Information pushing method and device and storage medium

Abstract

The embodiment of the invention provides an information pushing method and device and a storage medium, which can be applied to the technical field of big data mining analysis, in the method, customer features of a plurality of customer clusters are obtained through private domain data, and service features of a plurality of service entities are obtained through public domain data; the client features of the client cluster comprise client geographic positions and group behavior features; the service features of the service entity comprise a service address position, a service basic attribute and a service dynamic attribute; determining client clusters and service entities belonging to the same fusion cluster from the plurality of client clusters and the plurality of service entities; according to any fusion cluster, the in-cluster customer density meets a first preset condition and / or the in-cluster customer behavior consistency meets a second preset condition; the fusion cluster is determined based on the behavior consistency factor, the dynamic influence of the service entity and the average distance of the clients; and information pushing is performed on the client clusters in the same fusion cluster based on the service entities in the same fusion cluster, so that the accuracy of information pushing is improved.

Description

Technical Field

[0001] This application relates to the field of big data mining and analysis technology, and in particular to an information push method, device and storage medium. Background Technology

[0002] In mainstream information push solutions, customer data and static data of service entities (such as business district type, average property price, etc.) are obtained based on the built basic data platform. The customer data is converted into structured data and associated with other data of the customer by customer ID. Multiple customers are clustered based on customer ID, and the clustering results are matched with the static data of service entities to push information to the customer groups in the clustering results.

[0003] However, once a customer ID is lost, the associated data becomes "unknown data," making it impossible to determine the user group. Furthermore, simply matching the static data of the service entity with the customer group in the clustering results lacks extended analysis of the geographical attributes of the service entity. The limitations of static data and the relatively singular matching dimension lead to reduced matching accuracy, thereby reducing the accuracy of information push. Summary of the Invention

[0004] This invention provides an information push method, device, and storage medium to improve the accuracy of information push.

[0005] On the one hand, embodiments of this application provide an information push method, which includes: Customer characteristics of multiple customer clusters are obtained through private domain data, and service characteristics of multiple service entities are obtained through public domain data; the customer characteristics of any customer cluster include customer geographic location and group behavior characteristics; the service characteristics of any service entity include service address location, basic service attributes, and dynamic service attributes. From multiple customer clusters and multiple service entities, customer clusters and service entities belonging to the same fusion cluster are determined; wherein, any fusion cluster is defined as having a customer density within the cluster that meets a first preset condition and / or customer behavior consistency within the cluster that meets a second preset condition; any fusion cluster is determined based on a behavior consistency factor, the dynamic influence of service entities, and the average distance of customers; the dynamic influence is determined based on service basic attributes and service dynamic attributes; Information is pushed to customer clusters within the same converged cluster based on service entities within the same converged cluster.

[0006] Optionally, after determining the customer cluster and service entity belonging to the same fusion cluster, the method further includes: For any service entity in the converged cluster, determine the overlapping area between customers and service entities in the same converged cluster; Based on the business scenario corresponding to the information push, obtain customer characteristics in the customer cluster corresponding to the business scenario and service characteristics in the service entity corresponding to the business scenario; The degree of association of the service entity in the fusion cluster under the business scenario is determined by the overlapping area, customer characteristics and service characteristics corresponding to the business scenario; Service entities that meet the correlation requirements are included as service entities in the fusion cluster.

[0007] Optionally, the information push to customer clusters within the same fusion cluster based on service entities within the same fusion cluster includes: Based on the historical behavior data of customers in the fusion cluster, service entities that match the customer's behavioral preferences are determined from the service entities of the fusion cluster; Push service information from service entities that match customer behavior preferences to the corresponding customers.

[0008] Optionally, from multiple customer clusters and multiple service entities, customer clusters and service entities belonging to the same converged cluster are identified, including: Based on at least one customer cluster within the target range, determine whether the customer density within the target range meets a first preset condition and / or whether the consistency of customer behavior within the target range meets a second preset condition; If the conditions are met, then the center point is determined based on the weights of the service entities within the target range, and the average distance of customers within the target range is determined based on the center point. Based on behavioral consistency factors determined by customer behavior consistency, geographical factors, the dynamic influence of service entities, and the average distance of customers, it is determined whether the customer clusters and service entities within the target range are considered as a fusion cluster.

[0009] Optionally, it also includes: The overlapping region is updated to the target range of the fusion cluster.

[0010] Optionally, before determining the service entity that matches the customer's behavioral preferences from the service entities of the fusion cluster, the method further includes: Based on the offline historical behavior data of the customer clusters in the fusion cluster, the visit status of multiple customers in the customer cluster and any service entity is determined; different visit statuses correspond to different association radii; the better the visit status, the larger the association radius. With different association radii, service entities for customer behavior preference analysis are obtained from the service entities of the fusion cluster.

[0011] Optionally, after pushing information to a customer cluster within the same fusion cluster based on a service entity within the same fusion cluster, the method further includes: Collect the push response behavior of the customer clusters in the same fusion cluster to the information push; Based on the push response behavior and the service characteristics corresponding to the service entities in the same fusion cluster, the behavioral stability of the customer cluster in the same fusion cluster is determined; If the stability level does not meet the service matching requirements, then the customer clusters and service entities belonging to the same fusion cluster are re-determined.

[0012] On one hand, embodiments of this application provide an information push device, which includes: The acquisition module is used to obtain customer characteristics of multiple customer clusters through private domain data and service characteristics of multiple service entities through public domain data. The customer characteristics of any customer cluster include customer geographic location and group behavior characteristics; the service characteristics of any service entity include service address location, basic service attributes, and dynamic service attributes. The determination module is used to determine customer clusters and service entities belonging to the same fusion cluster from multiple customer clusters and multiple service entities; wherein, any fusion cluster is determined by the customer density within the cluster meeting a first preset condition and / or the consistency of customer behavior within the cluster meeting a second preset condition; any fusion cluster is determined based on a behavior consistency factor, the dynamic influence of service entities, and the average distance of customers; the dynamic influence is determined based on service basic attributes and service dynamic attributes; The push module is used to push information to customer clusters within the same converged cluster based on service entities within the same converged cluster.

