An intelligent CRM customer management method and system supporting cross-platform integration

Abstract

The application discloses an intelligent CRM customer management method and system supporting cross-platform integration, comprising the following steps: obtaining customer information of the same customer in at least two different network platforms; predicting first preference information and second preference information of the customer by using a trained federated learning model according to the customer information; selecting a service carrier consistent with the preference of the customer according to the first preference information, and displaying the second preference information to the service carrier; receiving demand information obtained based on the service carrier, and recommending target services or target goods to the customer according to the demand information and the second preference information. Thus, efficient and accurate customer management is realized.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to an intelligent CRM customer management method and system that supports cross-platform integration. Background Technology

[0002] In the context of current digital transformation, enterprises are placing increasing emphasis on the intelligence and cross-platform integration capabilities of their customer relationship management (CRM) systems in order to achieve unified data management and intelligent decision-making.

[0003] In existing technologies, such as the AIGC (Artificial Intelligence Generated Content)-based human-machine collaborative intelligent control system disclosed in CN119940425A, although it has made innovations in human-machine interaction and decision consistency, it has not proposed solutions to the cross-platform data integration problem unique to the CRM field. For example, CN118245677A discloses a development method for a front-end big data platform for power grid marketing. Although it has made efforts in data utilization efficiency and security, its applicability is mainly limited to the power grid industry and it is difficult to directly apply it to cross-industry, multi-source data synchronization and intelligent analysis scenarios of CRM systems.

[0004] Thus, existing technologies for cross-platform integrated customer management primarily suffer from data silos. For example, customer information is scattered across different business platforms, lacking unified management. This results in a lack of effectiveness in customer management systems, specifically inaccurate product or service recommendations, which negatively impacts customer conversion rates and ultimately leads to low commercial value. Therefore, improving the management effectiveness of existing customer management systems has become a pressing technical challenge. Summary of the Invention

[0005] In view of this, the main objective of the present invention is to provide an intelligent CRM customer management method and system that supports cross-platform integration, aiming to achieve more efficient, centralized and accurate customer management.

[0006] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0007] This invention provides an intelligent CRM customer management method that supports cross-platform integration, the method comprising:

[0008] Obtain customer information for the same customer from at least two different online platforms;

[0009] Based on customer information, a trained federated learning model is used to predict the customer's first and second preferences.

[0010] Based on the first preference information, select a service provider that matches the customer's preferences, and then display the second preference information to the service provider.

[0011] Receive information based on the service carrier or obtain demand information, and recommend target services or target products to customers based on demand information and second preference information.

[0012] In the above solution, obtaining customer information for the same customer from at least two different network platforms includes:

[0013] Obtain customer information of customers in N network platforms through the API interfaces of N network platforms, where the API interface is an allowed access interface and N is a positive integer greater than or equal to 2;

[0014] Aggregate N customer information entries to obtain the first customer information, and determine whether the first customer information is complete;

[0015] If the first customer information is complete, the aggregated first customer information will be used as the customer's customer information.

[0016] The method in the above scheme further includes:

[0017] If the first customer information is incomplete, obtain the customer information of the customer in N+1 network platforms through the API interface of N+1 network platforms; aggregate the N+1 customer information to obtain the second customer information, until the second customer information is complete.

[0018] In the above scheme, if the customer information is incomplete, the missing information of the customer information is determined;

[0019] Supplement missing information using a trained customer analysis model;

[0020] Based on the missing information, determine the customer's customer information.

[0021] In the above scheme, displaying the second preference information to the service carrier includes:

[0022] The second preference information is encrypted and then displayed to the service carrier.

[0023] or,

[0024] The second preference information is displayed to the service provider with access rights.

[0025] In the above scheme, customer information includes: first customer information and second customer information, where the second customer information represents the compliance attributes of the first customer information;

[0026] The step of predicting a customer's first and second preference information using a trained federated learning model based on customer information includes:

[0027] Based on the first customer information, select the first customer information with compliance attributes as the third customer information;

[0028] Based on third-party customer information, a trained federated learning model is used to predict the customer's second preference information.

[0029] In the above scheme, the service carrier includes: robot equipment or waiters; the first preference information includes: service preferences;

[0030] The step of selecting a service carrier that matches the customer's preferences based on the first preference information and displaying the second preference information to the service carrier includes:

[0031] If the customer's service preference is unmanned service, then a robot device that matches the customer's preference is selected as the service carrier, and the second preference information is displayed on the robot device's display screen;

[0032] or,

[0033] If the customer's service preference is to be served by someone, then a waiter who matches the customer's preference will be selected as the service provider, and the second preference information will be displayed on the waiter's wearable device.

