Financial account activation method and device, program product and electronic equipment

Abstract

The invention discloses a financial account activation method and device, a program product and electronic equipment, and relates to the field of financial science and technology, and the method comprises the steps: inputting target user data corresponding to a to-be-activated account into a target model, and obtaining a target probability and a target moment, the target probability is a probability that a target user corresponding to the to-be-activated account activates the account at a target moment, then, under the condition that the target probability is greater than or equal to a preset probability, generating a target program based on the target moment and the target user data, and activating the to-be-activated account through the target program, the target program is used for interacting with a target user through a preset mode, and the preset mode is a voice outbound mode, a short message prompt mode or a mail link mode. According to the method and the device, the technical problem that in a financial institution, outbound prompting is regularly carried out on a user of a financial account in a non-enabled state through a traditional outbound system, so that the user experience degree is low is solved.

Description

Technical Field

[0001] The present application relates to the field of financial technology, and specifically, to a method, device, program product, and electronic device for activating a financial account. Background Art

[0002] With the rapid development of financial technology, digital financial accounts have become an integral part of people's daily lives. However, many users fail to activate and use their registered financial accounts promptly for various reasons, resulting in accounts remaining inactive for extended periods. This situation not only deprives users of immediate access to financial services, diminishing the user experience, but also wastes resources for financial institutions, impacting their business performance and regulatory compliance.

[0003] Currently, most financial institutions employ traditional strategies to activate inactive accounts, relying primarily on customer-initiated activation mechanisms or widespread, blind, intelligent outbound calls, text messages, or email reminders. The former is inefficient because banks lack visibility into when customers are willing to activate their accounts. While the latter attempts proactive customer engagement, the lack of personalized and accurate activation prediction capabilities often results in subpar results. Excessive outbound calls and push notifications not only consume significant resources but also create customer frustration and diminish the user experience.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] The present application provides a method, apparatus, program product, and electronic device for activating a financial account, to at least address the technical problem of low user experience caused by regular outbound call reminders to users of inactive financial accounts through traditional outbound call systems in financial institutions.

[0006] According to one aspect of the present application, a method for activating a financial account is provided, comprising: inputting target user data corresponding to an account to be activated into a target model to obtain a target probability and a target time, wherein the account to be activated is a financial account registered by the target user and in an inactive state, the target model is a neural network model trained based on historical account data and historical user data corresponding to L users, the target probability is the probability that the target user corresponding to the account to be activated activates the account at the target time, and L is a positive integer; when the target probability is greater than or equal to a preset probability, generating a target program based on the target time and target user data, and activating the account to be activated through the target program, wherein the target program is used to interact with the target user in a preset manner, which is a voice outbound call manner, a text message prompt manner, or an email link manner.

[0007] Optionally, the target model is trained by the following steps: taking the historical account data and historical user data corresponding to L users as the initial training set, performing a first operation and a second operation on the initial training set to obtain a target training set, wherein the first operation is used to update missing parameters and abnormal parameters in the initial training set, and the second operation is used to unify the format of the data obtained by the first operation; performing feature extraction on the data in the target training set to obtain L first vectors and L second vectors, wherein each first vector is used to represent the status, consumption frequency, and consumption cycle of the historical account corresponding to one of the L users, and each second vector is used to represent the gender, age, financial outlet, and credit score of one of the L users; splicing each first vector in the L first vectors with the corresponding second vector in the L second vectors to obtain L training vectors; and generating a target model based on the L training vectors.

[0008] Optionally, the step of generating a target model based on L training vectors includes: randomly screening the L training vectors M times to obtain M training subsets, wherein each training subset includes N training vectors, M and N are both positive integers, and N is less than or equal to L; iteratively training each training subset in the M training subsets based on a preset algorithm to obtain M target decision trees, wherein each target decision tree corresponds to a weight value; and generating a target model based on the M target decision trees and the weight value of each target decision tree.

[0009] Optionally, the step of iteratively training each of the M training subsets based on a preset algorithm includes: dividing the N training vectors included in the i-th training subset into a first set and a second set corresponding to the i-th training subset, where i is a positive integer less than or equal to M; training an initial decision tree based on the training vectors in the first set to obtain a first decision tree; and updating the weight values ​​of the first decision tree based on the training vectors in the second set to obtain a target decision tree corresponding to the i-th training subset.

[0010] Optionally, when the preset method is a voice outbound call method, the step of generating a target program based on the target time and target user data includes: determining a first element and a second element based on the target user data, wherein the first element includes at least the target user's name, gender and contact number, and the second element includes at least the target user's contact number, user ID and ID of the account to be activated; generating a target program based on the target time, the first element and the second element.

[0011] Optionally, when the preset method is a voice outbound call method, the step of activating the account to be activated through the target program includes: controlling the outbound call system at the target time through the target program, and performing an outbound call operation on the target user based on the first factor, wherein the outbound call operation is used to prompt the target user to activate the account to be activated; after the target user allows the activation of the account to be activated, collecting a first verification code, a second verification code and a third verification code through the outbound call system, wherein the first verification code is the P-digit tail number of the communication number entered by the target user in real time, the second verification code is the user identifier entered by the target user in real time, and the third verification code is the identifier of the account to be activated entered by the target user in real time; performing security verification based on the first verification code, the second verification code, the third verification code and the second factor; if the security verification passes, updating the account to be activated to an activated state.

