Banking outlet personalized service method, system and equipment based on multi-dimensional identification

Abstract

The invention relates to the field of bank services, and discloses a bank outlet personalized service method, system and device based on multi-dimensional recognition, and the key point of the technical scheme is that the method comprises the following steps: S1, when a target customer enters a current outlet, collecting the multi-dimensional data of the target customer in real time; s2, performing real-time data fusion processing on the multi-dimensional data, performing data association on the multi-dimensional data through a unique identifier, and generating a target customer comprehensive portrait; s3, making a service decision according to the comprehensive portrait of the target customer, and generating a service scheme; s4, pushing the service scheme to service equipment matched with the real-time position of the target client; and S5, after the service equipment confirms that the target customer is docked, service operation is executed on the target customer according to the service scheme, and comprehensive, real-time and personalized bank outlet marketing service is truly realized through integrating image recognition, biological recognition and position recognition technologies.

Description

Technical Field

[0001] This invention relates to the field of banking services, and more specifically, to a method, system, and device for personalized services at bank branches based on multi-dimensional recognition. Background Technology

[0002] As the digital transformation of the banking industry deepens, personalized customer experience has become a crucial direction for improving bank branch services. Modern bank branches are not merely venues for financial transactions, but also vital nodes for customer interaction and the provision of value-added services. Against this backdrop, how to leverage advanced technologies to acquire multi-dimensional customer information and provide personalized services based on this information has become a key focus for the banking industry.

[0003] Image recognition, biometrics, and location recognition technologies, as typical applications of artificial intelligence, have been widely adopted in various fields. For example, image recognition technology can be used to identify customers' physical characteristics, such as age, gender, and clothing; biometrics can obtain customer identity information through fingerprint, iris, and facial recognition; and location recognition technology can obtain a customer's specific location within a bank branch in real time. These technologies provide banks with powerful tools for collecting customer information, making personalized marketing based on this information possible.

[0004] Currently, in the field of personalized marketing at bank branches, some technologies based on image recognition, biometrics, and location recognition have been applied in practice.

[0005] For example: Image recognition technology: Existing image recognition systems can capture customers' facial features through cameras and, combined with customer behavior analysis, infer their potential needs. This type of technology is often used in bank branch greeting systems to determine a customer's emotional state or approximate age in order to recommend appropriate products or services. Biometric technology: Primarily used for customer identity verification and access control. For example, facial recognition technology is already widespread in many bank branch self-service machines for customer identity verification and security authentication. Some advanced applications can also push relevant financial products to customers based on their historical transaction records. Location recognition technology: Currently, this mainly uses customers' mobile devices for positioning, such as using Wi-Fi, Bluetooth, or GPS technologies to determine the customer's location within the bank branch. Based on the customer's current location, the system can push service or product information related to that location; for example, when a customer approaches the financial advisory desk, the system automatically pushes financial product recommendations.

[0006] Although the above technologies have been applied in bank branches, some significant drawbacks still exist:

[0007] Information silos: Currently, most image recognition, biometric recognition, and location recognition technologies operate independently, lacking effective integration and collaboration. This results in customer information being obtained only from a single dimension, failing to form a comprehensive customer profile and limiting the accuracy of personalized marketing.

[0008] Lack of real-time capability: Existing systems suffer from delays in data processing and analysis, especially at high-traffic locations, making it difficult to achieve real-time personalized recommendations and potentially failing to meet customers' latent needs in a timely manner.

[0009] Limited application scenarios: Existing technologies are often concentrated in a specific scenario or device, lacking the ability to be applied across scenarios. The customer experience within the branch is fragmented due to limitations of equipment or area, making it impossible to achieve continuous, seamless, personalized service. Summary of the Invention

[0010] The purpose of this invention is to provide a method, system, and equipment for personalized services at bank branches based on multi-dimensional recognition. By integrating image recognition, biometric recognition, and location recognition technologies, it can truly achieve comprehensive, real-time, and personalized marketing services at bank branches.

[0011] The above-mentioned technical objective of the present invention is achieved through the following technical solution: a method for personalized services at bank branches based on multi-dimensional recognition, comprising the following steps:

[0012] S1. Collect multi-dimensional data of target customers in real time when they enter the current branch.

[0013] S2. Perform real-time data fusion processing on multi-dimensional data, associate multi-dimensional data with unique identifiers, and generate a comprehensive profile of the target customer.