[0034] Furthermore, embodiments of the present invention also provide an intelligent CRM customer management system that supports cross-platform integration, the system comprising:

[0035] The acquisition module is used to acquire customer information for the same customer from at least two different online platforms.

[0036] The prediction module is used to predict a customer's first and second preferences based on customer information using a trained federated learning model.

[0037] The selection module is used to select a service carrier that matches the customer's preferences based on the first preference information, and then display the second preference information to the service carrier.

[0038] The recommendation module is used to receive demand information obtained based on the service carrier, and recommend target services or target products to customers based on the demand information and second preference information.

[0039] To achieve the above objectives, embodiments of the present invention also provide a computing device, the computing device comprising:

[0040] processor;

[0041] A memory for storing processor-executable instructions; wherein the processor is used to execute a smart CRM customer management method supporting cross-platform integration as described in any of the above schemes.

[0042] To achieve the above objectives, embodiments of the present invention also provide a computer storage medium, comprising: the computer storage medium storing one or more programs, the one or more programs being executable by one or more processors to cause the one or more processors to execute the intelligent CRM customer management method supporting cross-platform integration as described in any of the above schemes.

[0043] This invention provides an intelligent CRM customer management method and system supporting cross-platform integration. It acquires customer information for the same customer from at least two different network platforms; predicts the customer's first and second preferences using a trained federated learning model based on this information; selects a service carrier matching the customer's preferences based on the first preference and displays the second preference information to the service carrier; receives demand information obtained from the service carrier; and recommends target services or products to the customer based on the demand information and the second preference information. Therefore, this invention constructs a more comprehensive customer information system by acquiring customer information for the same customer from different network platforms. The federated learning model protects customer information privacy, and the dual preference prediction—rapidly matching the first preference to the service carrier and then using the second preference to facilitate the service carrier's rapid acquisition of the customer's current real demand information—enables accurate recommendations of desired products or services based on the current real demand information and the second preference information, improving the conversion rate and thus enhancing the effectiveness and commercial value of customer analysis and management using an intelligent CRM customer management system supporting cross-platform integration. Attached Figure Description

[0044] Figure 1 A flowchart illustrating an intelligent CRM customer management method supporting cross-platform integration, provided for some embodiments of the present invention;

[0045] Figure 2 A schematic diagram of the federated learning model architecture provided for some embodiments of the present invention;

[0046] Figure 3 A schematic diagram illustrating the structural composition of an intelligent CRM customer management system supporting cross-platform integration, provided for some embodiments of the present invention;

[0047] Figure 4 This is a schematic diagram of the hardware structure of a computing device provided in an embodiment of the present invention. Detailed Implementation

[0048] The present invention aims to obtain multiple preference information of a customer by using a pre-trained federated learning model based on customer information of the same customer obtained from different network platforms, and to provide more accurate recommendations of goods or services based on the multiple preference information, thereby increasing the purchase rate of goods or services, increasing the customer conversion rate, and thus improving the management effectiveness and value of the customer management system.

[0049] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0050] Understandably, CRM (Customer Relationship Management) is a system of technologies, strategies, and processes used to manage interactions between a business and its customers. Its core objectives are to optimize customer experience, improve customer satisfaction, enhance customer loyalty, and ultimately increase revenue and profitability. Core CRM functions include, but are not limited to: customer data management, sales management, marketing automation, customer service and support, and data analytics and business intelligence. In short, CRM is not merely a tool for storing customer data; it is a core system for businesses to improve customer experience, optimize business processes, and drive growth. Choosing the right CRM and effectively utilizing technologies such as data analytics, automation, and AI can help businesses gain a competitive edge.

[0051] The intelligent CRM customer management mentioned in this embodiment is an intelligent customer management system that supports cross-platform integration. This system runs on a computing device and is used to execute the following method steps:

[0052] like Figure 1 As shown, Figure 1 This is a flowchart illustrating a cross-platform integrated intelligent CRM customer management method according to some embodiments of the present invention. The embodiments of the present invention provide a cross-platform integrated intelligent CRM customer management method applied to a computing device, the method comprising:

[0053] Step 101: Obtain customer information for the same customer from at least two different online platforms;

[0054] The computing device here can be a terminal device, which can be a fixed terminal or a mobile terminal. A fixed terminal can be a desktop computer or an all-in-one computer, and a mobile terminal can be a laptop, tablet, or mobile phone, etc. In some embodiments, the computing device here can be a device for running an intelligent CRM customer management system that supports cross-platform integration.