[0012] Optionally, the step of performing security verification based on the first verification code, the second verification code, the third verification code and the second factor includes: comparing the first verification code, the second verification code, the third verification code and the second factor; determining that the security verification is passed when the first verification code is consistent with the P-digit suffix of the communication number included in the second factor, the second verification code is consistent with the user identifier included in the second factor, and the third verification code is consistent with the identifier of the account to be activated included in the second factor.

[0013] According to another aspect of an embodiment of the present application, a financial account activation device is also provided, including: an input unit, used to input target user data corresponding to the account to be activated into a target model to obtain a target probability and a target time, wherein the account to be activated is a financial account registered by the target user and is in an unactivated state, the target model is a neural network model trained based on historical account data and historical user data corresponding to L users, the target probability is the probability that the target user corresponding to the account to be activated activates the account at the target time, and L is a positive integer; a program generation unit, used to generate a target program based on the target time and target user data when the target probability is greater than or equal to a preset probability, and activate the account to be activated through the target program, wherein the target program is used to interact with the target user through a preset method, which is a voice outbound call method, a text message prompt method, or an email link method.

[0014] According to another aspect of the present application, a computer program product is provided, in which a computer program is stored. When the computer program is run, the computer program product is controlled to execute any one of the above-mentioned methods for activating a financial account.

[0015] According to another aspect of the present application, an electronic device is also provided, wherein the electronic device includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by one or more processors, the one or more processors implement any of the above-mentioned methods for activating a financial account.

[0016] In the present application, first, the target user data corresponding to the account to be activated is input into the target model to obtain the target probability and target time, wherein the account to be activated is a financial account that has been registered by the target user and is in an unactivated state, and the target model is a neural network model trained based on the historical account data and historical user data corresponding to L users. The target probability is the probability that the target user corresponding to the account to be activated activates the account at the target time, and L is a positive integer. Afterwards, when the target probability is greater than or equal to the preset probability, the present application generates a target program based on the target time and target user data, and activates the account to be activated through the target program, wherein the target program is used to interact with the target user through a preset method, which is a voice call method, a text message prompt method, or an email link method.

[0017] From the above content, it can be seen that this application predicts the probability and time of target users activating their accounts through a pre-built target model, thereby achieving the purpose of accurately understanding the target users' willingness to activate their accounts, thereby achieving the purpose of intelligently activating accounts to be activated in financial institutions and providing personalized reminders to target users. The technical solution of this application can identify subtle differences in the willingness of different users to activate their accounts, and through the target model, predict which accounts to be activated may be activated at what time based on the target users' behavioral history data and account status data (i.e., target user data).

[0018] Compared with the existing technology, the technical solution of the present application does not make regular outbound calls to all users whose accounts are to be activated. Instead, it makes an outbound call only when it is known that the probability of the target user activating the account at the target time is greater than the preset probability. This avoids the problem of making outbound call reminders to users who have no intention of activating their accounts in the existing technology, achieves the technical effect of improving user experience, and further solves the technical problem of low user experience caused by regular outbound call reminders to users of financial accounts that are not activated in financial institutions through traditional outbound call systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1is a flowchart of an optional method for activating a financial account according to an embodiment of the present application;

[0021] Figure 2 This is a flowchart of an optional outbound call database activation method for a financial account according to an embodiment of the present application;

[0022] Figure 3 is a schematic diagram of an optional financial account activation device according to an embodiment of the present application;

[0023] Figure 4 This is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] It should also be noted that the relevant information (including but not limited to information used for display and analysis) and data (including but not limited to target user data) involved in this application are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set up between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain the relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0027] In addition, the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant information and data involved in this application comply with the relevant laws, regulations and standards of the relevant regions, and necessary confidentiality measures have been taken, and do not violate public order and good morals. In addition, this application provides corresponding operation entrances for users to choose to agree to authorization or refuse authorization. If the user chooses to refuse authorization, he / she will enter the corresponding expert decision-making process.

[0028] According to an embodiment of the present application, an embodiment of a method for activating a financial account is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0029] This application provides a financial account activation system (hereinafter referred to as the activation system) for executing the financial account activation method of this application. Figure 1 is a flow chart of an optional method for activating a financial account according to an embodiment of the present application, such as Figure 1 As shown, the method includes the following steps:

[0030] Step S101: Input the target user data corresponding to the account to be activated into the target model to obtain the target probability and target time, where the account to be activated is a financial account registered by the target user but not yet activated. The target model is a neural network model trained based on historical account data and historical user data corresponding to L users. The target probability is the probability that the target user corresponding to the account to be activated will activate the account at the target time, where L is a positive integer.

[0031] Optionally, the target user data includes at least the age, gender, identity identifier, contact information, credit status, historical transaction records, and time series data of account status changes of the target user corresponding to the account to be activated.

[0032] Optionally, the activation system regularly checks the account information in the database of the financial institution, and updates the account status of financial accounts that have no user active transactions within a continuous preset time period and have loan overdraft risks to a pending activation status.