[0013] Optionally, the determining module is further configured to: For any service entity in the converged cluster, determine the overlapping area between customers and service entities in the same converged cluster; Based on the business scenario corresponding to the information push, obtain customer characteristics in the customer cluster corresponding to the business scenario and service characteristics in the service entity corresponding to the business scenario; The degree of association of the service entity in the fusion cluster under the business scenario is determined by the overlapping area, customer characteristics and service characteristics corresponding to the business scenario; Service entities that meet the correlation requirements are included as service entities in the fusion cluster.

[0014] Optionally, the push module is specifically used for: Based on the historical behavior data of customers in the fusion cluster, service entities that match the customer's behavioral preferences are determined from the service entities of the fusion cluster; Push service information from service entities that match customer behavior preferences to the corresponding customers.

[0015] Optionally, the determining module is specifically used for: Based on at least one customer cluster within the target range, determine whether the customer density within the target range meets a first preset condition and / or whether the consistency of customer behavior within the target range meets a second preset condition; If the conditions are met, then the center point is determined based on the weights of the service entities within the target range, and the average distance of customers within the target range is determined based on the center point. Based on behavioral consistency factors determined by customer behavior consistency, geographical factors, the dynamic influence of service entities, and the average distance of customers, it is determined whether the customer clusters and service entities within the target range are considered as a fusion cluster.

[0016] Optionally, the determining module is further configured to: The overlapping region is updated to the target range of the fusion cluster.

[0017] Optionally, the determining module is further configured to: Based on the offline historical behavior data of the customer clusters in the fusion cluster, the visit status of multiple customers in the customer cluster and any service entity is determined; different visit statuses correspond to different association radii; the better the visit status, the larger the association radius. With different association radii, service entities for customer behavior preference analysis are obtained from the service entities of the fusion cluster.

[0018] Optionally, the push module is further configured to: Collect the push response behavior of the customer clusters in the same fusion cluster to the information push; Based on the push response behavior and the service characteristics corresponding to the service entities in the same fusion cluster, the behavioral stability of the customer cluster in the same fusion cluster is determined; If the stability level does not meet the service matching requirements, then the customer clusters and service entities belonging to the same fusion cluster are re-determined.

[0019] On one hand, embodiments of this application provide a computer device, including: Memory, used to store program instructions; The processor is used to call the program instructions stored in the memory and execute the steps of the above-mentioned information push method according to the obtained program.

[0020] On the one hand, embodiments of this application provide a computer-readable storage medium storing a computer program executable by a computer device, which, when run on the computer device, causes the computer to perform the steps of the above-described information push method.

[0021] On one hand, embodiments of this application provide a computer program product, including a computer program stored on a computer-readable storage medium. The computer program includes program instructions, which, when executed by a computer device, cause the computer device to perform the steps of the above-described information push method.

[0022] In this embodiment, customer characteristics of multiple customer clusters are obtained through private domain data, and service characteristics of multiple service entities are obtained through public domain data. The customer characteristics of any customer cluster include customer geographic location and group behavior characteristics. The service characteristics of any service entity include service address location, basic service attributes, and dynamic service attributes. From multiple customer clusters and multiple service entities, customer clusters and service entities belonging to the same fusion cluster are determined. Any fusion cluster is defined as one where the customer density within the cluster meets a first preset condition and / or the consistency of customer behavior within the cluster meets a second preset condition. Any fusion cluster is determined based on a behavior consistency factor, the dynamic influence of the service entity, and the average distance of the customer. The dynamic influence is determined based on the basic service attributes and the dynamic service attributes. Information is pushed to customer clusters within the same fusion cluster based on service entities within the same fusion cluster. This associates customer clusters and service entities with the same fusion cluster from multiple dimensions. Pushing information to customer clusters within the same fusion cluster has higher accuracy and does not rely on individual customer IDs but is based on customer geographic location, solving the problem of information not being pushed due to missing customer IDs. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 A schematic diagram of a system architecture provided in an embodiment of this application; Figure 2 A flowchart illustrating an information push method provided in an embodiment of this application; Figure 3 A flowchart illustrating an information push method provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of an information push device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0025] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0026] It should be noted that the brief descriptions of terms in this application are only for the convenience of understanding the embodiments described below, and are not intended to limit the embodiments of this application. Unless otherwise stated, these terms should be understood in their ordinary and common meaning.

[0027] The terms "first," "second," "third," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar or related objects or entities, and do not necessarily imply a specific order or sequence, unless otherwise specified. It should be understood that such terms are interchangeable where appropriate.

[0028] The terms “comprising” and “having”, and any variations thereof, are intended to cover but not exclude inclusion, for example, a product or device that includes a range of components is not necessarily limited to all of the components that are clearly listed, but may include other components that are not clearly listed or that are inherent to such product or device.

[0029] The term "module" refers to any known or subsequently developed hardware, software, firmware, artificial intelligence, fuzzy logic, or combination of hardware and / or software code that is capable of performing the functions associated with that element.

[0030] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0031] The following is a brief introduction to the system architecture diagram applicable to the technical solutions of the embodiments of this application. It should be noted that the process described below is only for illustrating the embodiments of this application and is not intended to limit the scope of the application.

[0032] refer to Figure 1This is a system architecture diagram applicable to the embodiments of this application. The system architecture includes at least terminal device 101 and server 102. The number of terminal devices 101 can be one or more, and the number of servers 102 can also be one or more. This application does not specifically limit the number of terminal devices 101 and servers 102.

[0033] Terminal device 101 is pre-installed with an application that has information push functionality. This application can be a client application, a web application, a mini-program application, etc. Terminal device 101 can be a smartphone, tablet computer, laptop computer, desktop computer, smart home appliance, smart voice interaction device, smart in-vehicle device, etc., but is not limited to these.

[0034] Server 102 is the backend server of the application. Server 102 can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms, but it is not limited to these.

[0035] It should be noted that the method in this embodiment can be executed by the terminal device 101 or the server 102 alone, or by the terminal device 101 and the server 102 together.

[0036] In this embodiment, the terminal device 101 and the server 102 can communicate directly or indirectly through one or more networks. The network can be a wired network or a wireless network; for example, the wireless network can be a mobile cellular network or a Wireless-Fidelity (WIFI) network, or other possible networks. This embodiment does not limit the types of networks used.