[0055] In some embodiments, before acquiring customer information for the same customer from at least two different online platforms, the method further includes: checking the API policies and user authorization status of at least two different online platforms. Acquiring customer information for the same customer from at least two different online platforms includes: if the API policies of at least two different online platforms are open, acquiring user data for the same customer that is authorized and allowed to be accessed by the user from at least two different online platforms. Thus, the obtained user data is compliant and legal, providing a legal and compliant foundation and strong guarantee for subsequent customer management based on customer data.

[0056] It should be noted that in some implementations, the "two different online platforms" can refer to two different online platforms of the same category, such as JD.com or Taobao, both being online shopping platforms. In this case, by obtaining customer information from two online platforms of the same category but different platforms, more targeted customer information, such as information specific to shopping, can be obtained, thus facilitating a more effective customer profile for shopping recommendations. In other implementations, the "two different online platforms" can refer to two online platforms of different categories, such as a social networking platform (e.g., Weibo) and an online shopping platform (e.g., Taobao). In this case, by obtaining customer information from at least two different categories of online platforms, customer information can be analyzed from different perspectives to obtain a more comprehensive customer profile. Of course, in other implementations, obtaining customer information for the same customer from at least two different online platforms can include obtaining customer information for the same customer from multiple online platforms of different and the same category, thus facilitating a more comprehensive and targeted customer profile, which in turn facilitates more accurate customer management based on this profile.

[0057] Step 102: Based on the customer information, use the trained federated learning model to predict the customer's first preference information and second preference information.

[0058] It's important to understand that federated learning is a distributed machine learning technique whose core goal is "the model moves while the data remains stationary." Multiple participants, such as enterprises or devices, collaboratively train a global model without sharing the original data, while simultaneously protecting data privacy. Traditional machine learning models centralize all data for training, posing a high risk of privacy breaches. Federated learning, on the other hand, keeps the data locally; each participant's original data does not leave their local server. The interaction involves each participant training their model with their local data, uploading model parameters (such as gradients or weights) to a central server, and then performing global model aggregation. The central server securely aggregates the model parameters from all parties, updates them, and distributes the new model to all participants. In this way, the original data remains on-site, complying with regulations such as GDPR (General Data Protection Regulation) and CCPA (California Consumer Privacy Act), thus providing data privacy protection; the distributed training method, which allows the model to be trained in parallel on multiple data sources, can improve model training efficiency; encryption technology is used to protect parameter transmission and prevent man-in-the-middle attacks, thereby ensuring the security of aggregation; and it supports data from different feature spaces, such as data from different network platforms, thus ensuring the compatibility of heterogeneous data.

[0059] It's understandable that federated learning can be categorized into three types: horizontal federated learning, vertical federated learning, and federated transfer learning. Horizontal federated learning is suitable for scenarios where participating parties have the same data characteristics but different samples. For example, jointly training an input method prediction model using the input habits of Apple iPhone users; however, the user data here is limited to mobile phones. Vertical federated learning is suitable for participating parties with the same data samples but different features. For example, the same user's e-commerce and social data on different platforms; such as banks and e-commerce platforms jointly modeling and assessing user credit risk, but the two parties do not exchange raw data. Federated transfer learning is suitable when there is little overlap in the participating parties' data and samples, improving efficiency through transfer learning. For example, Hospital A has a small amount of cancer data, and Hospital B has a large amount of diabetes data, jointly training a disease prediction model.

[0060] The first and second preference information here are information that represents customer preferences from two different dimensions. It can be understood that the first preference information is used to represent the user's preferences during the shopping process, while the second preference information represents the user's preferences for the purchase outcome.

[0061] It is understandable that step 102, in conjunction with Figure 2 As shown, predicting a customer's first and second preferences using a trained federated learning model based on customer information can include:

[0062] The central server 20 corresponding to the federated learning model distributes the initial model to at least two network platforms. The following example uses two network platforms A and B.

