Credit card user interaction method and device based on digital human, equipment and medium

Abstract

The invention provides a credit card user interaction method and device based on a digital human, equipment, a medium and a program product, and relates to the technical field of artificial intelligence. The method comprises the following steps: receiving a first interaction operation of a first card holding user; in response to the first interaction operation, a first category to which the first card holding user belongs is obtained, the first category is one of k categories, and each category corresponds to a combination of values in two dimensions of value and risk; a first digital person corresponding to the first category is obtained, the k categories are in one-to-one correspondence with the k digital persons, and the k digital persons are different from one another; and interacting with the first card-holding user through the first digital person.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular, to a credit card user interaction method, device, equipment, medium and program product based on digital human. Background Art

[0002] At present, when digital humans are used to interact with users in the credit card business, they mainly answer customers' questions, which is relatively passive and single. The diversity and liveliness of digital humans' interactions are not fully utilized. The user interaction method is poor and not very targeted, and its actual role in promoting the development of credit card business is limited. Summary of the invention

[0003] In view of this, the present disclosure provides a credit card user interaction method, device, equipment and medium based on digital human that can provide different interaction modes for different categories of cardholders.

[0004] In a first aspect of an embodiment of the present disclosure, a credit card user interaction method based on a digital person is provided. The method comprises: receiving a first interaction operation of a first cardholder user; in response to the first interaction operation, obtaining a first category to which the first cardholder user belongs, wherein the first category is one of k categories, each category corresponds to a combination of values ​​of two dimensions of value and risk, and k is an integer greater than 1; obtaining a first digital person corresponding to the first category, wherein the k categories correspond one-to-one to the k digital persons, wherein the k digital persons are different from each other; and interacting with the first cardholder user through the first digital person.

[0005] According to an embodiment of the present disclosure, obtaining the first digital human corresponding to the first category includes: obtaining a preset digital human image; obtaining a first interactive attribute corresponding to the first category; the interactive attributes corresponding to the k categories are different from each other; and configuring the digital human image using the first interactive attribute to obtain the first digital human.

[0006] According to an embodiment of the present disclosure, the interaction attributes include speech.

[0007] According to an embodiment of the present disclosure, the value ranges of the two dimensions of value and risk are both discrete level information, and the obtaining of the first interactive attribute corresponding to the first category includes: obtaining the level information of the value dimension in the first category to obtain the first value level; obtaining the level information of the risk dimension in the first category to obtain the first risk level; obtaining the first speech preset for the first value level, wherein a corresponding speech is preset for each level information in the value dimension; obtaining the second speech preset for the first risk level, wherein a corresponding speech is preset for each level information in the risk dimension; based on the first speech and the second speech, the first interactive attribute is obtained.

[0008] According to an embodiment of the present disclosure, the first category to which the first cardholder belongs is obtained in the following manner: upon obtaining the authorization of the first cardholder to use his transaction data, the transaction data of the first cardholder is obtained to obtain the first transaction data; value dimension transaction data and risk dimension transaction data are screened out from the first transaction data according to preset value dimension screening conditions and risk dimension screening conditions; features of the first cardholder are extracted from the value dimension transaction data and the risk dimension transaction data to obtain a first feature vector; the first feature vector is clustered by a trained clustering model to obtain the first cluster to which the first cardholder belongs; wherein the clustering result of the clustering model is k clusters; and the first category is determined based on the values ​​of the two dimensions of value and risk corresponding to the first cluster.

[0009] According to an embodiment of the present disclosure, extracting the characteristics of the first cardholder user from the value dimension transaction data and the risk dimension transaction data to obtain a first feature vector includes: converting the value dimension transaction data based on a coding rule corresponding to the value dimension screening condition to obtain a value characteristic; converting the risk dimension transaction data based on a coding rule corresponding to the risk dimension screening condition to obtain a risk characteristic; and combining the value characteristic and the risk characteristic to obtain a first feature vector.

[0010] According to an embodiment of the present disclosure, determining the first category based on the values ​​of the two dimensions of value and risk corresponding to the first cluster includes: obtaining a list of values ​​of the two dimensions of value and risk corresponding to each of the k clusters, wherein the values ​​of the value dimension and the risk dimension corresponding to each cluster are obtained according to the centroid of each cluster according to a preset mapping relationship; reading the values ​​of the two dimensions of value and risk corresponding to the first cluster from the list to obtain the first category.