[0037] The following is based on Figure 1 The system architecture diagram shown illustrates the flow of an information push method provided in this application embodiment. This method's flow can be... Figure 1 The execution can be performed by the terminal device 101 shown, or by the server 102, or by interaction between the terminal device 101 and the server 102, as shown. Figure 2 As shown, it includes the following steps: Step 201: Obtain customer characteristics of multiple customer clusters through private domain data and service characteristics of multiple service entities through public domain data.

[0038] In the embodiments of this application, the customer characteristics of any customer cluster include customer geographical location and group behavior characteristics; the service characteristics of any service entity include service address location, basic service attributes, and dynamic service attributes.

[0039] Specifically, basic information of existing customers is collected from the private domain data banking system. This information includes, but is not limited to, name, age, gender, occupation, asset size, and branch where the account was opened. Data cleaning is performed on each customer's basic information to remove duplicates; for example, if a customer has multiple accounts, these accounts are merged. Missing fields in the customer's basic information are filled in, such as inferring the customer's occupation and gender based on historical transaction records, thus forming a structured customer information table.

[0040] Each customer provides a permanent address during account opening or trading. A reverse geocoding engine (such as a map software API) converts this address into latitude and longitude coordinates. If no permanent address is provided, the customer's login IP address is converted into latitude and longitude coordinates, which are then stored in the customer's geographic information table. For example, if a customer's provided address is "No. XX, XX Road, XX District, XX City," it will be converted into latitude and longitude coordinates (30.123456°N, 120.654321°E) using a map software API, retaining six decimal places.

[0041] Online behavioral data of customers is collected through the bank's app or website. This historical online behavioral data includes: time spent on the app's homepage, browsing history of wealth management products, number of clicks on loan application portals, and number of loan inquiries. Offline behavioral data is also collected by deploying a tracking SDK at the bank's physical branches. This offline behavioral data includes: branch visit frequency, types of counter services (such as deposits and transfers), content of customer manager inquiries, wealth management transaction data, and loan transaction data. The online and offline historical behavioral data are then correlated with customer behavior to create a time-series behavioral profile. For example, the app / branch tracking might collect the number of loan inquiries in the past 15 days and the duration of wealth management product browsing in the past 30 days.

[0042] In summary, for each customer, there is basic information, geographical location, and time-series behavioral data.

[0043] The geographic locations (latitude and longitude coordinates) and temporal behavioral data of multiple customers are converted into feature vectors and input into the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) density clustering algorithm. This density clustering algorithm determines core points, boundary points, and isolated points based on the geographic locations of multiple customers. A core point indicates that the number of customers within an initial radius of that point exceeds a first threshold; a boundary point indicates that the point is located around a core point but does not meet the core point requirement; and an isolated point indicates an isolated customer. The final output is a customer cluster. Each customer cluster possesses customer geographic locations (i.e., center location and geographic boundaries), group behavioral characteristics, and customer attributes; the group behavioral characteristics are obtained based on the temporal behavioral data of multiple customers.

[0044] For example, among multiple customers located in the "Wensan Road Science and Technology Park", there are 30 customers within a 500-meter radius, with a behavioral similarity of ≥70% and 80% of them having inquired about loans in the past 15 days. These 30 customers constitute a customer cluster, and the area where this customer cluster is located is an irregular shape formed by the geographical locations of the multiple customers. The customer attributes of this cluster may include: an average age of 32 years old, 70% of the customers being employees in the technology industry, and an average monthly asset turnover of 50,000 yuan. The central location of the area where this customer cluster is located is at the latitude and longitude coordinates (30.274, 120.005), and the geographical boundaries of the area are 120.002-120.008° east longitude and 30.271-30.277° north latitude.

[0045] The density clustering algorithm can be either sorting points to identify cluster structure (OPTICS) or hierarchical density clustering (HDBSCAN). This application does not make any specific restrictions on this.

[0046] Specifically, data on all service entities near the current city or customer cluster is collected through third-party data service interfaces. This includes building data, such as the latitude and longitude, construction year, unit type, average price, and occupancy rate of any given building; business district data, such as the latitude and longitude, business type (e.g., catering, retail), average daily foot traffic, and number of merchants for any given business district; and transportation hub data, such as the latitude and longitude, average daily passenger flow, and connecting routes for subway / bus stations. This data undergoes preprocessing operations such as data cleaning and standardization. The preprocessed data is then stored using "InfluxDB + PostgreSQL + PostGIS." "InfluxDB + PostgreSQL + PostGIS" is a collaborative technology stack of "time-series data + structured data + spatial data," used to solve complex data scenarios that simultaneously process time-series data, structured business information, and geospatial location. It also supports spatial and dynamic queries, such as querying "office buildings within 1 kilometer of the customer cluster" or querying "business districts with ≥10,000 foot traffic in the past 2 hours." This results in a public domain data warehouse covering all service entities across the entire city.

[0047] For example, in preprocessing, the data is standardized, such as converting "pedestrian traffic" into "average daily visitors"; data cleaning is performed, such as removing abnormal data where the average daily visitors in the business district are 0.

[0048] Next, relevant data for each service entity is obtained from the public domain data. Based on this data, Points of Interest (PIO) analysis is performed on the service entities to extract their corresponding service characteristics. These characteristics include: service location, basic service attributes, and dynamic service attributes. Basic service attributes include: transportation convenience and business level; dynamic service attributes include: business type, traffic volume, transportation mode, entity size, and entity function, but are not limited to these. For example, the "West Lake Cultural Square Business District" is analyzed using POI as "restaurant cluster + high foot traffic + subway connection," and specific indicators are associated, such as 40% of the businesses in this business district being restaurants and 100% connection to Metro Line 1. Based on the PIO analysis results, a static influence score is assigned to the service entity. Different static influence baseline values ​​(IO) are assigned according to the service entity's business attributes (such as business type, traffic volume, transportation mode, entity size, and entity function). The baseline IO is first determined by business type, then fine-tuned by "entity size," and finally supplemented by "entity function," without involving any dynamic indicators. For example, if the service entity is a high-speed rail station with a business type IO=100, and the high-speed rail station is a top-tier station, then the entity size IO=10. Since there are no scarce entity functions, there is no supplement. Therefore, the static influence IO of the high-speed rail station is 100+10=110. If the service entity is a general business district with a business type IO=60, and the general level of the general business district is no supplement, and the entity function of the general business district is cultural and tourism integration, then the entity function IO=10. Therefore, the static influence IO of the general business district is 60+10=70.