[0014] S3. Make service decisions based on the comprehensive profile of the target customer and generate service plans;

[0015] S4. Push the service plan to the service device that matches the real-time location of the target customer;

[0016] S5. Once the service equipment confirms that it has been connected to the target customer, it will perform service operations on the target customer according to the service plan.

[0017] As a preferred embodiment of the present invention, the multidimensional data includes:

[0018] Real-time collection of target customer image data, biometric information, and real-time location information at the current branch.

[0019] As a preferred embodiment of the present invention, the data fusion processing includes:

[0020] The collected multidimensional data is preprocessed to obtain standardized data;

[0021] Generate temporary unique identifiers for standardized multidimensional data, and perform data association on the standardized multidimensional data;

[0022] Data identification is performed on standardized multidimensional data to obtain target customer profile data;

[0023] Information is integrated from the target customer profile data to obtain a comprehensive target customer profile.

[0024] As a preferred embodiment of the present invention, the preprocessing includes:

[0025] The collected multidimensional data is cleaned and filtered to remove noisy data.

[0026] Standardize multidimensional data by converting it into a unified standard format.

[0027] By adding timestamps to multidimensional data in a unified standard format, standardized multidimensional data is obtained.

[0028] As a preferred embodiment of the present invention, the data identification includes:

[0029] Transform the image data of the target customer into key feature vectors;

[0030] The biometric information of target customers is identified, converted into identity feature data, and linked to account information;

[0031] The real-time location information of target customers is identified and converted into real-time coordinate information within the network.

[0032] In a preferred embodiment of the present invention, in S3, the decision-making process is as follows: based on the comprehensive profile of the target customer and historical service information, the needs of the target customer are predicted through a behavior prediction model; based on the predicted needs of the target customer, a service plan is generated through a recommendation algorithm.

[0033] In a preferred embodiment of the present invention, in step S5, after performing service operations on the target customer according to the service plan, the target customer's behavioral response to the service plan is recorded and fed back to the behavior prediction model.

[0034] A personalized service system for bank branches based on multi-dimensional recognition includes:

[0035] The multi-dimensional information collection and recognition module is used to collect and recognize multi-dimensional data of target customers in real time when they enter the current branch.

[0036] The central data processing module is used to perform real-time data fusion processing on multi-dimensional data, associate multi-dimensional data with unique identifiers to generate a comprehensive profile of the target customer; make service decisions based on the comprehensive profile of the target customer and generate service plans; and push the service plans to service devices that match the real-time location of the target customer.

[0037] Service equipment is used to perform service operations on the target customer according to the service plan after confirming connection.

[0038] A computer device includes a processor and a memory, the memory storing a computer program executable by the processor, wherein the processor executes the computer program to implement the method described above.

[0039] In summary, this invention offers the following advantages: By integrating image recognition, biometrics, and location recognition technologies, it overcomes the problems of information isolation, lack of real-time performance, and limited application scenarios in existing technologies, truly achieving comprehensive, real-time, and personalized marketing services for bank branches; it possesses information integration capabilities, breaking down information isolation by simultaneously using multiple recognition technologies, and achieving comprehensive integration of customer appearance, identity, and location information. This multi-dimensional information acquisition makes marketing content more accurate and significantly enhances the customer experience.

[0040] With real-time response capabilities, the overall service approach, through streamlined steps and efficient algorithms, ensures that personalized marketing information can be pushed to customers in real time after they enter the bank branch. This improved real-time performance not only allows customer needs to be met quickly but also increases the bank's marketing conversion rate.

[0041] Multi-scenario coverage capability: This invention overcomes the limitations of existing technologies applied in a single scenario, and can adapt to the needs of multiple different areas within a bank branch. Even as customers move, the system can still seamlessly track their location and dynamically push relevant marketing content, enhancing the continuity and flexibility of marketing. Attached Figure Description

[0042] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0043] The present invention will be further described in detail below with reference to the accompanying drawings.

[0044] like Figure 1 As shown, this invention provides a method for personalized services at bank branches based on multi-dimensional recognition, comprising the following steps:

[0045] S1. When a target customer enters the current branch, collect multidimensional data of the target customer in real time. The multidimensional data includes: real-time image data, biometric information, and real-time location information of the target customer at the current branch.

[0046] S2. Perform real-time data fusion processing on multi-dimensional data, associate multi-dimensional data with unique identifiers, and generate a comprehensive profile of the target customer.