[0063] A uses, for example, a customer's purchase history to train model 21, and calculates the gradient ΔW. a B uses, for example, customer "like" behavior to train model 22, and calculates the gradient ΔW. β ;

[0064] Central server 20 obtains the global gradient ΔW = (ΔW) through methods such as weighted averaging or cryptographic aggregation. a + ΔW β ) / 2;

[0065] The central server 20 updates the global model and sends the new parameters to the local models of A and B, and A and B synchronize the new parameters.

[0066] Each online platform uses the final model to predict local customer preferences. For example, it predicts customer preferences for the purchase process and purchase outcome. For instance, it predicts that customers prefer a purchase process where they can make their own choices, such as not needing a salesperson to push the product. It also predicts that customers prefer to buy currently popular models or non-mainstream models that offer good value for money.

[0067] Furthermore, to ensure the privacy of federated learning models during the prediction process, in some implementations, federated learning can prevent data leakage in the following ways:

[0068] Differential privacy involves adding random noise to the gradient, making it impossible for external parties to reverse engineer the original data.

[0069] Homomorphic encryption means that the model parameters are calculated in an encrypted state, and the server cannot decrypt them;

[0070] Secure multi-party computation means that during collaborative computation among multiple parties, no single party can know the input of the others.

[0071] In this way, federated learning resolves the conflict between data privacy and sharing through a distributed collaborative model, making it particularly suitable for joint analysis across enterprises and platforms. Therefore, through federated learning, enterprises can legally and compliantly manage platform customers while meeting regulatory requirements such as GDPR without sharing raw data. This enables cross-platform customer management under compliance, thereby reducing data silos in cross-platform customer management in existing technologies and ensuring the effectiveness of customer management systems.

[0072] Step 103: Based on the first preference information, select a service carrier that matches the customer's preferences, and display the second preference information to the service carrier.

[0073] As mentioned above, first preference information represents a customer's preferences in the purchasing process. For example, some customers prefer to have the salesperson provide a detailed introduction and demonstration of the product or service during the purchase process; while others prefer to make a purchase based on their own understanding and do not want the salesperson to introduce or promote the product or service. They may believe that the salesperson's introduction and promotion are somewhat biased and not conducive to their true needs.

[0074] The service provider here can be understood as the service provider who offers this purchase activity to the client.

[0075] The second preference information here can include, but is not limited to: payment preferences, channel preferences, functional attribute preferences, experiential and emotional preferences, social and recommendation preferences, risk preferences, contextual preferences, and product geographic preferences. Payment preferences may include, for example, preferences for payment time and payment method; channel preferences may include preferences for purchase channels, interaction channels, and promotional channels; functional attribute preferences may include preferences for product characteristics, such as preference for sugar-free or sugar-free products, technical parameter preferences, and service add-on preferences; experiential and emotional preferences may include preferences for purchasing after experiencing the product; and social and recommendation preferences may include preferences for purchasing through friend referral links. Thus, by displaying this second preference information to the service provider, the service provider can select appropriate marketing strategies based on the customer's second preference. For example, if one of the customer's preferences is to purchase after experiencing the product, one or two experiential items can be recommended during the interaction; if one of the customer's preferences is a preference for product characteristics, detailed product characteristic displays or feature descriptions can be provided during the interaction.

[0076] In this embodiment, by first selecting a service provider that matches the customer's first preference to serve the customer, it is beneficial to reduce customer churn and increase the customer purchase rate during the purchase process. Then, by displaying the customer's second preference to the service provider, the service provider can interact based on the second preference, thereby more accurately grasping the customer's preferences and helping the customer select a product or service that better suits their needs, further improving the customer purchase rate and enhancing the commercial value of the customer management system.

[0077] In some embodiments, the service carrier includes a robot device or a waiter, and the first preference information includes service preferences.

[0078] Step 103, namely, selecting a service provider that matches the customer's preferences based on the first preference information, and displaying the second preference information to the service provider, includes:

[0079] If the customer's service preference is unmanned service, then select a robot device that matches the customer's preference as the service carrier, and display the second preference information on the robot device's display screen;

[0080] or,

[0081] If the customer's service preference is to be served by someone, then a waiter who matches the customer's preference will be selected as the service provider, and the second preference information will be displayed on the waiter's wearable device.

[0082] The wearable devices here could be smartwatches or smart glasses worn by waiters.

