weight and message importance weight, while feature parameter data includes scale data and distribution characteristic data. Each parameter is standardized to obtain a quantified value with a unified dimension. Based on the agent graph of the target audience, feature parameters for each individual in the group are extracted, including each user's social reach, interaction frequency, and information forwarding rate. A weighted calculation is then performed to obtain the social influence index for each user node, with weight coefficients pre-set based on the group's characteristic attributes. The social influence indices of each user in the target audience are then aggregated and combined with the standardized and normalized message timeliness and importance weight coefficients, scale data, and... The distribution feature data is weighted to obtain the group criticality value, which reflects the criticality evaluation result of the target population. Based on the group criticality value, the group is divided according to the threshold range to obtain the corresponding message propagation path level and the message push strategy corresponding to the level. For example, if the group criticality value of a target population is 0.82 and falls within the threshold range of [0.7, 0.9], the corresponding message propagation path level is level III (divided into levels I to IV). The corresponding message push strategy is a three-level push strategy. In this embodiment, the three-level push strategy is to use a non-urgent, secondary priority channel, such as unencrypted letters, for immediate message push. The push process is triggered immediately after the message is generated to push the unencrypted letter immediately.

[0033] According to an embodiment of the present invention, the step of processing message reach data to obtain a reach effectiveness level, updating the agent graph, and optimizing message generation and push strategies specifically includes: Acquire message delivery data, including core performance data, user feedback data, and behavioral trajectory data, and perform standardized processing. The standardized message delivery data is processed to obtain a comprehensive score for delivery effectiveness, and the delivery effectiveness is classified into levels. The reach effect level is processed by a preset analysis model to obtain core influencing factors. The agent graph is dynamically updated based on the core influencing factors, and the message generation and push strategies are optimized.

[0034] The acquisition of message reach data includes core performance data such as message delivery rate, read rate, click-through rate, and forwarding rate; user feedback data such as active ratings, comment keywords, unsubscribe records, and complaint feedback types; and behavioral trajectory data such as message open duration, click location distribution, and subsequent operation paths. This data is then cleaned, deduplicated, and standardized. Weights are assigned to the standardized message reach data to obtain a comprehensive reach performance score, which is used to classify reach performance levels. In this embodiment, the weights are determined using the Analytic Hierarchy Process (AHP), such as a click-through rate weight of 0.3, a read rate weight of 0.3, a positive user feedback rate weight of 0.2, and a subsequent conversion rate weight of 0.2. The comprehensive reach performance score for a single push is calculated, and reach performance levels are classified. A pre-defined analysis model, such as an attribution analysis model, is used to analyze and process the reach performance levels to obtain core influencing factors. If the message is delivered… If the rate is low, analyze the effectiveness of the push channel and the status of user devices. If the read rate / click rate is low, analyze the quality of the message content and the rationality of the push timing. If user negative feedback is high, analyze the message frequency and content relevance. Dynamically update the agent graph based on the core influencing factors, such as correcting the weight of user interest tags, supplementing user push preferences, marking sensitive push types, optimizing content feature tags, establishing content-user matching rules, and adjusting the relevant parameters of the push strategy. This will obtain the updated message propagation path hierarchy and message push strategy. Then, process and collect the comprehensive score data of the updated reach effect after the new strategy update, and compare it with the original effect score to obtain the improvement. If the improvement is greater than 15%, the new strategy is set as the standard strategy and enters the next round of push. If the improvement is less than 15%, re-analyze the influencing factors and optimize the graph until a strategy that meets the effect requirements is formed, and continue to optimize.

[0035] Please refer to Figure 3 , Figure 3 This is a high-level flowchart of the method for reaching specific groups of people using an intelligent agent graph in some embodiments of this application.