[0033] Optionally, the activation system can regularly screen Q accounts to be activated from the database of the financial institution, where Q is a positive integer. Then, the target user data corresponding to each of the Q accounts to be activated is input into the target model, thereby predicting the probability and time of user account activation.

[0034] Optionally, by applying the above-mentioned intelligent activation strategy, financial institutions can reduce the number of blind outbound calls and information pushes compared to traditional regular outbound calling methods, thereby saving resources of the financial institution's outbound calling system and information push platform and reducing operating costs.

[0035] Step S102: When the target probability is greater than or equal to the preset probability, a target program is generated based on the target time and the target user data, and the account to be activated is activated through the target program, wherein the target program is used to interact with the target user through a preset method, which is a voice call method, a text message prompt method, or an email link method.

[0036] Optionally, compared with the traditional periodic outbound calling method, the activation system only controls the outbound calling system or information push platform to reach the target user at the target time by running the target program, thereby reducing the frequency of invalid contact with the target user. While increasing the activation rate of accounts to be activated, it further reduces the resource utilization and operating costs of the outbound calling system and information push platform.

[0037] Optionally, when selecting the method of interaction with the target user, the activation system can also analyze the target user data to obtain the target user's preferred communication method. For example, for target users who use online applications more frequently than a preset frequency, the activation system gives priority to SMS prompts or email links to interact with the target user, while for target users who prefer direct voice communication, the activation system gives priority to voice outbound calls, thereby achieving the purpose of reaching the user based on the user's preferred communication method, thereby improving the user experience.

[0038] From the above content, it can be seen that this application predicts the probability and time of target users activating their accounts through a pre-built target model, thereby achieving the purpose of accurately understanding the target users' willingness to activate their accounts, thereby achieving the purpose of intelligently activating accounts to be activated in financial institutions and providing personalized reminders to target users. The technical solution of this application can identify subtle differences in the willingness of different users to activate their accounts, and through the target model, predict which accounts to be activated may be activated at what time based on the target users' behavioral history data and account status data (i.e., target user data).

[0039] Compared with the existing technology, the technical solution of the present application does not make regular outbound calls to all users whose accounts are to be activated. Instead, it makes an outbound call only when it is known that the probability of the target user activating the account at the target time is greater than the preset probability. This avoids the problem of making outbound call reminders to users who have no intention of activating their accounts in the existing technology, achieves the technical effect of improving user experience, and further solves the technical problem of low user experience caused by regular outbound call reminders to users of financial accounts that are not activated in financial institutions through traditional outbound call systems.

[0040] In an optional embodiment, the training steps of the target model include: first, the activation system takes the historical account data and historical user data corresponding to L users as the initial training set, performs a first operation and a second operation on the initial training set to obtain a target training set, wherein the first operation is used to update the missing parameters and abnormal parameters in the initial training set, and the second operation is used to unify the format of the data obtained by the first operation. Afterwards, the activation system performs feature extraction on the data in the target training set to obtain L first vectors and L second vectors, wherein each first vector is used to represent the status, consumption frequency, and consumption cycle of the historical account corresponding to one of the L users, and each second vector is used to represent the gender, age, financial outlet, and credit score of one of the L users. Then, the activation system splices each first vector in the L first vectors with the corresponding second vector in the L second vectors to obtain L training vectors. Finally, the activation system generates a target model based on the L training vectors.

[0041] Optionally, during the execution of the first operation, the activation system can fill in the missing parameters in the initial training set by interpolation, and can correct the identified abnormal values ​​that are far beyond the normal range by replacement or elimination; during the execution of the second operation, the activation system can unify the date format included in the data processed by the first operation, and then convert it into a unified text data encoding format.

[0042] Optionally, through the first operation and the second operation, the activation system can ensure that the processed data is presented in a consistent format, and can remove noise in the original data, converting the original data into a cleaner, structured and model-friendly encoding format to facilitate subsequent model understanding and machine learning.

[0043] Optionally, the first vector focuses on the user's account behavior, such as changes in account status, consumption frequency, and consumption cycle, reflecting the user's habits and activity level in using the account; the second vector covers the user's static attributes, including gender, age, financial branch, and credit score, providing an overview of the user's basic profile and credit level. The activation system converts complex and multi-dimensional user data into vector form, making it easier for subsequent models to understand and analyze the comprehensive impact of user characteristics.

[0044] Optionally, the activation system splices and merges each user's account behavior characteristics (i.e., the first vector) and their personal attribute characteristics (i.e., the second vector) into a unified training vector, wherein the splicing operation enables each piece of training data to contain comprehensive information on user behavior and attributes, thereby providing the model with a richer and more comprehensive data perspective and enhancing the predictive ability of the trained target model.

[0045] In an optional embodiment, the activation system first randomly screens L training vectors M times to obtain M training subsets, where each training subset includes N training vectors, M and N are both positive integers, and N is less than or equal to L. Afterwards, the activation system iteratively trains each of the M training subsets based on a preset algorithm to obtain M target decision trees, where each target decision tree corresponds to a weight value. Then, the activation system generates a target model based on the M target decision trees and the weight value of each target decision tree.