[0011] According to an embodiment of the present disclosure, the k categories include: high value-low risk, high value-medium risk, high value-high risk, medium value-low risk, medium value-medium risk, medium value-high risk, low value-low risk, low value-medium risk and low value-high risk.

[0012] According to a second aspect of an embodiment of the present disclosure, a credit card user interaction device based on a digital person is provided. The device includes a receiving module, a first acquisition module, a second acquisition module, and a user interaction module. The receiving module is used to receive a first interaction operation of a first cardholder user. The first acquisition module is used to obtain a first category to which the first cardholder user belongs in response to the first interaction operation, wherein the first category is one of k categories, each category corresponds to a combination of values ​​of the two dimensions of value and risk, and k is an integer greater than 1. The second acquisition module is used to obtain a first digital person corresponding to the first category, wherein the k categories correspond one-to-one to the k digital persons, and the k digital persons are different from each other. The user interaction module is used to interact with the first cardholder user through the first digital person.

[0013] According to an embodiment of the present disclosure, the first acquisition module includes a transaction acquisition submodule, a screening submodule, a feature extraction submodule, a clustering submodule and a category determination submodule. The transaction acquisition submodule is used to obtain the transaction data of the first cardholder and obtain the first transaction data when the first cardholder authorizes the use of the transaction data. The screening submodule is used to filter out the value dimension transaction data and the risk dimension transaction data from the first transaction data according to the preset value dimension screening conditions and risk dimension screening conditions. The feature extraction submodule is used to extract the features of the first cardholder from the value dimension transaction data and the risk dimension transaction data to obtain a first feature vector. The clustering submodule is used to cluster the first feature vector using a trained clustering model to obtain a first cluster to which the first cardholder belongs, wherein the clustering result of the clustering model is k clusters. The category determination submodule is used to determine the first category based on the values ​​of the two dimensions of value and risk corresponding to the first cluster.

[0014] In a third aspect of the embodiments of the present disclosure, an electronic device is provided. The electronic device comprises: one or more processors and a memory. The memory is used to store one or more computer programs. The one or more processors execute the one or more computer programs to implement the above.

[0015] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, on which a computer program or instruction is stored, wherein the computer program or instruction implements the above method when executed by a processor.

[0016] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, including a computer program or instructions, wherein the computer program or instructions implement the above method when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0018] Figure 1 Schematically illustrates a credit card user interaction method, apparatus, device, medium and program product based on a digital human in an embodiment of the present disclosure;

[0019] Figure 2 A flowchart of a credit card user interaction method based on a digital human according to an embodiment of the present disclosure is schematically shown;

[0020] Figure 3 The schematic diagram shows the concept of credit card user interaction based on digital human according to an embodiment of the present disclosure;

[0021] Figure 4 A flowchart of obtaining a first digital person in a credit card user interaction method based on a digital person according to an embodiment of the present disclosure is schematically shown;

[0022] Figure 5 Schematically shows a flow chart for obtaining a first interaction attribute for configuring a digital human in a credit card user interaction method based on a digital human according to an embodiment of the present disclosure;

[0023] Figure 6 A schematic diagram of an operation flow chart of determining the category to which a user belongs in a credit card user interaction method based on a digital human according to an embodiment of the present disclosure;

[0024] Figure 7 The processing flow of determining the category to which a user belongs according to the user's transaction data in a credit card user interaction method based on a digital human in an embodiment of the present disclosure is schematically shown;

[0025] Figure 8 A schematic diagram schematically shows encoding conversion of data of value dimension and risk dimension based on RFMDA model in a credit card user interaction method based on digital human according to an embodiment of the present disclosure;

[0026] Fig. 9 A block diagram schematically shows a credit card user interaction device based on a digital human according to an embodiment of the present disclosure;

[0027] Fig.10 A block diagram schematically shows a first acquisition module in a digital human-based credit card user interaction device according to an embodiment of the present disclosure; and

[0028] Fig.11 A block diagram of an electronic device suitable for implementing the digital human-based credit card user interaction method of an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0029] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0030] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0031] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0032] In the technical solution of the present invention, the user information (including but not limited to user personal information, user image information, user device information, transaction data, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0033] In the scenario of using personal information for automated decision-making, the methods, devices, and systems provided by the embodiments of the present invention provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0034] The disclosed embodiments provide a credit card user interaction method, apparatus, device, medium and program product based on digital humans. According to the category of the cardholder, a corresponding digital human can be selected to interact with the cardholder. The category of the cardholder is obtained by classifying the cardholder's transaction data from the dimensions of value and risk. Different categories of cardholders interact with different digital humans. In this way, the diversity and flexibility of digital human interaction can be utilized to provide cardholders of credit card business with targeted interactive experience.