[0049] For each service entity, a service entity profile is constructed based on the basic service attributes and dynamic service attributes. For example, if the service entity is a business district, then the static (business level × 0.4 + transportation convenience × 0.3) + dynamic (pedestrian traffic × 0.4 + consumption frequency × 0.3) weighted sum is calculated. Based on the summation result, the service entity is tagged, such as "business district with high demand for enterprise services".

[0050] For another example, dynamic influence can be determined based on traffic volume, transportation mode, and consumption level. For offline marketing scenarios in a typical business district, with traffic volume weighted at 0.4, consumption level at 0.3, and transportation mode at 0.3, the dynamic influence of a business district could be: Traffic Volume 80 points (out of 100) × 0.4 + Consumption Level 70 points × 0.3 + Transportation Convenience 90 points × 0.3 = 79 points. Based on this dynamic influence, the business district could be labeled as a "high-potential business district." These labels, obtained from calculating dynamic influence, also represent the dynamic attributes of the service entity.

[0051] Step 202: From multiple customer clusters and multiple service entities, determine the customer clusters and service entities that belong to the same converged cluster.

[0052] In the embodiments of this application, any fusion cluster is defined as having a customer density within the cluster that meets a first preset condition and / or a customer behavior consistency within the cluster that meets a second preset condition; any fusion cluster is determined based on a behavior consistency factor, the dynamic influence of the service entity, and the average distance of the customer; the dynamic influence is determined based on the service's basic attributes and service dynamic attributes.

[0053] Specifically, a density-based clustering algorithm will be used to merge the geographical location and group behavior characteristics of customer clusters with the service address, basic service attributes, and dynamic service attributes of service entities into a new dataset, and the features will be standardized. During the clustering process, if the customer density within the clustering result meets a first preset condition, such as the number of customers in the corresponding area reaching 800, then the first preset condition is considered met. If the behavioral similarity between customers in the clustering result meets a second preset condition, such as the proportion of similar business interactions among customers in the cluster in the past 15 days being greater than 70%, then the second preset condition is considered met, and clustering can be stopped, with the current clustering result being considered as a fused cluster. The first and second preset conditions are set according to actual circumstances, and this application does not specify them in detail.

[0054] For example, the geographical location of the "Technology Park Customer Cluster" is (30.274, 120.005), and the service address location of the public domain geographic entity center point—the "Business District"—is (30.275, 120.006). These two are merged into a new dataset, and density clustering with dynamic threshold adjustment is used: the silhouette coefficient is calculated, and when the coefficient approaches 1, the optimal initial EPS is determined. For example, when the target neighborhood radius (EPS) = 800 meters, the silhouette coefficient = 0.85, which is the optimal value. During the iterative optimization process of density clustering, the gradient boosting tree (GBDT) model is used to iteratively optimize EPS. The distribution density of the customer cluster in the target neighborhood, the total number of service entities in the target neighborhood, and the consistency of customer behavior within the customer cluster are used as inputs to the clustering algorithm. Following pre-set adjustment rules, the adjusted EPS value is output. The pre-defined adjustment rules can be as follows: if the target neighborhood has a high customer density (e.g., ≥500 customers per square kilometer) or high consistency in customer behavior (e.g., ≥70% of customers in the cluster have similar business interactions in the past 15 days), the EPS will be automatically reduced, for example, to 600 meters, to avoid excessive differences in features within the cluster; if the regional customer density is low or the behavior consistency is low, the EPS will be expanded accordingly, for example, to 800-1000 meters, to ensure cluster integrity.

[0055] Furthermore, the central location bias of the clusters obtained from clustering is determined based on the transportation mode and traffic volume of any service entity. For example, if an office building has a transportation mode score of 60 and a traffic volume score of 90, then 0.6 × 60 + 0.4 × 90 = 72 is used as the geographical location score for the office building. If a business district has a transportation mode score of 90 and a traffic volume score of 80, then 0.6 × 90 + 0.4 × 80 = 86 is used as the geographical location score for the business district. Since 86 > 72, the central location of the clusters is biased towards the business district. Specifically, this can be achieved by using the latitude and longitude coordinates corresponding to the central location of the business district. Location weight 1 + latitude and longitude coordinates of the center location of the office building area Position weight 2 is used as the center position of the fusion cluster, where position weight 1 is greater than position weight 2.

[0056] In some embodiments, for any service entity in the fusion cluster, the overlapping area between customers and service entities in the same fusion cluster is determined; based on the business scenario corresponding to the information push, customer characteristics in the customer cluster corresponding to the business scenario and service characteristics in the service entity corresponding to the business scenario are obtained; through the overlapping area, customer characteristics and service characteristics corresponding to the business scenario, the relevance of the service entity in the fusion cluster under the business scenario is determined; and the service entity whose relevance meets the requirements is taken as the service entity in the fusion cluster.

[0057] Specifically, customer characteristics corresponding to the business scenario in the customer cluster include: customer assets, customer occupation, etc., without specific limitations; service characteristics corresponding to the business scenario in the service entity include: consumption level, historical conversion rate, etc., without specific limitations, where the historical conversion rate represents the conversion rate of customers after the service entity has pushed information in the past.

[0058] The overlapping area represents the area of ​​overlap between the geographical location of any customer cluster and the service address of any service entity. The geographical fit between the customer cluster and the service entity is determined based on the sum of the overlapping area and the area of ​​the geographical location of the customer cluster and the service address of the service entity. The correlation between the customer cluster and the service entity is then determined based on the fit, customer characteristics, and service characteristics.