[0046] Optionally, the preset algorithm can be set to a logistic regression algorithm, a random forest algorithm, and an SVM (Support Vector Machine) algorithm.

[0047] Optionally, during the iterative training process, the activation system can split each training subset into a training set (i.e., a first set) and a test set (a second set), and use cross-validation technology to improve the generalization ability of the trained target model. Afterwards, the activation system uses grid search or other parameter optimization methods to adjust the algorithm hyperparameters (such as regularization parameters and tree depth). Finally, the activation system statistically calculates the performance indicators of the target model, such as accuracy, recall, F1-Score, and AUC-ROC curve (Area Under the Receiver Operating Characteristic Curve, a curve used to show the model's ability to recognize positive and negative classes at different thresholds).

[0048] Optionally, the activation system can randomly select training vectors that occupy a preset proportion of the L training vectors from the L training vectors as a training subset by random sampling.

[0049] Optionally, the activation system creates diverse training subsets through multiple random samplings, which can increase the diversity of the trained target model, reduce the risk of overfitting, and thus improve the generalization ability of the model for probabilistic prediction.

[0050] Optionally, during the training process, each training subset will be used to train an independent decision tree model. In this way, the activation system will eventually obtain M target decision trees that have been iteratively trained, where each target decision tree represents an account activation pattern. At the same time, a weight value will be assigned to each decision tree during the training process. The weight value reflects the degree of trust of the target model in the target decision tree. The weight value is a quantitative indicator for evaluating the contribution of the target decision tree. The size of the weight value determines the influence of the conclusion drawn by each target decision tree in the final target model.

[0051] Optionally, the activation system obtains a target model by integrating M target decision trees and their respective weight values. Through model integration technology, not only the predictive ability of the integrated target model is enhanced, but also its robustness is improved, thereby ensuring that it can maintain good predictive performance even in the face of unevenly distributed data sets or noise.

[0052] In an optional embodiment, the activation system first divides the N training vectors included in the i-th training subset into a first set and a second set corresponding to the i-th training subset, where i is a positive integer less than or equal to M. Then, the activation system trains the initial decision tree based on the training vectors in the first set to obtain a first decision tree. Then, the activation system updates the weight values ​​of the first decision tree based on the training vectors in the second set to obtain a target decision tree corresponding to the i-th training subset.

[0053] Optionally, the i-th training subset here refers to one of the subsets obtained in the M random screening processes, where i identifies the position of the training subset in the overall training process.

[0054] Optionally, the activation system can process the training data in stages by splitting each training subset into a first set and a second set again, that is, first using the data in the first set to train the initial decision tree to form a preliminary learning and understanding of the data, and then using the data in the second set to optimize the weight distribution of the decision tree, thereby improving the accurate cognition and adaptability of the data characteristics.

[0055] Optionally, the activation system adjusts the weight values ​​of the first decision tree by utilizing the data in the second set, which is essentially a correction and optimization of the model's prediction accuracy. Through continuous iterative updates, the target decision tree can better adapt to the characteristics of the data set and more closely fit the real data distribution, thereby reducing prediction bias and improving prediction accuracy and reliability. This series of operations not only strengthens the capabilities of a single decision tree, but also provides high-quality components for subsequent model integration, helping to build a more intelligent and efficient account activation prediction model.

[0056] In an optional embodiment, the activation system first determines the first factor and the second factor based on the target user data, wherein the first factor includes at least the target user's name, gender and contact number, and the second factor includes at least the target user's contact number, user ID and ID of the account to be activated. Afterwards, the activation system generates the target program based on the target time, the first factor and the second factor.

[0057] Optionally, when the preset method is a voice outbound call method, the activation system controls the outbound call system to perform an outbound call operation on the target user based on the first factor. Here, the first factor is the key channel for the outbound call system to reach the target user.

[0058] Optionally, when the preset method is a voice outbound call method, after the activation system contacts the target user through the outbound call system, if the target user allows the account to be activated, the activation system needs to compare the collected communication number, user ID and ID of the account to be activated input in real time with the second factor, so as to achieve the purpose of securely verifying the identity of the target user.

[0059] Optionally, by constructing the first element and the second element, it can ensure that the account activation instructions can be correctly sent and received by the operating end of the corresponding account. At the same time, it is also convenient for financial institutions to track the account activation process internally. The activation system defines the key user data elements required for the activation process and generates a personalized target program based on the target moment predicted by the model, thereby realizing effective activation of the account to be activated. Among them, the determination of the first element and the second element ensures the customization and personalization of the activation strategy, the prediction of the target moment optimizes the activation timing, and the generation of the target program converts this information into actual operational guidance, thereby improving the efficiency of the activation process and user experience.

[0060] In an optional embodiment, when the preset method is a voice outbound call method, the activation system first controls the outbound call system at the target time through the target program, and performs an outbound call operation on the target user based on the first factor, wherein the outbound call operation is used to prompt the target user to activate the account to be activated. Afterwards, after the target user allows the activation of the account to be activated, the activation system collects the first verification code, the second verification code and the third verification code through the outbound call system, wherein the first verification code is the P-digit tail number of the communication number entered by the target user in real time, the second verification code is the user identification entered by the target user in real time, and the third verification code is the identification of the account to be activated entered by the target user in real time. Then, the activation system performs security verification based on the first verification code, the second verification code, the third verification code and the second factor. Finally, if the security verification passes, the activation system updates the account to be activated to an activated state.