[0035] Figure 1 The digital human-based credit card user interaction method, apparatus, device, medium and program product of an embodiment of the present disclosure are schematically illustrated.

[0036] like Figure 1 As shown, the application scenario 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0037] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as mobile banking, shopping applications, web browser applications, instant messaging tools, etc. (only as examples). Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, etc.

[0038] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the terminal devices 101, 102, and 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0039] It should be noted that the credit card user interaction method based on digital human provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the credit card user interaction device, equipment, medium and program product based on digital human provided in the embodiment of the present disclosure can generally be set in the server 105. The credit card user interaction method based on digital human provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105. Accordingly, the credit card user interaction device, equipment, medium and program product device based on digital human provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105.

[0040] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0041] The following combination Figure 1 Application scenarios, through Figure 2~Figure 8 An exemplary description of the credit card user interaction method based on digital human in the embodiment of the present disclosure is given

[0042] Figure 2 The flowchart of the credit card user interaction method based on digital human according to an embodiment of the present disclosure is schematically shown.

[0043] like Figure 2 As shown, the digital human-based credit card user interaction method may include operations S201 to S204.

[0044] In operation S201, a first interactive operation of a first cardholder is received. The first interactive operation is an operation to initiate an interaction, such as an interactive operation of a user logging into a mobile banking client, or changing the mobile banking client from a background application to a foreground application.

[0045] In operation S202, in response to the first interactive operation, a first category to which the first cardholder belongs is obtained, wherein the first category is one of k categories, each category corresponds to a combination of values ​​of two dimensions of value and risk, and k is an integer greater than 1.

[0046] The two dimensions of value and risk are extracted from the transaction data of the credit card business. Since the credit card business has the characteristics of high returns and high risks, the credit card business pays great attention to the consumption and repayment capabilities of cardholders. Among them, the value dimension can be measured from the data such as the latest consumption time, consumption frequency and total consumption amount in the transaction data. The risk dimension can be measured from the data such as the total amount of arrears, total amount of cash withdrawal, number of arrears, average duration of arrears, etc. in the transaction data.

[0047] In one embodiment, the value ranges of both the value and risk dimensions may include: high, medium, and low values ​​(which may be represented by corresponding coded values). Thus, the k categories may include: Figure 3 The 9 categories shown are: high value-low risk, high value-medium risk, high value-high risk, medium value-low risk, medium value-medium risk, medium value-high risk, low value-low risk, low value-medium risk and low value-high risk.

[0048] In operation S203, a first digital person corresponding to the first category is obtained, wherein k categories correspond to k digital persons one by one, and the k digital persons are different from each other.

[0049] The K digital humans are different from each other. For example, the digital humans may have different images, different intonations or voices when using voice interactions, different interaction contents, different numbers or types of interactive functions, or different words used during interactions.

[0050] In operation S204, the first digital person interacts with the first card-holding user.

[0051] Figure 3 The following schematically shows the concept of credit card user interaction based on digital human in one embodiment of the present disclosure. It can be understood that Figure 3 The nine categories shown are merely an example, and the present disclosure is not limited thereto.

[0052] like Figure 3 As shown, when credit card holders are segmented into 9 categories, namely, high value-low risk, high value-medium risk, high value-high risk, medium value-low risk, medium value-medium risk, medium value-high risk, low value-low risk, low value-medium risk and low value-high risk, the card issuer can provide different categories of services to cardholders of different categories through different digital humans, so as to provide targeted customer service, thereby achieving better customer service results under the constraints of limited marketing resources.