[0059] For example, the first step is to determine the sum of the areas of the geographical locations of customers in a customer cluster and the service addresses of any service entity. The geographical boundaries of the customer geographical locations and the geographical boundaries of the service addresses of service entities are clustered to form irregular polygons. For example, the geographical boundary of the "Wensan Road Technology Enterprise Main Customer Cluster" is generated by fitting the latitude and longitude coordinates of all customers in the cluster, with a boundary accuracy of 10 meters. The geographical boundary of the service entity is an irregular polygon formed by combining the basic attributes and dynamic attributes of the service. For example, the geographical boundary of the "Wensan Road Science and Technology Innovation Business District" is generated by fitting the latitude and longitude coordinates and dynamic influence radiation range of each merchant in the business district, ensuring that both types of boundaries can reflect the actual spatial distribution and functional coverage of the entity, rather than a simple administrative division. Next, the intersection area (overlapping area) and union area of ​​customer clusters and service entities are calculated using spatial overlay analysis tools to obtain the geographical fit. For example, the area of ​​the "Wensan Road Technology Enterprise Main Customer Cluster" is 10 square kilometers, the area of ​​the "Wensan Road Science and Technology Innovation Business Circle" is 8 square kilometers, the overlap area is 6 square kilometers, and the geographical fit is 6 / (10+8-6)=50%. Assuming the current information push business scenario is a wealth management promotion scenario, we can set the customer asset weight in the customer characteristics to 0.3, the matching degree weight to 0.2, the historical conversion rate (e.g., the historical conversion rate of wealth management business) in the service characteristics to 0.3, and the consumption level weight in the service characteristics to 0.2. Finally, we calculate the correlation. Assuming the customer cluster is a high-asset customer cluster (asset size score 90) and the service entity is a high-end wealth management business district (wealth management response rate score 85, consumption level score 80), then the correlation between the customer cluster and the service entity = 90×0.3+85×0.3+50×0.2+80×0.2=27+25.5+10+16=78.5 points.

[0060] For different levels of relevance, the degree of information binding between different service entities and customer clusters is determined. When the relevance score is ≥60, it is judged as "strong relevance," and the service entity's scenario-based tags and resources are accurately assigned to the customer cluster. For example, if the high-asset customer cluster has a relevance score of 78.5 with the loan business district, the customer cluster is tagged as "high-conversion potential customer group," and information on bank wealth management service outlets within the business district (such as outlet address and dedicated manager contact information) is simultaneously linked. Activities of business district outlets are prioritized for this customer cluster. When the relevance score is between 40 and 60, it is judged as "weak relevance," and only the basic attributes of the service entity (such as business district type) are used as supplementary information for the customer cluster, but priority recommendation is not triggered. When the relevance score is <40, it is judged as "no relevance," and the service entity is excluded to avoid interference from invalid data.

[0061] In this embodiment, by combining overlapping areas of the geographical dimension, customer characteristics of customer clusters, and service characteristics of service entities, the correlation between customer clusters and service entities is determined. This breaks through the limitations of a single geographical dimension, combines the geographical dimension with the business scenarios of service entities, and improves the accuracy of subsequent information push.

[0062] In some embodiments, the overlapping region is updated to the target range of the fusion cluster.

[0063] Specifically, the overlapping area can be used as the location of the fusion cluster. Alternatively, customer clusters and service entities can be further filtered based on the overlapping area to determine the final fusion cluster.

[0064] In some embodiments, based on at least one customer cluster within the target range, it is determined whether the customer density within the target range meets a first preset condition and / or whether the consistency of customer behavior within the target range meets a second preset condition; if so, a center point is determined based on the weight of the service entity within the target range, and the average distance of customers within the target range is determined based on the center point; based on the behavioral consistency factor, geographical factor, dynamic influence of the service entity, and the average distance of the customer determined by the consistency of customer behavior, it is determined whether the customer cluster and service entity within the target range are a fusion cluster.

[0065] Specifically, the customer clusters and service entities within the target area are further clustered. The neighborhood radius is dynamically determined from the inside out within the target area, but the process stops when the customer density within the range defined by this neighborhood radius meets a first preset condition or the consistency of the behavior of each customer meets a second preset condition. Simultaneously, the location weights of service entities are determined based on their transportation modes and traffic volumes. Based on the location weights of each service entity and the latitude and longitude coordinates corresponding to its central location, the center location of the target area is determined. The newly determined center location must not exceed a distance preset condition from the average distance of customers in the fused cluster. The first distance preset condition can be set according to actual conditions, and this application does not impose specific limitations on it.

[0066] In some embodiments, based on the offline historical behavior data of the customer cluster in the fusion cluster, the visit status of multiple customers in the customer cluster and any service entity is determined; different visit statuses correspond to different association radii; the better the visit status, the larger the association radius; with different association radii, service entities for customer behavior preference analysis are obtained from the service entities of the fusion cluster.

[0067] Specifically, based on the offline historical behavior data of customer groups, the percentage of visits to different service entities in the customer group over a period of time is determined. Different association radii are set for different visit percentages. The higher the visit percentage, the larger the association radius, which means that there are more customers within the scope of the association radius.

[0068] For any service entity falling within the range defined by the association radius, the service matching degree between the customer cluster and the service entity is determined based on the customer's offline historical behavior data and the service behavior data or service characteristics corresponding to the service entity. This is equivalent to determining whether the behavioral preferences of the customer cluster correspond to the services that the service entity can provide.

[0069] Taking the offline historical behavior data of the past three months as an example, in the past three months, the number of visits to office building service entities accounted for 70% of the total visits, while the number of visits to community supermarket service entities accounted for only 15%. Differentiated association radii were set according to the visit ratio: for high-preference entity types with a visit ratio of ≥60%, a 1.2-kilometer association radius was set, which can cover the area of ​​customers' daily high-frequency activities; for medium-preference entity types with a visit ratio of 30%-60%, a 0.8-kilometer association radius was set; and for low-preference entity types with a visit ratio of <30%, a 0.5-kilometer association radius was set to ensure that the radius range is highly matched with the actual behavioral preferences of customers.

[0070] Taking science and technology innovation services and loan consultations from the offline historical behavior data of a customer cluster as examples, a weighted sum is calculated based on the offline historical behavior data of the customer cluster and the service characteristics corresponding to the service entities to determine the service matching degree. If the customer cluster's loan consultation frequency score is 85, the customer cluster's science and technology innovation service query score is 90, the score for the proportion of physical science and technology innovation merchants is 80, and the score for physical loan cooperation records is 75, then the service matching degree is 85×0.4+90×0.3+80×0.2+75×0.1=34+27+16+7.5=84.5 points. For multiple service entities, the service matching degree between the customer cluster and the service entities is arranged in descending order. At least one service entity at the top of the list is selected as the service entity in the same fusion cluster as the customer cluster. Such a service entity is also the service entity that conforms to the customer's behavioral preferences.