[0061] Optionally, the activation system can ensure accurate contact with the target user by controlling the outbound call system based on the first factor, wherein the outbound call content includes prompting the target user to start the activation process of his / her account to be activated. The activation system accurately controls the outbound call system through the target program and communicates with the target user at the predicted best time (i.e., the target moment), which not only avoids invalid outbound call attempts, but also enhances the user's experience of the financial institution's services, thereby improving the response rate of activation of accounts to be activated.

[0062] Optionally, the activation system collects the first verification code, second verification code and third verification code input by the target user in real time, so that the outbound call system can further confirm the identity of the target user and the accuracy of the account to be activated after the user clearly expresses the intention to activate the account, thereby increasing the security of the activation process and reducing the probability of misoperation and the risk of user impersonation.

[0063] Optionally, the security verification process is a key step in the account activation process. Through security verification, the activation system not only verifies the identity of the target user, but also confirms the legitimacy of the activation request, thereby providing security for subsequent account activation operations and maintaining the asset security of the target user and the financial institution.

[0064] In an optional embodiment, the activation system first compares the first verification code, the second verification code, and the third verification code with the second factor. Then, if the first verification code is consistent with the P-digit suffix of the communication number included in the second factor, the second verification code is consistent with the user identifier included in the second factor, and the third verification code is consistent with the identifier of the account to be activated included in the second factor, the activation system determines that the security verification is passed.

[0065] Optionally, when the first verification code is inconsistent with the P-digit suffix of the communication number included in the second factor, the second verification code is inconsistent with the user identifier included in the second factor, or the third verification code is inconsistent with the identifier of the account to be activated included in the second factor, the activation system determines that the security verification has failed.

[0066] Optionally, by using the last digit of the P digit in the communication number as the first verification code, the user input is simplified while maintaining sufficient information uniqueness to distinguish the communication channel from other users.

[0067] Optionally, the activation system can verify the authenticity of the target user's identity and the accuracy of the activation request by comparing the three verification codes provided by the user in real time with the pre-stored second factor. This is an important security barrier in the activation process. By comparing the first, second, and third verification codes input in real time with the pre-stored second factor in detail, and applying a multi-condition verification mechanism, the activation system achieves double confirmation of the target user's identity and activation request, which not only improves the accuracy of security verification, but also ensures the legality and accuracy of the activation operation.

[0068] From the above content, it can be seen that this application predicts the probability and time of target users activating their accounts through a pre-built target model, thereby achieving the purpose of accurately understanding the target users' willingness to activate their accounts, thereby achieving the purpose of intelligently activating accounts to be activated in financial institutions and providing personalized reminders to target users. The technical solution of this application can identify subtle differences in the willingness of different users to activate their accounts, and through the target model, predict which accounts to be activated may be activated at what time based on the target users' behavioral history data and account status data (i.e., target user data).

[0069] Compared with the existing technology, the technical solution of the present application does not make regular outbound calls to all users whose accounts are to be activated. Instead, it makes an outbound call only when it is known that the probability of the target user activating the account at the target time is greater than the preset probability. This avoids the problem of making outbound call reminders to users who have no intention of activating their accounts in the existing technology, achieves the technical effect of improving user experience, and further solves the technical problem of low user experience caused by regular outbound call reminders to users of financial accounts that are not activated in financial institutions through traditional outbound call systems.

[0070] In an optional embodiment, Figure 2 This is a flow chart of an optional outbound call database activation method for a financial account according to an embodiment of the present application. Figure 2 As shown, the method includes the following steps:

[0071] First, the activation system filters out accounts to be activated from the database of the financial institution. Then, the target user data corresponding to the account to be activated is input into the target model to obtain prediction information, wherein the prediction information includes the activation probability and the activation time (for example, within one week, two weeks or four weeks). Then, the activation system makes an outbound call to the target user based on the prediction information. After the target user answers the call, the outbound call system asks the target user whether he agrees to activate the account. If the target user does not agree to activate the account, the outbound call is ended.

[0072] If the target user agrees to activate the account, the activation system controls access to the information database system and performs security verification. If the security verification passes, the activation operation is performed for the target user's pending account and the target user is informed that the account activation is successful. If the security verification fails, the target user is prompted that the activation operation is risky.

[0073] According to another aspect of the embodiments of the present application, a financial account activation device (hereinafter referred to as the activation device) is also provided. Figure 3 is a schematic diagram of an optional financial account activation device according to an embodiment of the present application, such as Figure 3 As shown, the financial account activation device includes: an input unit 301 and a program generation unit 302.

[0074] Optionally, the input unit 301 is used to input the target user data corresponding to the account to be activated into the target model to obtain a target probability and a target time, wherein the account to be activated is a financial account registered by the target user and is in an inactivated state, the target model is a neural network model trained based on historical account data and historical user data corresponding to L users, the target probability is the probability that the target user corresponding to the account to be activated activates the account at the target time, and L is a positive integer; the program generation unit 302 is used to generate a target program based on the target time and target user data when the target probability is greater than or equal to a preset probability, and activate the account to be activated through the target program, wherein the target program is used to interact with the target user through a preset method, which is a voice call method, a text message prompt method, or an email link method.