[0036] This invention also discloses a message delivery system for a specific group of people based on an intelligent agent graph, including a memory and a processor. The memory includes a method program for delivering messages to a specific group of people based on an intelligent agent graph. When the processor executes the method program for delivering messages to a specific group of people based on an intelligent agent graph, it performs the following steps: Acquire and process multi-source population data within a specific region to construct an intelligent agent map; The target audience is obtained by processing the attribute feature data of the message push target; Personalized messages are obtained by processing the personalized feature data of the target population in combination with core information. Based on the feature parameter data of the agent graph of the target population, the individual characteristic parameters are combined with the attribute parameter data to obtain the group criticality value, and the corresponding message propagation path level and message push strategy are obtained. The message reach data is processed to obtain the reach effect level, the agent graph is updated, and the message generation and push strategy is optimized.

[0037] This process involves acquiring and processing multi-source data on the population within a specific region to construct an intelligent agent graph. It then processes the attribute characteristics of the message push target to identify the target population. Next, it combines personalized characteristic data of the target population with core information to generate personalized messages. Finally, it processes feature parameters, individual characteristic parameters, and attribute parameters from the intelligent agent graph of the target population to obtain a group keyness value, leading to the corresponding message propagation path hierarchy and message push strategy. Finally, it processes message reach data to determine the reach effectiveness level, updates the intelligent agent graph, and optimizes message generation and push strategies. Through constructing an intelligent agent graph, selecting target populations, generating personalized messages, formulating message push strategies, and evaluating and optimizing message reach effectiveness, the process achieves precise and personalized message push, improving message reach and user acceptance, and adapting to the needs of multiple application scenarios.

[0038] According to an embodiment of the present invention, the step of acquiring and processing multi-source data of the population in a specific area to construct an intelligent agent map specifically includes: Acquire multi-source data on people within a specific region, including basic user information, behavioral data, and social relationship data; The multi-source data is cleaned, fused, and standardized to obtain standardized multi-source data; A graph structure is constructed based on the node and boundary relationships of the standardized multi-source data to build an agent graph.

[0039] This process involves acquiring multi-source data on people within a specific region, including basic user information such as age, gender, and occupation; behavioral data such as location trajectories, consumption records, and online behavior; and social relationship data such as friend relationships and group relationships in social networks. The multi-source data is then cleaned, fused, and standardized to obtain standardized multi-source data. A graph structure is constructed based on the nodes and boundary relationships of the standardized multi-source data, representing each user as a node. Relationships between users and between users and various entities such as locations, events, and points of interest are represented as edges. The weights of the edges are set according to the strength of the association, such as interaction frequency and visit frequency. This constructs an intelligent agent graph, forming a complex network structure that characterizes the individual characteristics, relationships, and scene adaptability of people within the region.

[0040] According to an embodiment of the present invention, the step of processing the attribute feature data of the message push target to obtain the target audience specifically involves: Obtain attribute feature data of the message push target, including specific geographic location range data, user attribute data, and behavioral pattern data; The attribute feature data is processed by combining the intelligent agent graph with a preset algorithm model to obtain the target population.

[0041] The process involves acquiring attribute feature data of the message push target, including specific geographical location range data, specific user attribute data such as age range and occupation category, and specific behavioral pattern data such as frequent visits to certain types of places and recent consumption behaviors. Further, to determine the screening criteria for the target audience, a pre-set algorithm model is used, employing graph traversal and node matching models. This model is based on graph neural networks such as the GAT algorithm, which extracts features from a large number of sample agent graphs to train classification or prediction models such as logistic regression. This allows for the classification or prediction of the target audience. The attribute feature data of the message push target is then input into the trained model, which outputs the probability or classification result of each individual belonging to the target audience. Based on a set threshold, the target audience is selected, thus leveraging the structured characteristics of the graph to quickly locate the user group that meets the target criteria, serving as the initial target audience for message pushes.

[0042] According to an embodiment of the present invention, the step of processing the personalized feature data of the target population in combination with core information to obtain personalized messages specifically includes: Obtain personalized feature data of the target population and construct a feature tag library; The personalized feature data includes the user's interests, language style, and message receiving habits; Obtain the core information of the message push target, including key information, supporting instructions, and action instructions; The personalized feature data is processed in conjunction with core information to generate and present personalized messages.