[0071] It should be noted that the implementation examples in step 202 that are highly correlated with customer clusters and have a high degree of service matching can be executed in parallel or in a progressive manner, that is, there is no specific limitation on the execution order of different filtering methods.

[0072] In this embodiment, by combining the customer characteristics of the customer cluster and the service characteristics of the service entity through overlapping regions, and then by using the customer's offline historical behavior data, the service characteristics of the service entity, or the service data corresponding to the offline historical behavior data, the area of ​​the final fusion cluster is further reduced, and the number of customers is also simplified, but the customer density in the fusion cluster is increased. Furthermore, the behavioral preferences of customers in these fusion clusters have a high degree of matching with the service characteristics of the service entity, thereby greatly improving the accuracy of subsequent information push.

[0073] Step 203: Push information to customer clusters in the same converged cluster based on service entities in the same converged cluster.

[0074] In some embodiments, based on the historical behavior data of customers in the fusion cluster, service entities that match the customer's behavioral preferences are determined from the service entities in the fusion cluster; and the service information of the service entities that match the customer's behavioral preferences is pushed to the corresponding customers.

[0075] Specifically, this could involve a service entity within the same converged cluster conducting an event and pushing the event information to the customer cluster within that cluster. Alternatively, a service entity could tailor event information based on the behavioral preferences of the customer cluster. In addition to pushing event information, service entities can also push information such as past client case studies and service hours to the customer cluster. For example, the service time slot of a dedicated account manager on-site every Wednesday afternoon can be pushed to customers within the customer cluster.

[0076] In this embodiment, by determining the service matching between the service entity and the customer cluster, personalized service packages are pushed to customers within the customer cluster, and consultation time or consultation entry is provided. This achieves a closed-loop association of "customer behavior preferences - service characteristics of service entities - service resources", avoiding the drawback of simply filtering locations without matching needs in conventional radius association.

[0077] In some embodiments, the push response behavior of customer clusters in the same fusion cluster to information push is collected; based on the push response behavior and the service characteristics corresponding to the service entities in the same fusion cluster, the behavior stability of customer clusters in the same fusion cluster is determined; if the stability does not meet the service matching requirements, the customer clusters and service entities belonging to the same fusion cluster are re-determined.

[0078] Specifically, after determining the fusion cluster, the fusion cluster can be evaluated in terms of clustering quality, accuracy of association, and stability, and then further optimization can be carried out.

[0079] The clustering quality of the fusion clusters is determined by the following formula (1): …(1) in, The standard profile coefficient represents the spatial clustering compactness of the fusion cluster before filtering based on relevance and service matching, and its value ranges from [-1, 1]. The improved profile coefficient represents the spatial clustering compactness of the final determined fusion cluster; Sim_behavior represents the similarity of customer behavior within the cluster, such as the proportion of similar business interactions among customers within the cluster in the past 15 days (value range [0,1]); β represents the dynamic influence decay coefficient of the service entity; and d represents the average geographical distance of customers within the cluster. Different improved profile coefficients correspond to different clustering qualities. For example, SC_adapted≥0.8 is excellent, indicating high clustering quality of the fusion cluster, with concentrated geographical distribution and consistent behavior among customers within the cluster; 0.6≤SC_adapted<0.8 is good; and SC_adapted<0.8 is poor, requiring optimization or re-clustering.

[0080] For example, before filtering based on relevance and service matching, the standard profile coefficient SC_standard = 0.75 for customer clusters and multiple service entities, the similarity of customer behavior within the cluster Sim_behavior = 0.7 (70% of loan inquiries in the past 15 days), and the average distance within the cluster d = 800 meters, then at this point... = 0.619, indicating that the consistency of behavior within the customer cluster needs improvement. After filtering by relevance, removing service entities or customers with low relevance (such as weakly related community business districts and non-technology industry customers), the remaining service entities and customer clusters show the following improvements: SC_standard increases to 0.82, intra-cluster behavioral similarity Sim_behavior increases to 0.85 (loan inquiries accounted for 85% in the past 15 days), and the average intra-cluster distance d decreases to 600 meters. = 0.713>0.619, indicating an improvement in cluster quality.

[0081] Additionally, if the improved profile coefficient decreases instead of increasing after removing low-relevance service entities or their corresponding customers, the weight of the clustering process's neighborhood radius or the offline historical behavior data of the customer clusters can be reduced.

[0082] Furthermore, the correlation between each service entity in the final merged cluster and the customer cluster can be compared to determine the priority order of information push.

[0083] For example, for the science and technology innovation business district within the integrated cluster, if the customer cluster's loan inquiry frequency score (C_score) is 90, the business district's loan cooperation score (E_value) is 85, the regional coverage overlap rate (Overlap_rate) is 53.8%, and the business district's foot traffic score (E_dynamic) is 88, then the correlation is 0.4×90 + 0.3×85 + 0.2×53.8 + 0.1×88 = 81.06. For the community business district within the same integrated cluster, if the customer cluster's loan inquiry frequency score (C_score) is 90, the community business district's loan cooperation score (E_value) is 30, the regional coverage overlap rate (Overlap_rate) is 13.3%, and the community business district's foot traffic score (E_dynamic) is 20, then the correlation is 0.4×90 + 0.3×30 + 0.2×13.3 + 0.1×20 == 49.66 < 81.06. Therefore, information about the science and technology innovation business district will be prioritized for recommendation to the customer cluster within the integrated cluster. If the correlation between any service entity in the fusion cluster and the customer cluster is less than the pre-specified value (e.g., 50), the correlation can be recalculated by adjusting the weight of each dimension.

[0084] Furthermore, data on the push response behavior of customer clusters in the same fusion cluster to information push is periodically acquired, the divergence is calculated, and the stability is then determined. The specific calculation formula for the stability is shown in the following formula (2): …………………(2) Wherein, Pt is the service matching degree between the service entity in the fusion cluster and the customer cluster in the t-th period. For example, the service matching degree distribution of the three science and technology office buildings in the same fusion cluster is [89.4, 85.2, 82.6]); P{t+1} is the service matching degree between the information push response behavior of the customer cluster and the service characteristics of the service entity in the fusion cluster. For example, the service matching degree of the three science and technology office buildings at this time is [88.8, 86.1, 83.3]; JS_divergence is the JS divergence (the value ranges from [0,1], and the smaller the value, the closer the distribution).