[0075] In an optional embodiment, the financial account activation device further includes: a determination unit, a feature extraction unit, a splicing unit, and a model generation unit.

[0076] Optionally, a determination unit is used to use the historical account data and historical user data corresponding to L users as an initial training set, perform a first operation and a second operation on the initial training set to obtain a target training set, wherein the first operation is used to update missing parameters and abnormal parameters in the initial training set, and the second operation is used to unify the format of the data obtained by the first operation; a feature extraction unit is used to extract features from the data in the target training set to obtain L first vectors and L second vectors, wherein each first vector is used to represent the status, consumption frequency, and consumption cycle of a historical account corresponding to one of the L users, and each second vector is used to represent the gender, age, financial outlet, and credit score of one of the L users; a splicing unit is used to splice each first vector in the L first vectors with the corresponding second vector in the L second vectors to obtain L training vectors; and a model generation unit is used to generate a target model based on the L training vectors.

[0077] In an optional embodiment, the model generation unit includes: a random screening subunit, an iterative training subunit and a model generation subunit.

[0078] Optionally, the random screening subunit is used to perform random screening on L training vectors M times to obtain M training subsets, wherein each training subset includes N training vectors, M and N are both positive integers, and N is less than or equal to L; the iterative training subunit is used to perform iterative training on each training subset in the M training subsets based on a preset algorithm to obtain M target decision trees, wherein each target decision tree corresponds to a weight value; the model generation subunit is used to generate a target model based on the M target decision trees and the weight value of each target decision tree.

[0079] In an optional embodiment, the iterative training subunit includes: a division module, an initial training module and an update module.

[0080] Optionally, a partitioning module is used to divide the N training vectors included in the i-th training subset into a first set and a second set corresponding to the i-th training subset, where i is a positive integer less than or equal to M; an initial training module is used to train an initial decision tree based on the training vectors in the first set to obtain a first decision tree; and an updating module is used to update the weight values ​​of the first decision tree based on the training vectors in the second set to obtain a target decision tree corresponding to the i-th training subset.

[0081] In an optional embodiment, when the preset mode is a voice outbound call mode, the program generation unit includes: an element determination subunit and a program generation subunit.

[0082] Optionally, the element determination subunit is used to determine the first element and the second element based on the target user data, wherein the first element includes at least the target user's name, gender and contact number, and the second element includes at least the target user's contact number, user ID and ID of the account to be activated; the program generation subunit is used to generate the target program based on the target time, the first element and the second element.

[0083] In an optional embodiment, when the preset mode is a voice outbound call mode, the program generation unit further includes: an outbound call control subunit, a verification code collection subunit, a security verification subunit, and an update subunit.

[0084] Optionally, the outbound call control subunit is used to control the outbound call system at the target time through the target program, and perform an outbound call operation on the target user based on the first factor, wherein the outbound call operation is used to prompt the target user to activate the account to be activated; the verification code collection subunit is used to collect the first verification code, the second verification code and the third verification code through the outbound call system after the target user allows the activation of the account to be activated, wherein the first verification code is the P-digit tail number of the communication number entered by the target user in real time, the second verification code is the user identification entered by the target user in real time, and the third verification code is the identification of the account to be activated entered by the target user in real time; the security verification subunit is used to perform security verification based on the first verification code, the second verification code, the third verification code and the second factor; the update subunit is used to update the account to be activated to the activated state if the security verification passes.

[0085] In an optional embodiment, the security verification subunit includes: a comparison module and a determination module.

[0086] Optionally, the comparison module is used to compare the first verification code, the second verification code, the third verification code and the second factor; the determination module is used to determine that the security verification is passed when the first verification code is consistent with the P-digit suffix of the communication number included in the second factor, the second verification code is consistent with the user identifier included in the second factor, and the third verification code is consistent with the identifier of the account to be activated included in the second factor.

[0087] From the above content, it can be seen that the activation device predicts the probability and time of the target user activating the account through a pre-built target model, thereby achieving the purpose of accurately understanding the target user's willingness to activate the account, thereby achieving the purpose of intelligently activating the accounts to be activated in the financial institution and providing personalized reminders to the target user. The technical solution of the activation device can identify subtle differences in the willingness of different users to activate their accounts, and predict which accounts to be activated and at what time they may be activated based on the target user's behavioral history data and account status data (i.e., target user data) through the target model.

[0088] Compared with the existing technology, the technical solution of the activation device does not make regular outbound calls to all users whose accounts are to be activated. Instead, it only makes outbound calls when it is known that the probability of the target user activating the account at the target time is greater than the preset probability. This avoids the problem of making outbound call reminders to users who have no intention of activating their accounts in the existing technology, achieves the technical effect of improving user experience, and further solves the technical problem of low user experience caused by regular outbound call reminders to users of financial accounts that are not activated in financial institutions through traditional outbound call systems.