[0053] For example, for "high-value-low-risk" cardholders, the digital human provided can have more interactivity, functional flexibility, and real-time interaction, and can also be equipped with functions such as emotion recognition and eye capture to enhance the user experience. In terms of interactive content design for cardholders, more content related to the user's consumption interests can be provided, and the risk reminders for repayment can be moderated or reduced.

[0054] For example, for "low-value-high-risk" cardholders, the digital human provided may be weaker in terms of functionality and interactivity. At the same time, during the user interaction process, it can convey concerns about the cardholder's financial situation and explain to the user in more detail the risks of defaulting on repayments.

[0055] In this way, after dividing customers into categories, the disclosed embodiment can use different digital humans to interact with different categories of cardholders, thereby improving the pertinence and effectiveness of digital human interaction in credit card business.

[0056] Figure 4 The flowchart of operation S203 of obtaining the first digital human in the credit card user interaction method based on digital human in one embodiment of the present disclosure is schematically shown.

[0057] like Figure 4 As shown, according to this embodiment, operation S203 may include operations S2031 to S2033.

[0058] In operation S2031, a preset digital human image is obtained, wherein the digital human image is a digital image close to a human image designed by computer graphics technology, including a specific face image, and in some embodiments, decorative images such as hairstyle and clothing matching may also be provided.

[0059] In operation S2032, a first interaction attribute corresponding to the first category is obtained; the interaction attributes corresponding to the k categories are different from each other. The interaction attributes include but are not limited to: intonation or tone of voice interaction, interaction content, number or type of interaction functions, and speech.

[0060] In operation S2033, the digital human image is configured using the first interactive attribute to obtain a first digital human.

[0061] By configuring the preset digital human images through interactive attributes, it is possible to provide different interactive experiences for different categories of cardholders while avoiding the trouble and cost of setting up a digital human image for each category.

[0062] Figure 5 The flowchart of operation S2032 for obtaining the first interaction attribute for configuring the digital human in the credit card user interaction method based on the digital human according to one embodiment of the present disclosure is schematically shown.

[0063] like Figure 5 As shown, according to this embodiment, operation S2032 may include operations S501 to S505. In this embodiment, the value ranges of the two dimensions of value and risk are both discrete level information, and the interaction attribute is speech, where speech refers to the words and expressions used in the interaction process (such as the emotional color of words used in the same semantics, etc.).

[0064] In operation S501, the level information of the value dimension in the first category is obtained to obtain a first value level.

[0065] In operation S502, level information of the risk dimension in the first category is obtained to obtain a first risk level.

[0066] In operation S503, a first speech preset for the first value level is obtained, wherein a corresponding speech is preset for each level information in the value dimension.

[0067] by Figure 3 For example, we can collect the interactive corpus of historical interactions with the high-value-low-risk cardholder user group, extract keywords from it, and then filter the topN keywords through certain means (such as clustering, or word frequency-inverse document frequency TF-IDF algorithm). Then, based on these keywords, set the corresponding words for each value level. For example, first set the topN keywords to the words corresponding to high value. Then replace or modify each keyword in the words corresponding to high value with an alternative expression with a slightly negative emotional concentration to obtain the words corresponding to medium value. For example, if the keyword "strongly recommended" is used in the words corresponding to high value, it can be replaced with "recommended" in the words corresponding to medium value, and replaced with "learn more" in the words corresponding to low value.

[0068] In operation S504, a second speech preset for the first risk level is obtained, wherein a corresponding speech is preset for each level of information in the risk dimension.

[0069] For example, you can collect words related to risks in the past interactions with a large number of cardholders. Then, you can treat these words with similar semantics as a set according to their semantic similarity. When setting the words in each risk level, you can select words and expressions whose emotional color (or concentration) is appropriate to the risk level from each set as the corresponding words for the risk level. For example, for the expression "pay attention to a certain risk", the words set in the low-risk category are "it is recommended that you pay attention to a certain risk", the words set in the medium-risk category can be "it is recommended that you pay close attention to a certain risk, what kind of trouble it may bring to you", and the words set in the high-risk category can be "please do not behave in a certain way, otherwise it will seriously affect your credit rating."

[0070] It can be seen that when presetting the corresponding words for each level of information in the value dimension or risk dimension, a large number of keywords or common expressions related to the value dimension or risk dimension in historical user interactions can be collected first. Then, for the same keyword or common expression, the emotional color can be changed to form a new expression, thereby forming a word corresponding to the information of the corresponding level.