[0043] Specifically, for the selected target audience, personalized feature data is extracted for each user, including user interests and preferences such as technical fields, content depth, language style such as concise professionalism or colloquialism, message receiving habits such as reading time and preferred media. A structured feature tag library for the user group is established. This data is then processed in conjunction with the core information of the message push target, including key information, auxiliary instructions, and action instructions, to generate personalized messages that match the user. This includes adjusting the expression, emphasis, and presentation format such as text, image and text combination, short video, and voice, making the message more in line with the user's focus and receiving habits, and improving the message's attractiveness and readability. For example, for users who prefer concise and clear information, a short and concise message text is generated; for users who are more sensitive to visual elements, appropriate images or video content is added. For the same message topic, differentiated versions are generated for users with different characteristics to ensure that the message matches the user's focus and receiving habits.

[0044] According to an embodiment of the present invention, the step of processing the feature parameter data, individual characteristic parameters, and attribute parameter data of the agent graph of the target population to obtain the group criticality value, and obtaining the corresponding message propagation path hierarchy and message push strategy, specifically includes: Obtain the attribute parameter data of the message to be pushed, including the message timeliness weight coefficient and the message importance weight coefficient; Obtain characteristic parameter data of the target population, including the size and distribution characteristics of the target population; The attribute parameter data and feature parameter data are standardized. Individual characteristic parameters, including social reach, interaction frequency, and information forwarding rate, are extracted from the intelligent agent graph of the target population. The social influence index is obtained by weighting the individual characteristic parameters. The social influence index is aggregated and combined with standardized message timeliness weight coefficients and message importance weight coefficients, as well as scale data and distribution characteristic data, to obtain the group keyness value. Based on the group criticality value, a threshold judgment is performed to divide the message propagation path hierarchy and the corresponding message push strategy.

[0045] This process involves acquiring attribute parameter data for the message to be pushed and feature parameter data for the target audience. Attribute parameter data includes message timeliness weight and message importance weight, while feature parameter data includes scale data and distribution characteristic data. Each parameter is standardized to obtain a quantified value with a unified dimension. Based on the agent graph of the target audience, feature parameters for each individual in the group are extracted, including each user's social reach, interaction frequency, and information forwarding rate. A weighted calculation is then performed to obtain the social influence index for each user node, with weight coefficients pre-set based on the group's characteristic attributes. The social influence indices of each user in the target audience are then aggregated and combined with the standardized and normalized message timeliness and importance weight coefficients, scale data, and... The distribution feature data is weighted to obtain the group criticality value, which reflects the criticality evaluation result of the target population. Based on the group criticality value, the group is divided according to the threshold range to obtain the corresponding message propagation path level and the message push strategy corresponding to the level. For example, if the group criticality value of a target population is 0.82 and falls within the threshold range of [0.7, 0.9], the corresponding message propagation path level is level III (divided into levels I to IV). The corresponding message push strategy is a three-level push strategy. In this embodiment, the three-level push strategy is to use a non-urgent, secondary priority channel, such as unencrypted letters, for immediate message push. The push process is triggered immediately after the message is generated to push the unencrypted letter immediately.

[0046] According to an embodiment of the present invention, the step of processing message reach data to obtain a reach effectiveness level, updating the agent graph, and optimizing message generation and push strategies specifically includes: Acquire message delivery data, including core performance data, user feedback data, and behavioral trajectory data, and perform standardized processing. The standardized message delivery data is processed to obtain a comprehensive score for delivery effectiveness, and the delivery effectiveness is classified into levels. The reach effect level is processed by a preset analysis model to obtain core influencing factors. The agent graph is dynamically updated based on the core influencing factors, and the message generation and push strategies are optimized.