[0089] According to another aspect of an embodiment of the present application, a computer program product is further provided, which includes a stored computer program, wherein when the computer program is running, the computer program product is controlled to execute any one of the above-mentioned methods for activating a financial account.

[0090] According to another aspect of the embodiments of the present application, an electronic device is provided. Figure 4 is a structural block diagram of an electronic device according to an embodiment of the present application, such as Figure 4 As shown, the electronic device may include: one or more ( Figure 4 Only one is shown) processor 402, memory 404, storage controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0091] Among them, the memory can be used to store software programs and modules, such as program instructions / modules corresponding to the methods and devices in the embodiments of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implementing the above-mentioned method. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network and a combination thereof.

[0092] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: input the target user data corresponding to the account to be activated into the target model to obtain the target probability and target time, wherein the account to be activated is a financial account registered by the target user and is in an unactivated state, the target model is a neural network model trained based on historical account data and historical user data corresponding to L users, the target probability is the probability that the target user corresponding to the account to be activated activates the account at the target time, and L is a positive integer; when the target probability is greater than or equal to the preset probability, a target program is generated based on the target time and target user data, and the account to be activated is activated through the target program, wherein the target program is used to interact with the target user in a preset manner, which is a voice call method, a text message prompt method, or an email link method.

[0093] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: taking the historical account data and historical user data corresponding to L users as the initial training set, performing a first operation and a second operation on the initial training set to obtain a target training set, wherein the first operation is used to update the missing parameters and abnormal parameters in the initial training set, and the second operation is used to unify the format of the data obtained by the first operation; performing feature extraction on the data in the target training set to obtain L first vectors and L second vectors, wherein each first vector is used to represent the status, consumption frequency, and consumption cycle of the historical account corresponding to one of the L users, and each second vector is used to represent the gender, age, financial outlet, and credit score of one of the L users; splicing each first vector in the L first vectors with the corresponding second vector in the L second vectors to obtain L training vectors; and generating a target model based on the L training vectors.

[0094] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: randomly screen L training vectors M times to obtain M training subsets, where each training subset includes N training vectors, M and N are both positive integers, and N is less than or equal to L; iteratively train each training subset in the M training subsets based on a preset algorithm to obtain M target decision trees, where each target decision tree corresponds to a weight value; generate a target model based on the M target decision trees and the weight value of each target decision tree.

[0095] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: divide the N training vectors included in the i-th training subset into a first set and a second set corresponding to the i-th training subset, where i is a positive integer less than or equal to M; train the initial decision tree based on the training vectors in the first set to obtain a first decision tree; update the weight value of the first decision tree based on the training vectors in the second set to obtain a target decision tree corresponding to the i-th training subset.

[0096] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: determine the first factor and the second factor based on the target user data, wherein the first factor includes at least the name, gender and contact number of the target user, and the second factor includes at least the contact number, user ID and ID of the account to be activated of the target user; generate the target program based on the target time, the first factor and the second factor.

[0097] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: control the outbound call system at the target time through the target program, and perform an outbound call operation on the target user based on the first factor, wherein the outbound call operation is used to prompt the target user to activate the account to be activated; after the target user allows the activation of the account to be activated, collect the first verification code, the second verification code and the third verification code through the outbound call system, wherein the first verification code is the P-digit tail number of the communication number entered by the target user in real time, the second verification code is the user identification entered by the target user in real time, and the third verification code is the identification of the account to be activated entered by the target user in real time; perform security verification based on the first verification code, the second verification code, the third verification code and the second factor; if the security verification passes, update the account to be activated to the activated state.

[0098] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: compare the first verification code, the second verification code, and the third verification code with the second factor; when the first verification code is consistent with the P-digit suffix of the communication number included in the second factor, the second verification code is consistent with the user identifier included in the second factor, and the third verification code is consistent with the identifier of the account to be activated included in the second factor, it is determined that the security verification is passed.

[0099] An embodiment of the present application provides a financial account activation solution. The embodiment of the present application predicts the probability and time of the target user's account activation through a pre-built target model, thereby achieving the purpose of accurately understanding the target user's willingness to activate the account, thereby achieving the purpose of intelligently activating accounts to be activated in financial institutions and providing personalized reminders to target users. The technical solution of the embodiment of the present application can identify subtle differences in the willingness of different users to activate accounts, and predict which accounts to be activated and at what time may be activated based on the target user's behavioral history data and account status data (i.e., target user data) through the target model.

[0100] Compared with the prior art, the technical solution of the embodiment of the present application does not make regular outbound calls to all users whose accounts are to be activated. Instead, it makes an outbound call only when it is known that the probability of the target user activating the account at the target time is greater than the preset probability. This avoids the problem of making outbound call reminders to users who have no intention of activating their accounts in the prior art, achieves the technical effect of improving user experience, and further solves the technical problem of low user experience caused by regular outbound call reminders to users of financial accounts that are not activated in financial institutions through traditional outbound call systems.

[0101] It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only, and the electronic device may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (MID), a PAD, or other terminal devices. Figure 4 It does not limit the structure of the above electronic device. For example, the electronic device may also include Figure 4 More or fewer components (such as network interfaces, display devices, etc.) shown in, or with Figure 4 Different configurations shown.