[0071] In operation S505, a first interaction attribute is obtained based on the first speech and the second speech.

[0072] The first speech and the second speech are combined to form the first interactive attribute, so that when the first cardholder interacts, the corresponding words and expressions can be selected according to the progress of the conversation. For example, when it comes to consumer recommendations or consultations, words or expressions can be selected from the first speech and combined with the specific communication content for output. If risk warnings are required, words or expressions can be selected from the second speech and combined with the specific communication content for output.

[0073] In this way, differentiated digital human services can be provided to different categories of cardholders based on the characteristics of the credit card business, thereby achieving precise support for the credit card business through digital human interactive services.

[0074] Figure 6 The operational flow chart of determining the category to which a user belongs in the digital human-based credit card user interaction method according to an embodiment of the present disclosure is schematically shown.

[0075] like Figure 6 As shown, according to this embodiment, the first category to which the first cardholder belongs is determined through operations S601 to S605.

[0076] First, in operation S601, after obtaining authorization from the first cardholder to use its transaction data, the transaction data of the first cardholder is obtained to obtain first transaction data. It should be noted that all transaction data in the present disclosure need to obtain authorization from the corresponding user before obtaining.

[0077] Next, in operation S602, value dimension transaction data and risk dimension transaction data are screened out from the first transaction data according to preset value dimension screening conditions and risk dimension screening conditions.

[0078] The value dimension filtering conditions may include, for example, one or more keywords related to consumption behavior, such as "consumption amount" or "payment amount" keywords, which are used to filter value dimension transaction data.

[0079] The risk dimension screening conditions may include, for example, one or more keywords related to repayment and withdrawal behaviors, such as "repayment", "deferred repayment", "withdrawal", "adjustment amount", etc., which are used to screen risk dimension transaction data.

[0080] Then in operation S603, the features of the first cardholder are extracted from the value dimension transaction data and the risk dimension transaction data to obtain a first feature vector.

[0081] In operation S604, the first feature vector is clustered by the trained clustering model to obtain a first cluster to which the first cardholder belongs; wherein the clustering result of the clustering model is k clusters. By clustering and classifying, the value and risk categories of the cardholder can be more accurately and objectively classified.

[0082] In operation S605 , a first category is determined based on the values ​​of the two dimensions of value and risk corresponding to the first cluster.

[0083] The clustering model can be formed by analyzing and clustering the feature vectors of a large number of cardholders using any clustering algorithm (such as the k-means algorithm, the k-nearest neighbor algorithm, etc.), wherein the feature vectors of these cardholders are obtained in the same way as the first feature vector mentioned above.

[0084] After the clustering model training is completed, k clusters are formed, and the correspondence between each cluster in the k clusters and the values ​​of the two dimensions of value and risk can be determined by the correspondence between the characteristic vector of the centroid in each cluster and the values ​​of the two dimensions of value and risk. In this way, the clustering results of the clustering model are associated with the value and risk of the cardholder concerned by the credit card business. Accordingly, in one embodiment, in operation S605, a list of the values ​​of the two dimensions of value and risk corresponding to each cluster in the k clusters can be obtained first, and then the values ​​of the two dimensions of value and risk corresponding to the first cluster can be read from the above list to obtain the first category.

[0085] Reference below Figure 7 and Figure 8 ,right Figure 6 The specific implementation process is further described as follows. Figure 7 The present invention schematically illustrates a processing flow of determining a user's category based on the user's transaction data in a digital human-based credit card user interaction method according to an embodiment of the present disclosure.

[0086] like Figure 7 As shown in the figure, it is assumed that the transaction data of the value dimension is analyzed and processed to extract the latest consumption time, consumption frequency, and total consumption amount; the transaction data of the risk dimension is analyzed and processed to extract the arrears amount and the cash-out amount. Then, the extracted latest consumption time, consumption frequency, total consumption amount, arrears amount, and cash-out amount can be standardized or encoded to obtain the corresponding eigenvalues, such as R, F, M, D, and A. Next, R, F, M, D, and A are combined to obtain the first eigenvector.