[0083] In this embodiment, by predicting the customer's purchase process preferences, such as liking detailed explanations during the purchase process or disliking human intervention, the service carrier is first matched based on the purchase process preferences, and then the second preference information is informed to the service carrier. This can effectively retain customers at least during the purchase process, reduce customer churn rate, and increase purchase rate.

[0084] Step 104: Receive demand information obtained based on the service carrier, and recommend target services or target products to customers based on the demand information and second preference information.

[0085] The demand information here can be the customer's current purchase needs obtained based on the interaction between the service provider and the customer, such as the information about the items the customer needs to buy after entering the store. This item information includes, but is not limited to, the signal, function, and appearance description of the item. Then, combined with the secondary preference information, the system can accurately recommend target services or target products to the customer. For example, based on the customer's price range preference, the system can recommend products or services that the customer may want to learn more about, thereby increasing the customer's purchase rate.

[0086] In some implementations, before receiving demand information obtained based on the service carrier, the method further includes:

[0087] The service carrier generates query information that matches the second preference information based on the second preference information;

[0088] Based on the inquiry information, we obtained the customer's needs information.

[0089] Understandably, taking offline store shopping as an example, the service provider could generate an inquiry that matches the customer's secondary preference, such as a preference for organic cotton materials, based on the information the salesperson has learned. For example, the inquiry could directly ask if the customer needs a certain type of product from the organic cotton section, thereby reducing communication costs and improving the efficiency of transactions for both merchants and customers.

[0090] In some embodiments, the method further includes: receiving demand information obtained based on the service carrier, and correcting the first preference information according to the demand information.

[0091] Understandably, for some customers, their preferences during the purchasing process are not static. In some scenarios, when the initial preference information represents the customer's detailed explanatory preferences during the service process, if the customer is detected to exhibit a tendency to make independent decisions, such as frequently looking at their phone to compare, it is necessary to notify the store staff in real time to switch to a low-intervention mode. That is, based on the received real-time demand information, the initial preference information is adjusted, which is conducive to the smooth progress of the purchasing process, reduces customer churn, and increases the purchase rate.

[0092] In summary, the above embodiments construct a more comprehensive customer information system by acquiring customer information from the same customer across different network platforms. The federated learning model protects customer privacy from leakage. Furthermore, by predicting the customer's dual preferences, the system matches the service provider with the first preference and then uses the second preference to help the service provider obtain the customer's current and accurate needs. Based on these needs and the second preference, the system can accurately recommend desired products or services to the customer, improving recommendation conversion rates and ultimately increasing purchase rates. Therefore, compared to traditional cross-platform customer management, the intelligent CRM customer management method supporting cross-platform integration provided in this embodiment makes customer management more effective and has higher business value.

[0093] In some embodiments, step 101, namely obtaining customer information for the same customer from at least two different network platforms, may include:

[0094] Obtain customer information of customers in N network platforms through the API interfaces of N network platforms, where the API interface is an allowed access interface, and N is a positive integer greater than or equal to 2;

[0095] Aggregate N customer information entries to obtain the first customer information, and determine whether the first customer information is complete;

[0096] If the first customer information is complete, the aggregated first customer information will be used as the customer's customer information.

[0097] In other embodiments, the method further includes:

[0098] If the first customer information is incomplete, obtain customer information from N+1 network platforms through the API interface of the N+1 network platform; aggregate the N+1 customer information to obtain the second customer information, until the second customer information is complete.

[0099] For example, obtaining customer information of a customer from N network platforms through the API interfaces of N network platforms can include: using an asynchronous thread pool to concurrently call the API interfaces of each network platform, thereby shortening the response time; the asynchronous thread pool here is, for example, Java's CompletableFuture or Python's asyncio.

[0100] For example, each API interface is configured with an OAuth 2.0 token or API key and authentication information is attached to the request header to ensure the security of customer information.

[0101] For example, aggregating N pieces of customer information can include: using a priority strategy to aggregate customer information from the same category of online platforms with higher credibility, thereby ensuring the credibility of the customer information obtained.

[0102] For example, determining whether the first customer information is complete includes: based on preset customer information attributes, determining whether all attributes of the first customer information are marked, if so, the customer information is considered complete; if any attribute in the first customer information is not marked, the customer information is considered incomplete.

[0103] The attributes here represent a certain type of information in customer information. Different attributes correspond to different types of information. For example, account information and gender information in customer information are different types of information.