[0102] A person skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0103] The embodiment of the present application further provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the service reservation method provided in the first embodiment.

[0104] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0105] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing the steps of the service reservation method.

[0106] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0107] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0108] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0109] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0110] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0111] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0112] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for activating a financial account, characterized in that: include: Input the target user data corresponding to the account to be activated into the target model to obtain a target probability and a target time, wherein the account to be activated is a financial account registered by the target user but not yet activated. The target model is a neural network model trained based on historical account data and historical user data corresponding to L users. The target probability is the probability that the target user corresponding to the account to be activated will activate the account at the target time, where L is a positive integer. When the target probability is greater than or equal to a preset probability, a target program is generated based on the target time and the target user data, and the account to be activated is activated through the target program, wherein the target program is used to interact with the target user through a preset method, and the preset method is a voice call method, a text message prompt method, or an email link method.

2. The method for activating a financial account according to claim 1, wherein: The target model is trained by the following steps: Using the historical account data and historical user data corresponding to the L users as an initial training set, performing a first operation and a second operation on the initial training set to obtain a target training set, wherein the first operation is used to update missing parameters and abnormal parameters in the initial training set, and the second operation is used to unify the format of the data obtained by the first operation; Performing feature extraction on the data in the target training set to obtain L first vectors and L second vectors, wherein each first vector is used to represent the status, consumption frequency, and consumption cycle of a historical account corresponding to one of the L users, and each second vector is used to represent the gender, age, financial institution affiliated, and credit score of one of the L users; Concatenate each first vector of the L first vectors with a corresponding second vector of the L second vectors to obtain L training vectors; The target model is generated based on the L training vectors.

3. The method for activating a financial account according to claim 2, wherein: Generating the target model based on the L training vectors includes: Performing M random screening on the L training vectors to obtain M training subsets, wherein each training subset includes N training vectors, M and N are both positive integers, and N is less than or equal to L; Iteratively training each of the M training subsets based on a preset algorithm to obtain M target decision trees, wherein each target decision tree corresponds to a weight value; The target model is generated according to the M target decision trees and the weight value of each target decision tree.

4. The method for activating a financial account according to claim 3, wherein: Performing iterative training on each of the M training subsets based on a preset algorithm includes: Divide the N training vectors included in the i-th training subset into a first set and a second set corresponding to the i-th training subset, where i is a positive integer less than or equal to M; Training the initial decision tree based on the training vectors in the first set to obtain a first decision tree; The weight values ​​of the first decision tree are updated based on the training vectors in the second set to obtain a target decision tree corresponding to the i-th training subset.

5. The method for activating a financial account according to claim 1, wherein: When the preset mode is the voice outbound call mode, generating a target program based on the target time and the target user data includes: Determining a first factor and a second factor based on the target user data, wherein the first factor includes at least the target user's name, gender, and contact number, and the second factor includes at least the target user's contact number, user ID, and ID of the account to be activated; The target program is generated based on the target time, the first element, and the second element.

6. The method for activating a financial account according to claim 5, characterized in that: When the preset method is the voice outbound call method, activating the account to be activated through the target program includes: controlling the outbound call system at the target time by the target program, and performing an outbound call operation on the target user based on the first factor, wherein the outbound call operation is used to prompt the target user to activate the account to be activated; After the target user allows activation of the account to be activated, a first verification code, a second verification code, and a third verification code are collected through the outbound call system, wherein the first verification code is the P-digit suffix of the communication number input by the target user in real time, the second verification code is the user ID input by the target user in real time, and the third verification code is the ID of the account to be activated input by the target user in real time; Perform security verification based on the first verification code, the second verification code, the third verification code, and the second factor; When the security verification is passed, the account to be activated is updated to an activated state.

7. The method for activating a financial account according to claim 6, wherein: Performing security verification based on the first verification code, the second verification code, the third verification code, and the second factor includes: comparing the first verification code, the second verification code, the third verification code, and the second factor; If the first verification code is consistent with the P-digit suffix in the communication number included in the second factor, the second verification code is consistent with the user identifier included in the second factor, and the third verification code is consistent with the identifier of the account to be activated included in the second factor, it is determined that the security verification is passed.

8. A financial account activation device, characterized in that: include: an input unit, configured to input target user data corresponding to the account to be activated into a target model to obtain a target probability and a target time, wherein the account to be activated is a financial account registered by the target user but not yet activated, the target model is a neural network model trained based on historical account data and historical user data corresponding to L users, the target probability is the probability that the target user corresponding to the account to be activated will activate the account at the target time, and L is a positive integer; A program generation unit is configured to generate a target program based on the target time and the target user data when the target probability is greater than or equal to a preset probability, and activate the account to be activated through the target program, wherein the target program is configured to interact with the target user through a preset method, which is a voice call method, a text message prompt method, or an email link method.

9. A computer program product, characterized in that The computer program product comprises a computer program, wherein when the computer program is executed, the computer program product is controlled to execute the method for activating a financial account according to any one of claims 1 to 7.

10. An electronic device, characterized in that: The invention comprises one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method for activating a financial account as described in any one of claims 1 to 7.

Images