[0085] When the stability level is greater than or equal to 0.95, it indicates that the customer cluster is highly stable, meaning that the behavioral preferences of the customer cluster are well-matched with the service characteristics of the service entities in the fusion cluster over a long period of time. When the stability level is less than or equal to 0.95 but greater than or equal to 0.9, it indicates that the customer cluster is moderately stable. When the stability level is less than 0.9, it indicates that the customer cluster is not highly stable, meaning that the behavioral preferences of the customer cluster are not well-matched with the service characteristics of the service entities in the fusion cluster. In this case, the fusion cluster is re-determined according to the technical solution of this application. During the process of re-determining the fusion cluster, the historical market of the customer's offline historical behavior data can be shortened, for example, from three months to two months. The weight corresponding to the conversion rate of the service entity can also be increased, and service entities with high service matching degree can be re-selected.

[0086] To better explain the embodiments of this application, the following describes an information push method provided by the embodiments of this application in conjunction with a practical scenario. The process of this method is as follows: Figure 1 The server shown executes, as follows: Figure 3 As shown: The big data platform includes both internal private domain data and external public domain data. Internal private domain data, acquired from the bank's core business systems, includes existing customer data and transaction data. This customer data comprises customer information and behavior data, while transaction data includes financial transactions and branch operation data. Based on this data, an initial customer cluster (i.e., a private domain customer profile data model) is formed. External data services, real estate, business districts, and geographic locations are used to determine the geographic data characteristics and business district data profiles of any entity, i.e., a public domain geographic feature model. Through big data mining and a rule engine, customer clusters and service entities are filtered and merged according to adaptive rules of density clustering and customer grouping classification to obtain a fused feature model (i.e., the final fused cluster). This fused feature model is then evaluated and optimized. In the data application phase, offline marketing and business opportunity promotion are conducted for customer clusters based on service entities within the same fused cluster.

[0087] Based on the same technical concept, this application provides a schematic diagram of the structure of an information push device, such as... Figure 4 As shown, the information push device 400 includes: The acquisition module 401 is used to obtain customer characteristics of multiple customer clusters through private domain data and service characteristics of multiple service entities through public domain data; the customer characteristics of any customer cluster include customer geographical location and group behavior characteristics; the service characteristics of any service entity include service address location, basic service attributes and dynamic service attributes. The determining module 402 is used to determine customer clusters and service entities belonging to the same fusion cluster from multiple customer clusters and multiple service entities; wherein, any fusion cluster is determined by the customer density within the cluster meeting a first preset condition and / or the consistency of customer behavior within the cluster meeting a second preset condition; any fusion cluster is determined based on a behavior consistency factor, the dynamic influence of service entities, and the average distance of customers; the dynamic influence is determined based on service basic attributes and service dynamic attributes; The push module 403 is used to push information to customer clusters in the same converged cluster based on service entities in the same converged cluster.

[0088] Optionally, the determining module 402 is further configured to: For any service entity in the converged cluster, determine the overlapping area between customers and service entities in the same converged cluster; Based on the business scenario corresponding to the information push, obtain customer characteristics in the customer cluster corresponding to the business scenario and service characteristics in the service entity corresponding to the business scenario; The degree of association of the service entity in the fusion cluster under the business scenario is determined by the overlapping area, customer characteristics and service characteristics corresponding to the business scenario; Service entities that meet the correlation requirements are included as service entities in the fusion cluster.

[0089] Optionally, the push module 403 is specifically used for: Based on the historical behavior data of customers in the fusion cluster, service entities that match the customer's behavioral preferences are determined from the service entities of the fusion cluster; Push service information from service entities that match customer behavior preferences to the corresponding customers.

[0090] Optionally, the determining module 402 is specifically used for: Based on at least one customer cluster within the target range, determine whether the customer density within the target range meets a first preset condition and / or whether the consistency of customer behavior within the target range meets a second preset condition; If the conditions are met, then the center point is determined based on the weights of the service entities within the target range, and the average distance of customers within the target range is determined based on the center point. Based on behavioral consistency factors determined by customer behavior consistency, geographical factors, the dynamic influence of service entities, and the average distance of customers, it is determined whether the customer clusters and service entities within the target range are considered as a fusion cluster.

[0091] Optionally, the determining module 402 is further configured to: The overlapping region is updated to the target range of the fusion cluster.

[0092] Optionally, the determining module is further configured to: Based on the offline historical behavior data of the customer clusters in the fusion cluster, the visit status of multiple customers in the customer cluster and any service entity is determined; different visit statuses correspond to different association radii; the better the visit status, the larger the association radius. With different association radii, service entities for customer behavior preference analysis are obtained from the service entities of the fusion cluster.

[0093] Optionally, the push module 403 is further configured to: Collect the push response behavior of the customer clusters in the same fusion cluster to the information push; Based on the push response behavior and the service characteristics corresponding to the service entities in the same fusion cluster, the behavioral stability of the customer cluster in the same fusion cluster is determined; If the stability level does not meet the service matching requirements, then the customer clusters and service entities belonging to the same fusion cluster are re-determined.

[0094] In this embodiment, customer characteristics of multiple customer clusters are obtained through private domain data, and service characteristics of multiple service entities are obtained through public domain data. The customer characteristics of any customer cluster include customer geographic location and group behavior characteristics. The service characteristics of any service entity include service address location, basic service attributes, and dynamic service attributes. From multiple customer clusters and multiple service entities, customer clusters and service entities belonging to the same fusion cluster are determined. Any fusion cluster is defined as one where the customer density within the cluster meets a first preset condition and / or the consistency of customer behavior within the cluster meets a second preset condition. Any fusion cluster is determined based on a behavior consistency factor, the dynamic influence of the service entity, and the average distance of the customer. The dynamic influence is determined based on the basic service attributes and the dynamic service attributes. Information is pushed to customer clusters within the same fusion cluster based on service entities within the same fusion cluster. This associates customer clusters and service entities with the same fusion cluster from multiple dimensions. Pushing information to customer clusters within the same fusion cluster has higher accuracy and does not rely on individual customer IDs but is based on customer geographic location, solving the problem of information not being pushed due to missing customer IDs.