[0104] In this embodiment, the aggregation of customer information from multiple network platforms ensures the integrity of customer information, providing a more accurate and favorable guarantee for subsequent prediction of customer preferences based on customer information, and improving the reliability of the customer management system.

[0105] In some embodiments, the method further includes:

[0106] If customer information is incomplete, identify the missing information.

[0107] The missing information can be supplemented by a well-trained customer analysis model;

[0108] Based on the missing information, determine the customer's customer information.

[0109] It should be added that the customer analysis model can be trained using complete historical customer data to train the following models: classification model, regression model, and generative model. The classification model is used to predict discrete fields, such as occupation and education level; the regression model is used to predict continuous fields, such as income and credit score; and the generative model is used to generate reasonable but fictitious temporary values, such as generating a temporary email address based on a name.

[0110] For example, if customer information fields cannot be completed across platforms, they can be completed by calling a model and using retrospective prediction.

[0111] In some embodiments, after supplementing the missing information using a trained customer analysis model, the missing information is validated. For example, it is checked whether the completed value is logically correct, such as whether the age and date of birth are consistent.

[0112] To ensure the accuracy of the customer analysis model, manually corrected data can be added to the training set to continuously optimize the model.

[0113] In this embodiment, by using a customer analysis model to supplement incomplete customer information, the integrity of customer information is ensured, providing a more accurate and favorable guarantee for subsequent prediction of customer preferences based on customer information, and improving the reliability of the customer management system.

[0114] In some embodiments, displaying the second preference information to the service carrier includes:

[0115] The second preference information is encrypted and then displayed to the service carrier.

[0116] or,

[0117] The second preference information is displayed to the service provider with access rights.

[0118] In this embodiment, by keeping the second preference information confidential, the reliability and trustworthiness of the customer management system are improved, providing a strong guarantee for protecting customer information and the privacy of customer preferences.

[0119] In some embodiments, the customer information includes: first customer information and second customer information, wherein the second customer information represents the compliance attributes of the first customer information;

[0120] The step of predicting a customer's first and second preference information using a trained federated learning model based on customer information includes:

[0121] Based on the first customer information, determine the second customer information within the first customer information;

[0122] Based on the second customer information, select the first customer information with compliance attributes as the third customer information;

[0123] Based on third-party customer information, a trained federated learning model is used to predict the customer's second preference information.

[0124] It is understandable that the first customer information may include, but is not limited to: customer basic information, customer environmental information, customer social relationship information, customer historical behavior information, customer credit information, etc.

[0125] The second customer information is the compliance attribute of the first customer information. In other words, for basic customer information, the second customer information is the attribute information indicating whether the basic customer information is compliant. In this embodiment, by predicting only the first customer information with compliant attributes, the use of non-compliant customer information is reduced, thus ensuring the security and reliability of the customer management system.

[0126] To achieve the above objectives, embodiments of the present invention also provide an intelligent CRM customer management system that supports cross-platform integration. Please refer to [link to relevant documentation]. Figure 3 The system includes:

[0127] Module 31 is used to acquire customer information for the same customer from at least two different network platforms;

[0128] Prediction module 32 is used to predict the customer's first preference information and second preference information based on the customer information and using a trained federated learning model.

[0129] The selection module 33 is used to select a service carrier that matches the customer's preferences based on the first preference information, and to display the second preference information to the service carrier.

[0130] The recommendation module 34 is used to receive demand information obtained based on the service carrier, and recommend target services or target products to customers based on the demand information and second preference information.

[0131] In some embodiments, the acquisition module 31 is further configured to:

[0132] Customer information of the customer in the N network platforms is obtained through the API interfaces of the N network platforms, wherein the API interface is an allowed access interface, and N is a positive integer greater than or equal to 2;

[0133] Aggregate N customer information entries to obtain the first customer information, and determine whether the first customer information is complete;

[0134] If the first customer information is complete, then the aggregated first customer information will be used as the customer information of the customer.

[0135] In some embodiments, the system further includes: a first processing module, configured to:

[0136] If the first customer information is incomplete, obtain the customer information of the customer from the N+1 network platforms through the API interfaces of the N+1 network platforms; aggregate the N+1 customer information to obtain the second customer information, until the second customer information is complete.

[0137] In some embodiments, the system further includes: a second processing module, configured to:

[0138] If the customer information is incomplete, determine the missing information in the customer information;

[0139] The missing information is supplemented by a trained customer analysis model;

[0140] Based on the missing information, the customer information of the customer is determined.