[0087] Next, the k-means algorithm can be used to perform clustering on the first eigenvector based on the first eigenvector to determine the cluster to which the first cardholder belongs. The k-means algorithm uses the mean of all data samples in each cluster subset as the representative point of the cluster. The main idea of ​​the algorithm is to divide the data set into different categories through an iterative process so that the criterion function for evaluating clustering performance is optimal, thereby making each generated cluster compact and independent between clusters. When calculating the distance between data samples, one of the Euclidean distance, Manhattan distance, or Minckhorst distance can be selected as the similarity measure of the algorithm according to actual needs.

[0088] Figure 8The schematic diagram of encoding and converting data of value dimension and risk dimension based on RFMDA model in the credit card user interaction method based on digital human in one embodiment of the present disclosure is schematically shown. The RFMDA model here is an extension of the RFM model, wherein the RFM model is used to measure and evaluate customer value, and the indicators used include: the time of the most recent purchase (Recency), the frequency of purchase (Frequency) and the amount of purchase (Monetary). On the basis of the RFM model, the RFMDA model further expands the two risk indicators of total arrears and total cash withdrawal amount.

[0089] In some embodiments, Figure 7 R, F, M, D, and A in the code are the values ​​after the transaction data is encoded and converted. Specifically, the value dimension transaction data can be converted based on the encoding rules corresponding to the value dimension screening conditions to obtain value features (such as R, F, M). Correspondingly, the risk dimension transaction data can be converted based on the encoding rules corresponding to the risk dimension screening conditions to obtain risk features (such as D, A), and then the value features and risk features are combined to obtain the first feature vector (such as {R, F, M, D, A}).

[0090] Among them, encoding and converting the transaction data can avoid diluting or ignoring the smaller amount when there is a significant difference in the amount of data in different dimensions in the first feature vector. This conversion process can be done by Figure 8 This can be achieved using the RFMDA model in .

[0091] Specifically, the cardholder's recent consumption time, consumption frequency, total consumption, total arrears, and total suspected cash withdrawal are first calculated from the cardholder's transaction data, and then these data are used to form the R, F, M, D, and A feature values ​​of each cardholder to obtain five types of features. The values ​​of R, F, and M can be, for example, between 1 and 5, and the values ​​of D and A can be, for example, between 0 and 5. Then, the cardholder's feature vector can be used to call the K-means clustering algorithm to cluster the cardholder into k clusters.

[0092] In one embodiment, the calculation method of the above five types of characteristics of the card holder is shown in Figure 8 .

[0093] The calculation method of the three types of features R, F and M can be: first extract the most recent consumption time, consumption frequency, and total consumption amount from the transaction data, sort them according to each dimension, and then divide the sorted data into 5 parts with equal numbers of credit card customers, and assign numbers from 1 to 5 to the customers in each part. Among them, when sorting by the most recent consumption time, the group of customers farthest from the present time is assigned 1, then 2, 3, 4 in turn, and the group of customers closest to the present time is assigned 5, and the assigned data is recorded as R. For consumption frequency and total consumption amount, the group of customers with the smallest value after sorting is assigned 1, then 2, 3, 4, and the group of customers with the largest frequency and amount values ​​is assigned 5, and the assigned data is recorded as F and M. After this processing, the value ranges of R, F and M are all {1, 2, 3, 4, 5}, so the higher the R, F and M of the cardholder, the higher its value.

[0094] In the calculation of the two types of features, D and A, the D value of cardholders who have not defaulted is 0, and the A value of cardholders who have not made suspected cash-out transactions is 0. The calculation of the D and A feature values ​​of other cardholders adopts a processing method similar to the aforementioned extraction of R, F, and M features, that is, sorting from large to small according to the two dimensions of default amount and suspected cash-out amount, and then they can be divided into 5 groups, with the feature value of the largest amount being 5 and the feature value of the smallest amount being 1. In this way, the overdue risk of D being 5 is the highest, and the cash-out risk of A being 5 is the highest.