[0047] The acquisition of message reach data includes core performance data such as message delivery rate, read rate, click-through rate, and forwarding rate; user feedback data such as active ratings, comment keywords, unsubscribe records, and complaint feedback types; and behavioral trajectory data such as message open duration, click location distribution, and subsequent operation paths. This data is then cleaned, deduplicated, and standardized. Weights are assigned to the standardized message reach data to obtain a comprehensive reach performance score, which is used to classify reach performance levels. In this embodiment, the weights are determined using the Analytic Hierarchy Process (AHP), such as a click-through rate weight of 0.3, a read rate weight of 0.3, a positive user feedback rate weight of 0.2, and a subsequent conversion rate weight of 0.2. The comprehensive reach performance score for a single push is calculated, and reach performance levels are classified. A pre-defined analysis model, such as an attribution analysis model, is used to analyze and process the reach performance levels to obtain core influencing factors. If the message is delivered… If the rate is low, analyze the effectiveness of the push channel and the status of user devices. If the read rate / click rate is low, analyze the quality of the message content and the rationality of the push timing. If user negative feedback is high, analyze the message frequency and content relevance. Dynamically update the agent graph based on the core influencing factors, such as correcting the weight of user interest tags, supplementing user push preferences, marking sensitive push types, optimizing content feature tags, establishing content-user matching rules, and adjusting the relevant parameters of the push strategy. This will obtain the updated message propagation path hierarchy and message push strategy. Then, process and collect the comprehensive score data of the updated reach effect after the new strategy update, and compare it with the original effect score to obtain the improvement. If the improvement is greater than 15%, the new strategy is set as the standard strategy and enters the next round of push. If the improvement is less than 15%, re-analyze the influencing factors and optimize the graph until a strategy that meets the effect requirements is formed, and continue to optimize.

[0048] A third aspect of the present invention provides a readable storage medium comprising a method program for reaching specific groups of people in an agent graph. When the method program is executed by a processor, it implements the steps of the method program for reaching specific groups of people in an agent graph as described in any of the preceding claims.

[0049] This invention discloses a method, system, and readable storage medium for delivering messages to specific groups using an intelligent agent graph. It constructs an intelligent agent graph by processing multi-source data of a population in a specific region. It then obtains the target population based on message push target attribute features, generates personalized messages based on the personalized characteristics of the target population combined with core information, and obtains a group keyness value by processing intelligent agent graph feature parameters, individual characteristic parameters, and attribute parameters. This yields the corresponding message propagation path hierarchy and message push strategy. Finally, it obtains the reach effect level based on message reach data processing, updates the intelligent agent graph, and optimizes the message generation and push strategy. Through constructing an intelligent agent graph, selecting the target population, generating personalized messages, formulating message push strategies, and evaluating and optimizing message reach effects, it achieves precise and personalized message delivery, improves message reach rate and user acceptance, and adapts to the needs of multiple application scenarios.

[0050] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0051] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0052] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0053] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0054] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for delivering messages to specific groups of people using an intelligent agent graph, characterized in that: Includes the following steps: Acquire and process multi-source population data within a specific region to construct an intelligent agent map; The target audience is obtained by processing the attribute feature data of the message push target; Personalized messages are obtained by processing the personalized feature data of the target population in combination with core information. Based on the feature parameter data of the agent graph of the target population, the individual characteristic parameters are combined with the attribute parameter data to obtain the group criticality value, and the corresponding message propagation path level and message push strategy are obtained. The message reach data is processed to obtain the reach effect level, the agent graph is updated, and the message generation and push strategy is optimized.

2. The method for reaching specific groups of people using an intelligent agent graph according to claim 1, characterized in that, The process of acquiring and processing multi-source data on the population within a specific area to construct an intelligent agent map includes: Acquire multi-source data on people within a specific region, including basic user information, behavioral data, and social relationship data; The multi-source data is cleaned, fused, and standardized to obtain standardized multi-source data; A graph structure is constructed based on the node and boundary relationships of the standardized multi-source data to build an agent graph.

3. The method for reaching specific groups of people using an intelligent agent graph according to claim 2, characterized in that, The process of processing the attribute feature data of the message push target to obtain the target audience includes: Obtain attribute feature data of the message push target, including specific geographic location range data, user attribute data, and behavioral pattern data; The attribute feature data is processed by combining the intelligent agent graph with a preset algorithm model to obtain the target population.