[0047] Data fusion processing includes:

[0048] The collected multidimensional data undergoes preprocessing to obtain standardized data. Preprocessing includes: cleaning and filtering the collected multidimensional data to remove noisy data, such as low-quality images and misidentified signals; standardizing the multidimensional data by converting it into a unified standard format, for example, converting image data into key feature vectors, biological data into unique identifiers, and location data into coordinate information within the network; and adding timestamps to the unified standard format multidimensional data to obtain standardized multidimensional data, ensuring that data fusion can match customer behavior at the same point in time.

[0049] To generate temporary unique identifiers for standardized multidimensional data, data association is performed on the standardized multidimensional data. For example, the feature vector generated by the image recognition module through the camera will be temporarily bound to this identifier, and the customer account information identified by the biometric module will also be associated with this identifier.

[0050] Standardized multidimensional data is used for data identification to obtain target customer profile data. Data identification includes: converting target customer image data into key feature vectors, such as age, gender, and clothing; converting target customer biometric information into identity feature data and linking it to account information, for example, using facial feature vectors to match registered identity feature databases in the biometric module to verify customer identity; if a customer has not registered identity information, the feature vector generated by the image recognition module can be used as a temporary identity label; and converting target customer real-time location information into real-time coordinate information within the branch. For example, based on the customer's unique identifier, their appearance and identity information are bound to their real-time location data. Another example is if a customer with the unique identifier "12345" is in the branch waiting area, the waiting area information is bound to their profile.

[0051] The process of fusing information from target customer profile data to obtain a comprehensive target customer profile includes: The central data processing module combines data from different sources for each customer, including physical characteristics, identity information, and location data, into a comprehensive customer profile.

[0052] Appearance characteristics: such as "male, 30-35 years old, business attire".

[0053] Identity information: such as "VIP customer, account balance of 500,000 yuan".

[0054] Real-time location: such as "next to self-service terminal No. 2".

[0055] At this time, various types of information and multi-dimensional data are collected and analyzed simultaneously to form a comprehensive customer profile, ensuring the accuracy of personalized marketing information.

[0056] S3. Make service decisions and generate service plans based on the comprehensive profile of the target customer. The decision-making process is as follows: Based on the comprehensive profile of the target customer and historical service information, predict the needs of the target customer through a behavior prediction model. For example, if customer A frequently makes large transfers, combined with location data - being near a self-service terminal, it is inferred that he may need to use the transfer function.

[0057] Based on the predicted needs of target customers, service plans are generated using recommendation algorithms. Specifically, recommendation system algorithms, such as collaborative filtering and content recommendation algorithms, are used to generate a list of products or services that the customer may be interested in. For example, if customer B appears young and is wearing sportswear, sports-related credit card offers might be recommended.

[0058] Specifically, behavioral prediction models are rule-based predictive models that map customer behavior using predefined rules or decision tables. For example, based on a customer's account information and historical transaction data, the type of service the customer might need can be inferred. In most financial service scenarios, customer behavior exhibits certain patterns. For instance, customers who frequently make large transfers are likely to repeat similar actions at specific locations (such as self-service terminals).

[0059] The design principle of the rule table is as follows: A customer behavior rule table is designed to record the relationship between different customer types, account balances, historical transactions, and their potential behaviors. For example, if a customer's account balance is greater than 500,000 yuan, and the customer previously made a large deposit at the counter, then the predicted behavior of this customer at the same location is "may need to conduct wealth management business." If the customer made an interbank transfer in the previous transaction, it can be inferred that the customer may continue to make interbank transfers at the self-service terminal.

[0060] In practical applications of behavioral prediction models: when a customer arrives at a branch and completes identity verification, the system quickly determines the customer's likely next action by querying a rule table, providing data support for subsequent marketing recommendations.

[0061] Specifically, there are several types of recommendation algorithms, and the choice depends on the characteristics of the data and the needs of the system:

[0062] Content-based recommendations: Recommendations are made based on the customer's personal characteristics (such as gender, age, account type, etc.) and product features. For example, for a young female customer, the system might recommend financial products or services related to fashion and health.

[0063] Collaborative filtering recommendation: Recommends products based on a customer's historical behavior and the behavior of other similar users. For example, suppose user A and user B behave similarly in multiple scenarios. If user A is interested in a certain product, user B may also become interested in that product. The recommendation system will then recommend the same product to user B based on this similarity.