[0095] The centroid of each cluster also corresponds to five types of features. By sorting and dividing the centroids of clusters according to the above five types of features, we can get the values ​​of the centroid of each cluster in the value dimension and risk dimension, and get the mapping relationship between each cluster and the values ​​of the value dimension and risk dimension. Based on this feature calculation method, cardholders with high R, F and M values ​​are high-value customers, and those with low values ​​are low-value customers; similarly, cardholders with high D and A values ​​are high-risk customers, and those with low D and A values ​​are low-risk customers. Figure 3 As shown, in one embodiment, cardholders can be divided into 9 categories according to the value and risk dimensions, namely: high value-low risk, high value-medium risk, high value-high risk, medium value-low risk, medium value-medium risk, medium value-high risk, low value-low risk, low value-medium risk, and low value-high risk. After the subdivision, the card issuer can provide different categories of digital people for interaction with credit card holders who do not fit the category, so that better credit card customer service effects can be achieved under limited resource constraints.

[0096] Based on the digital human-based credit card user interaction method of each of the above embodiments, the present disclosure also provides a digital human-based credit card user interaction device. Fig. 9 and Fig.10The credit card user interaction device based on digital human according to the embodiment of the present disclosure is exemplarily described.

[0097] Fig. 9 A block diagram of a credit card user interaction device based on a digital human according to an embodiment of the present disclosure is schematically shown.

[0098] like Fig. 9 As shown, the digital human-based credit card user interaction device 900 may include a receiving module 910, a first acquisition module 920, a second acquisition module 930 and a user interaction module 940. The device 900 may implement reference Figure 2~Figure 8 The method described.

[0099] The receiving module 910 is used to receive a first interactive operation of a first card-holding user. In one embodiment, the receiving module 910 may perform the operation S201 described above.

[0100] The first acquisition module 920 is used to obtain the first category to which the first cardholder belongs in response to the first interactive operation, wherein the first category is one of k categories, each category corresponds to a combination of values ​​of the two dimensions of value and risk, and k is an integer greater than 1. In one embodiment, the first acquisition module 920 can perform the operation S202 described above.

[0101] The second acquisition module 930 is used to acquire the first digital person corresponding to the first category, wherein the k categories correspond to the k digital persons one by one, and the k digital persons are different from each other. In one embodiment, the second acquisition module 930 may perform the operation S203 described above.

[0102] The user interaction module 940 is used to interact with the first card-holding user through the first digital person. In one embodiment, the user interaction module 940 can perform the operation S204 described above.

[0103] Fig.10 The block diagram schematically shows a first acquisition module 920 in a digital human-based credit card user interaction device according to an embodiment of the present disclosure.

[0104] like Fig.10 As shown, according to this embodiment, the first acquisition module 920 may include a transaction acquisition submodule 921 , a screening submodule 922 , a feature extraction submodule 923 , a clustering submodule 924 and a category determination submodule 925 .

[0105] The transaction acquisition submodule 921 is used to acquire the transaction data of the first cardholder and obtain the first transaction data when the first cardholder authorizes the first cardholder to use the transaction data. In one embodiment, the transaction acquisition submodule 921 may perform the operation S601 described above.

[0106] The screening submodule 922 is used to screen out value dimension transaction data and risk dimension transaction data from the first transaction data according to preset value dimension screening conditions and risk dimension screening conditions. In one embodiment, the screening submodule 922 may perform the operation S602 described above.

[0107] The feature extraction submodule 923 is used to extract the features of the first cardholder from the value dimension transaction data and the risk dimension transaction data to obtain a first feature vector. In one embodiment, the feature extraction submodule 923 may perform the operation S603 described above.

[0108] The clustering submodule 924 is used to cluster the first feature vector using the trained clustering model to obtain the first cluster to which the first cardholder belongs, wherein the clustering result of the clustering model is k clusters. In one embodiment, the clustering submodule 924 may perform the operation S604 described above.

[0109] The category determination submodule 925 is used to determine the first category based on the values ​​of the two dimensions of value and risk corresponding to the first cluster. In one embodiment, the category determination submodule 925 may perform the operation S605 described above.

[0110] The device 900 can implement reference Figure 2~Figure 8 The described method can be specifically referred to in the previous article, and will not be described in detail here.