[0095] Based on the same technical concept, embodiments of this application provide a computer device, which can be... Figure 1 The server shown is as follows: Figure 5 As shown, it includes at least one processor 501 and a memory 502 connected to at least one processor. In this embodiment, the specific linking medium between the processor 501 and the memory 502 is not limited. Figure 5Taking the connection between processor 501 and memory 502 via a bus as an example, the bus can be divided into address bus, data bus, control bus, etc.

[0096] In this embodiment of the application, the memory 502 stores instructions executed by at least one processor 501. By executing the instructions stored in the memory 502, at least one processor 501 can perform the steps of the above-described information push method.

[0097] The processor 501 is the control center of the computer device, capable of connecting various parts of the computer device via various interfaces and lines. It enables information push by running or executing instructions stored in the memory 502 and retrieving data stored in the memory 502. Optionally, the processor 701 may include one or more processing modules. The processor 501 may integrate an application processor and a modem processor. The application processor primarily handles the operating system, user interface, and applications, while the modem processor primarily handles wireless communication. It is understood that the modem processor may not be integrated into the processor 501. In some embodiments, the processor 501 and the memory 502 may be implemented on the same chip; in other embodiments, they may be implemented on separate chips.

[0098] Processor 501 can be a general-purpose processor, such as a central processing unit (CPU), digital signal processor, application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0099] Memory 502, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory 502 may include at least one type of storage medium, such as flash memory, hard disk, multimedia card, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), magnetic memory, magnetic disk, optical disk, etc. Memory 502 can be any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer device, but is not limited thereto. In the embodiments of this application, memory 502 can also be a circuit or any other device capable of implementing storage functions for storing program instructions and / or data.

[0100] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium storing a computer program executable by a computer device. When the program is run on the computer device, it causes the computer device to perform the steps of the above-described information push method.

[0101] Based on the same inventive concept, this application provides a computer program product, including a computer program stored on a computer-readable storage medium. The computer program includes program instructions, which, when executed by a computer device, cause the computer device to perform the steps of the above-described information push method.

[0102] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0104] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0105] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0106] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. An information push method, characterized in that, include: By using private domain data, we can obtain customer characteristics of multiple customer clusters and service characteristics of multiple service entities through public domain data. The customer characteristics of any customer cluster include customer geographic location and group behavior characteristics; The service characteristics of any service entity include service address location, basic service attributes, and dynamic service attributes; From multiple customer clusters and multiple service entities, customer clusters and service entities belonging to the same fusion cluster are determined; wherein, any fusion cluster is defined as having a customer density within the cluster that meets a first preset condition and / or customer behavior consistency within the cluster that meets a second preset condition; any fusion cluster is determined based on a behavior consistency factor, the dynamic influence of service entities, and the average distance of customers; the dynamic influence is determined based on service basic attributes and service dynamic attributes; Information is pushed to customer clusters within the same converged cluster based on service entities within the same converged cluster.

2. The method as described in claim 1, characterized in that, After determining the customer clusters and service entities belonging to the same fusion cluster, the process further includes: For any service entity in the converged cluster, determine the overlapping area between customers and service entities in the same converged cluster; Based on the business scenario corresponding to the information push, obtain customer characteristics in the customer cluster corresponding to the business scenario and service characteristics in the service entity corresponding to the business scenario; The degree of association of the service entity in the fusion cluster under the business scenario is determined by the overlapping area, customer characteristics and service characteristics corresponding to the business scenario; Service entities that meet the correlation requirements are included as service entities in the fusion cluster.

3. The method as described in claim 2, characterized in that, The method of pushing information to customer clusters within the same fusion cluster based on service entities within the same fusion cluster includes: Based on the historical behavior data of customers in the fusion cluster, service entities that match the customer's behavioral preferences are determined from the service entities of the fusion cluster; Push service information from service entities that match customer behavior preferences to the corresponding customers.

4. The method according to any one of claims 1 to 3, characterized in that, From multiple customer clusters and multiple service entities, identify customer clusters and service entities belonging to the same converged cluster, including: Based on at least one customer cluster within the target range, determine whether the customer density within the target range meets a first preset condition and / or whether the consistency of customer behavior within the target range meets a second preset condition; If the conditions are met, then the center point is determined based on the weights of the service entities within the target range, and the average distance of customers within the target range is determined based on the center point. Based on behavioral consistency factors determined by customer behavior consistency, geographical factors, the dynamic influence of service entities, and the average distance of customers, it is determined whether the customer clusters and service entities within the target range are considered as a fusion cluster.

5. The method as described in claim 4, characterized in that, Also includes: The overlapping region is updated to the target range of the fusion cluster.

6. The method as described in claim 3, characterized in that, Before determining the service entities that match the customer's behavioral preferences from the service entities of the fusion cluster, the process also includes: Based on the offline historical behavior data of the customer clusters in the fusion cluster, the visit status of multiple customers in the customer cluster and any service entity is determined; different visit statuses correspond to different association radii; the better the visit status, the larger the association radius. With different association radii, service entities for customer behavior preference analysis are obtained from the service entities of the fusion cluster.

7. The method as described in claim 1, characterized in that, After pushing information to customer clusters within the same fusion cluster based on service entities within the same fusion cluster, the method further includes: Collect the push response behavior of the customer clusters in the same fusion cluster to the information push; Based on the push response behavior and the service characteristics corresponding to the service entities in the same fusion cluster, the behavioral stability of the customer cluster in the same fusion cluster is determined; If the stability level does not meet the service matching requirements, then the customer clusters and service entities belonging to the same fusion cluster are re-determined.

8. A computer device, characterized in that, include: Memory, used to store program instructions; A processor is configured to invoke program instructions stored in the memory and execute the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, It stores a computer program executable by a computer device, which, when run on the computer device, causes the computer device to perform the steps of the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a computer program stored on a computer-readable storage medium, the computer program including program instructions that, when executed by a computer device, cause the computer device to perform the steps of the method as described in any one of claims 1 to 7.

Images