[0064] Hybrid Recommendation System: This system combines content-based recommendation and collaborative filtering methods to provide more accurate personalized recommendations. For example, in this invention, the customer's real-time location, identity information, and historical behavior are comprehensively considered to generate hybrid recommendations.

[0065] S4. Push service plans to service devices that match the real-time location of the target customer; generate suitable marketing plans based on the customer's comprehensive information through behavior prediction models and recommendation algorithms, and accurately push them to devices that match the customer's location.

[0066] Dynamic push rules are used when pushing content. The rules are as follows: the system dynamically adjusts the push content and device based on the customer's current comprehensive profile and location information.

[0067] Example of push notification rules: If the customer is a VIP user and is located in the counter area, recommend high-yield wealth management products. If the customer is a young user and is located in the waiting area, push promotional activities in the form of short videos.

[0068] Push notification channel selection: Based on the customer's real-time location and device status, select the optimal push device. For example, when a customer is near a self-service terminal, display personalized recommendations directly on the device screen. When a customer is in a waiting area, push notifications are sent to a display screen or via the customer's mobile device.

[0069] S5. Once the service equipment confirms that it has been connected to the target customer, it will perform service operations on the target customer according to the service plan.

[0070] As one embodiment of the method of the present invention, after performing service operations on target customers, the central data processing module records the target customer's behavioral response to the service plan, such as whether they clicked or completed a transaction, and feeds this feedback to the behavior prediction model for iterative optimization of the algorithm, thereby improving the accuracy of personalization. Furthermore, the entire central data processing module can continuously collect customer behavior data and regularly update customer profiles and behavior prediction models to adapt to changes in customer preferences.

[0071] It is important to note that throughout the personalized service process, this invention continuously collects, transmits, and processes multi-dimensional data in real time, enabling real-time responses to customer information. It can quickly analyze a target customer's appearance, identity, and location information upon entering the branch, and push customized marketing content based on the customer's needs or behaviors in the shortest possible time, solving the latency problem in existing technologies. The method of this invention can adapt to different areas within the branch, such as self-service areas, waiting areas, and counter areas. Regardless of where the customer is in the branch, corresponding personalized service information can be pushed in real time based on their location information and behavioral habits. This method provides a continuous marketing experience across devices and areas, eliminating the limitations of existing technologies in terms of application scenarios.

[0072] Corresponding to the above method, the present invention also provides a personalized service system for bank branches based on multi-dimensional recognition, including:

[0073] The multi-dimensional information collection and recognition module is used to collect and recognize multi-dimensional data of target customers in real time when they enter the current branch.

[0074] Most of these are information collection and recognition modules, including image recognition modules, biometric recognition modules, and location recognition modules.

[0075] Image recognition module: Captures real-time images of target customers using cameras installed in bank branches, which are then used to extract and identify the target customer's gender, age, clothing, and other physical characteristics.

[0076] Biometric module: Collects customer identity features, such as face, fingerprint, iris, etc., through biometric sensors and links them to their account information.

[0077] Location identification module: Utilizes positioning sensors, such as Wi-Fi, Bluetooth beacons, infrared, or GPS, to determine the customer's specific location at the branch.

[0078] Each information collection and identification module collects data in real time through independent sensors or devices and sends it to the central data processing module through a data interface.

[0079] The central data processing module performs real-time data fusion processing on multi-dimensional data. It associates multi-dimensional data using unique identifiers to generate a comprehensive profile of the target customer. Based on this profile, it makes service decisions and generates service plans. These plans are then pushed to service devices matched to the target customer's real-time location. As the central hub for all information aggregation and processing, the module integrates and analyzes multi-dimensional data acquired through different technologies. Furthermore, it links various data sources together using unique identifiers, ensuring that different types of data accurately correspond to the same customer and achieving cross-module information fusion.

[0080] Service equipment is used to perform service operations on the target customer according to the service plan after confirming connection.

[0081] In this invention, a central data processing module enables push devices in various branches, such as self-service terminals, counters, and mobile devices, to flexibly connect to different data sources. This flexibility ensures that regardless of the customer's location, the system can select the optimal push device based on their location and behavior, enhancing the system's adaptability to different scenarios. This allows personalized services to cover the entire branch, ensuring that every customer interaction is precisely matched to their needs.