[0141] In some embodiments, the recommendation module 34 is further configured to:

[0142] The second preference information is encrypted, and the encrypted second preference information is displayed to the service carrier.

[0143] or,

[0144] The second preference information is displayed to the service carrier that has access rights.

[0145] In some embodiments, the customer information includes: first customer information and second customer information, wherein the second customer information characterizes the compliance attributes of the first customer information;

[0146] The prediction module 32 is further configured to:

[0147] Based on the first customer information, determine the second customer information in the first customer information;

[0148] Based on the second customer information, the first customer information with compliance attributes is selected as the third customer information;

[0149] Based on the third customer information, the trained federated learning model is used to predict the customer's second preference information.

[0150] In some embodiments, the service carrier includes: a robotic device or a waiter; the first preference information includes: service preferences;

[0151] The selection module 33 is also used for:

[0152] If the customer's service preference is unmanned service, then a robot device that matches the customer's preference is selected as the service carrier, and the second preference information is displayed on the display screen of the robot device;

[0153] or,

[0154] If the customer's service preference is to have someone serve them, then a waiter who matches the customer's preference is selected as the service provider, and the second preference information is displayed on the waiter's wearable device.

[0155] It should be noted that the description of the intelligent CRM customer management system supporting cross-platform integration described above is similar to the description of the intelligent CRM customer management method supporting cross-platform integration described above. The beneficial effects of the same method will not be repeated. For technical details not disclosed in the embodiments of the intelligent CRM customer management system supporting cross-platform integration of this invention, please refer to the description of the embodiments of the intelligent CRM customer management method supporting cross-platform integration of this invention.

[0156] To achieve the above objectives, embodiments of the present invention also provide a computing device, such as... Figure 4 As shown, the computing device includes a processor 401 and a memory 402 connected to the processor 401 via a communication bus 403; wherein, the memory 402 is used to support a cross-platform integrated intelligent CRM customer management program; the processor 401 is used to execute the cross-platform integrated intelligent CRM customer management program to implement the cross-platform integrated intelligent CRM customer management method described in any of the above schemes.

[0157] Optionally, the processor 401 may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. Here, the program executed by the processor 401 may be stored in a memory 402 connected to the processor 401 via a communication bus 403. The memory 402 may be volatile memory or non-volatile memory, or may include both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache.By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Sync Link Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM). The memory 402 described in this embodiment is intended to include, but is not limited to, these and any other suitable types of memory 402. The memory 402 in this embodiment is used to store various types of data to support the operation of the processor 401. Examples of this data include: any computer programs operated by the processor 401, such as operating systems and applications; contact data; phone book data; messages; pictures; videos, etc. The operating system contains various system programs, such as the framework layer, core library layer, and driver layer, used to implement various basic business functions and handle hardware-based tasks.

[0158] In some embodiments of the present invention, the memory 402 may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DRRAM). The memory 402 of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0159] The processor 401 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 401 or by software instructions. The processor 401 can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory 402, and the processor 401 reads the information in memory 402 and, in conjunction with its hardware, completes the steps of the above method. In some embodiments, the embodiments described herein can be implemented using hardware, software, firmware, middleware, microcode, or a combination thereof. For hardware implementation, the processing unit can be implemented in one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontrollers, microprocessors, other electronic units for performing the functions described herein, or combinations thereof.

[0160] For software implementation, the techniques described herein can be achieved through modules (e.g., procedures, functions, etc.) that perform the functions described herein. The software code can be stored in memory and executed by a processor. The memory can be implemented within the processor or externally.

[0161] Another embodiment of the present invention provides a computer storage medium storing an executable program, which, when executed by a processor 401, can implement the steps of an intelligent CRM customer management method supporting cross-platform integration applied to the computing device. For example, as... Figures 1-2 One or more of the methods shown.

[0162] In some embodiments, the computer storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0163] It should be noted that the technical solutions described in the embodiments of the present invention can be combined arbitrarily without conflict.