4. The method for reaching specific groups of people using an intelligent agent graph according to claim 1, characterized in that, The process of processing personalized feature data of the target population in conjunction with core information to obtain personalized messages includes: Obtain personalized feature data of the target population and construct a feature tag library; The personalized feature data includes the user's interests, language style, and message receiving habits; Obtain the core information of the message push target, including key information, supporting instructions, and action instructions; The personalized feature data is processed in conjunction with core information to generate and present personalized messages.

5. The method for reaching specific groups of people using an intelligent agent graph according to claim 1, characterized in that, The process of obtaining a group criticality value by combining feature parameter data, individual characteristic parameters, and attribute parameter data of the agent graph of the target population, and obtaining the corresponding message propagation path hierarchy and message push strategy, includes: Obtain the attribute parameter data of the message to be pushed, including the message timeliness weight coefficient and the message importance weight coefficient; Obtain characteristic parameter data of the target population, including the size and distribution characteristics of the target population; The attribute parameter data and feature parameter data are standardized. Individual characteristic parameters, including social reach, interaction frequency, and information forwarding rate, are extracted from the intelligent agent graph of the target population. The social influence index is obtained by weighting the individual characteristic parameters. The social influence index is aggregated and combined with standardized message timeliness weight coefficients and message importance weight coefficients, as well as scale data and distribution characteristic data, to obtain the group keyness value. Based on the group criticality value, a threshold judgment is performed to divide the message propagation path hierarchy and the corresponding message push strategy.

6. The method for reaching specific groups of people using an intelligent agent graph according to claim 1, characterized in that, The process of processing message reach data to obtain the reach effectiveness level, updating the agent graph, and optimizing message generation and push strategies includes: Acquire message delivery data, including core performance data, user feedback data, and behavioral trajectory data, and perform standardized processing. The standardized message delivery data is processed to obtain a comprehensive score for delivery effectiveness, and the delivery effectiveness is classified into levels. The reach effect level is processed by a preset analysis model to obtain core influencing factors. The agent graph is dynamically updated based on the core influencing factors, and the message generation and push strategies are optimized.

7. A system for delivering messages to specific groups of people based on an intelligent agent graph, characterized in that: The system includes a memory and a processor. The memory contains a program for a method of reaching specific groups of people using an intelligent agent graph. When the program for reaching specific groups of people using an intelligent agent graph is executed by the processor, it performs the following steps: Acquire and process multi-source population data within a specific region to construct an intelligent agent map; The target audience is obtained by processing the attribute feature data of the message push target; Personalized messages are obtained by processing the personalized feature data of the target population in combination with core information. Based on the feature parameter data of the agent graph of the target population, the individual characteristic parameters are combined with the attribute parameter data to obtain the group criticality value, and the corresponding message propagation path level and message push strategy are obtained. The message reach data is processed to obtain the reach effect level, the agent graph is updated, and the message generation and push strategy is optimized.

8. The intelligent agent graph-based message delivery system for specific groups of people according to claim 7, characterized in that, The process of acquiring and processing multi-source data on the population within a specific area to construct an intelligent agent map includes: Acquire multi-source data on people within a specific region, including basic user information, behavioral data, and social relationship data; The multi-source data is cleaned, fused, and standardized to obtain standardized multi-source data; A graph structure is constructed based on the node and boundary relationships of the standardized multi-source data to build an agent graph.

9. The intelligent agent graph-based message delivery system for specific groups of people according to claim 8, characterized in that, The process of processing the attribute feature data of the message push target to obtain the target audience includes: Obtain attribute feature data of the message push target, including specific geographic location range data, user attribute data, and behavioral pattern data; The attribute feature data is processed by combining the intelligent agent graph with a preset algorithm model to obtain the target population.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a method program for reaching specific groups of people through an intelligent agent graph. When the method program is executed by a processor, it implements the steps of the method for reaching specific groups of people through an intelligent agent graph as described in any one of claims 1 to 6.

Images