[0082] Corresponding to the methods and systems described above, the present invention also provides a computer device, including: a processor and a memory, wherein the memory stores a computer program executable by the processor, and the processor implements the methods described above when executing the computer program.

[0083] The advantages of this invention are: by integrating image recognition, biometric recognition and location recognition technologies, it can overcome the problems of information isolation, lack of real-time performance and limited application scenarios in the existing technology, and truly realize comprehensive, real-time and personalized bank branch marketing services;

[0084] It possesses information integration capabilities, breaking down information silos by simultaneously using multiple identification technologies to achieve comprehensive integration of customer appearance, identity, and location information. This multi-dimensional information acquisition makes marketing content more precise and significantly enhances the customer experience.

[0085] With real-time response capabilities, the overall service approach, through streamlined steps and efficient algorithms, ensures that personalized marketing information can be pushed to customers in real time after they enter the bank branch. This improved real-time performance not only allows customer needs to be met quickly but also increases the bank's marketing conversion rate.

[0086] Multi-scenario coverage capability: This invention overcomes the limitations of existing technologies applied in a single scenario, and can adapt to the needs of multiple different areas within a bank branch. Even as customers move, the system can still seamlessly track their location and dynamically push relevant marketing content, enhancing the continuity and flexibility of marketing.

[0087] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for personalized services at bank branches based on multidimensional recognition, characterized by: Includes the following steps: S1. Collect multi-dimensional data of target customers in real time when they enter the current branch. S2. Perform real-time data fusion processing on multi-dimensional data, associate multi-dimensional data with unique identifiers, and generate a comprehensive profile of the target customer. S3. Make service decisions based on the comprehensive profile of the target customer and generate service plans; S4. Push the service plan to the service device that matches the real-time location of the target customer; S5. Once the service equipment confirms that it has been connected to the target customer, it will perform service operations on the target customer according to the service plan.

2. The method for personalized bank branch services based on multi-dimensional recognition according to claim 1, characterized in that: The multidimensional data includes: Real-time collection of target customer image data, biometric information, and real-time location information at the current branch.

3. The method for personalized bank branch services based on multi-dimensional recognition according to claim 2, characterized in that: The data fusion process includes: The collected multidimensional data is preprocessed to obtain standardized data; Generate temporary unique identifiers for standardized multidimensional data, and perform data association on the standardized multidimensional data; Data identification is performed on standardized multidimensional data to obtain target customer profile data; Information is integrated from the target customer profile data to obtain a comprehensive target customer profile.

4. The method for personalized bank branch services based on multi-dimensional recognition according to claim 3, characterized in that: The preprocessing includes: The collected multidimensional data is cleaned and filtered to remove noisy data. Standardize multidimensional data by converting it into a unified standard format. Multidimensional data in a unified standard format is given a timestamp to obtain standardized multidimensional data.

5. The method for personalized bank branch services based on multi-dimensional recognition according to claim 4, characterized in that: The data identification includes: Transform the image data of the target customer into key feature vectors; The biometric information of target customers is identified, converted into identity feature data, and linked to account information; The real-time location information of target customers is identified and converted into real-time coordinate information within the network.

6. The method for personalized bank branch services based on multi-dimensional recognition according to claim 5, characterized in that: in S3, the decision-making process is as follows: based on the comprehensive profile of the target customer and historical service information, the needs of the target customer are predicted through a behavior prediction model; based on the predicted needs of the target customer, a service plan is generated through a recommendation algorithm.

7. The method for personalized bank branch services based on multi-dimensional recognition according to claim 1, characterized in that: S5 In this process, after performing service operations on target customers according to the service plan, the behavioral responses of target customers to the service plan are recorded and fed back to the behavior prediction model.

8. A personalized service system for bank branches based on multi-dimensional recognition, characterized by including: The multi-dimensional information collection and recognition module is used to collect and recognize multi-dimensional data of target customers in real time when they enter the current branch. The central data processing module is used to perform real-time data fusion processing on multi-dimensional data, and to associate multi-dimensional data through unique identifiers to generate a comprehensive profile of the target customer. Based on a comprehensive profile of the target customer, service decisions are made and service plans are generated. Push service plans to service devices that match the real-time location of the target customer; Service equipment is used to perform service operations on the target customer according to the service plan after confirming connection.

9. A computer device, characterized by: include: A processor and a memory, the memory storing a computer program executable by the processor, wherein the processor, when executing the computer program, implements the method of any one of claims 1-7.

Images