[0164] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. A smart CRM customer management method supporting cross-platform integration, characterized in that, The method includes: Obtain customer information for the same customer from at least two different online platforms; Based on the customer information, a trained federated learning model is used to predict the customer's first preference information and second preference information. The first preference information represents the customer's preference for service interaction modes and is used to decide the type of service carrier. The second preference information includes payment preferences, channel preferences, functional attribute preferences, experience and emotional preferences, social and recommendation preferences, risk preferences, contextual preferences, and product regional preferences, and is used to guide the interaction content between the service carrier and the customer. Based on the first preference information, a service carrier matching the customer's preferences is selected from robot devices and waiters, and the second preference information is displayed to the service carrier to guide the service carrier to interact with the customer based on the second preference information; Receive demand information obtained by the service carrier through interaction with the customer under the guidance of the second preference information, and recommend target services or target products to the customer based on the demand information and the second preference information; The acquisition of customer information for the same customer from at least two different network platforms includes: Customer information of the customer in the N network platforms is obtained through the API interfaces of the N network platforms, wherein the API interface is an allowed access interface, and N is a positive integer greater than or equal to 2; Aggregating N customer information to obtain first customer information, and determining whether the first customer information is complete, the aggregation of N customer information includes: using a priority strategy to aggregate customer information from the same category of online platforms with higher credibility; If the first customer information is complete, then the aggregated first customer information will be used as the customer information of the customer.

2. The method according to claim 1, characterized in that, The method further includes: If the first customer information is incomplete, obtain the customer information of the customer from the N+1 network platforms through the API interfaces of the N+1 network platforms; aggregate the N+1 customer information to obtain the second customer information, until the second customer information is complete.

3. The method according to claim 1, characterized in that, The method further includes: If the customer information is incomplete, determine the missing information in the customer information; The missing information is supplemented by a trained customer analysis model; Based on the missing information, the customer information of the customer is determined.

4. The method according to claim 1, characterized in that, The step of displaying the second preference information to the service carrier includes: The second preference information is encrypted, and the encrypted second preference information is displayed to the service carrier. or, The second preference information is displayed to the service carrier that has access rights.

5. The method according to claim 1, characterized in that, The customer information includes: first customer information and second customer information, wherein the second customer information represents the compliance attributes of the first customer information; The step of predicting the customer's first preference information and second preference information using a trained federated learning model based on the customer information includes: Based on the first customer information, determine the second customer information in the first customer information; Based on the second customer information, the first customer information with compliance attributes is selected as the third customer information; Based on the third customer information, the trained federated learning model is used to predict the customer's second preference information.

6. The method according to claim 1 or 4, characterized in that, Service carriers include: robotic devices or waiters; primary preference information includes: service preferences; The step of selecting a service carrier that matches the customer's preferences based on the first preference information and displaying the second preference information to the service carrier includes: If the customer's service preference is unmanned service, then a robot device that matches the customer's preference is selected as the service carrier, and the second preference information is displayed on the display screen of the robot device; or, If the customer's service preference is to have someone serve them, then a waiter who matches the customer's preference is selected as the service provider, and the second preference information is displayed on the waiter's wearable device.

7. A smart CRM customer management system that supports cross-platform integration, characterized in that, The system includes: The acquisition module is used to acquire customer information for the same customer from at least two different network platforms, including: acquiring customer information of the customer from the N network platforms through API interfaces of the N network platforms, wherein the API interfaces are allowed access interfaces, and the N is a positive integer greater than or equal to 2; aggregating the N customer information to obtain first customer information; determining whether the first customer information is complete; the aggregating of the N customer information includes: aggregating customer information from the same category of network platforms with higher credibility using a priority strategy; if the first customer information is complete, then using the aggregated first customer information as the customer information of the customer. The prediction module is used to predict the customer's first preference information and second preference information based on the customer information using a trained federated learning model. The first preference information represents the customer's preference for service interaction modes and is used to decide the type of service carrier. The second preference information includes payment preferences, functional attribute preferences, experience and emotional preferences, social and recommendation preferences, risk preferences, contextual preferences, and product regional preferences, and is used to guide the interaction content between the service carrier and the customer. The selection module is used to select a service carrier that matches the customer's preferences from robot devices and waiters based on the first preference information, and to display the second preference information to the service carrier to guide the service carrier to interact with the customer based on the second preference information; The recommendation module is used to receive demand information obtained by the service carrier through interaction with the customer under the guidance of the second preference information, and to recommend target services or target products to the customer based on the demand information and the second preference information.

8. A computing device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to perform the method as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, include: A computer storage medium stores one or more programs that can be executed by one or more processors to cause the one or more processors to perform the method as described in any one of claims 1 to 6.

Images