[0111] According to an embodiment of the present disclosure, any multiple modules among the receiving module 910, the first acquisition module 920, the second acquisition module 930, the user interaction module 940, the transaction acquisition submodule 921, the screening submodule 922, the feature extraction submodule 923, the clustering submodule 924 and the category determination submodule 925 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the receiving module 910, the first acquisition module 920, the second acquisition module 930, the user interaction module 940, the transaction acquisition sub-module 921, the screening sub-module 922, the feature extraction sub-module 923, the clustering sub-module 924 and the category determination sub-module 925 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in an appropriate combination of any of them. Alternatively, at least one of the receiving module 910, the first acquisition module 920, the second acquisition module 930, the user interaction module 940, the transaction acquisition sub-module 921, the screening sub-module 922, the feature extraction sub-module 923, the clustering sub-module 924 and the category determination sub-module 925 can be at least partially implemented as a computer program module, which can perform the corresponding function when the computer program module is executed.

[0112] Fig.11 A block diagram of an electronic device suitable for implementing the digital human-based credit card user interaction method of an embodiment of the present disclosure is schematically shown.

[0113] like Fig.11 As shown, the electronic device 1100 according to an embodiment of the present disclosure includes a processor 1101, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1102 or a program loaded from a storage part 1108 to a random access memory (RAM) 1103. The processor 1101 may include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (for example, an application-specific integrated circuit (ASIC)), etc. The processor 1101 may also include an onboard memory for caching purposes. The processor 1101 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0114] In RAM 1103, various programs and data required for the operation of electronic device 1100 are stored. Processor 1101, ROM 1102 and RAM 1103 are connected to each other through bus 1104. Processor 1101 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 1102 and / or RAM 1103. It should be noted that the program can also be stored in one or more memories other than ROM 1102 and RAM 1103. Processor 1101 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0115] According to an embodiment of the present disclosure, the electronic device 1100 may further include an input / output (I / O) interface 1105, which is also connected to the bus 1104. The electronic device 1100 may further include one or more of the following components connected to the I / O interface 1105: an input portion 1106 including a keyboard, a mouse, etc.; an output portion 1107 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 1108 including a hard disk, etc.; and a communication portion 1109 including a network interface card such as a LAN card, a modem, etc. The communication portion 1109 performs communication processing via a network such as the Internet. A drive 1110 is also connected to the I / O interface 1105 as needed. A removable medium 1111, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1110 as needed, so that a computer program read therefrom is installed into the storage portion 1108 as needed.

[0116] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0117] According to an embodiment of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus, or a device. For example, according to an embodiment of the present disclosure, a computer-readable storage medium may include the ROM 1102 and / or RAM 1103 described above and / or one or more memories other than ROM 1102 and RAM 1103.

[0118] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the credit card user interaction method based on digital human provided by the embodiment of the present disclosure.

[0119] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 1101. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0120] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 1109, and / or installed from a removable medium 1111. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0121] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1109, and / or installed from the removable medium 1111. When the computer program is executed by the processor 1101, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0122] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0123] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations and / or combinations are not explicitly described in the present disclosure. In particular, the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways without departing from the spirit and teachings of the present disclosure. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0124] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the attached claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A credit card user interaction method based on digital human, comprising: receiving a first interaction operation of a first card-holding user; In response to the first interactive operation, obtaining a first category to which the first cardholder belongs, wherein the first category is one of k categories, each category corresponds to a combination of values ​​of two dimensions of value and risk, and k is an integer greater than 1; Acquire a first digital person corresponding to the first category, wherein the k categories correspond to the k digital persons one by one, and the k digital persons are different from each other; and Interact with the first card-holding user through the first digital person.

2. The method according to claim 1, wherein: The acquiring of the first digital person corresponding to the first category comprises: Obtain a preset digital human image; Acquire a first interaction attribute corresponding to the first category; the interaction attributes corresponding to the k categories are different from each other; and The digital human image is configured using the first interactive attribute to obtain the first digital human.

3. The method according to claim 2, wherein: The value ranges of the two dimensions of value and risk are both discrete level information, and the obtaining of the first interactive attribute corresponding to the first category includes: Obtaining level information of the value dimension in the first category to obtain a first value level; Obtaining level information of the risk dimension in the first category to obtain a first risk level; Obtaining a first speech preset for the first value level, wherein a corresponding speech is preset for each level of information in the value dimension, wherein the speech refers to the words and expressions used in the interaction process; Obtaining a second speech preset for the first risk level, wherein a corresponding speech is preset for each level of information in the risk dimension; Based on the first speech and the second speech, the first interaction attribute is obtained.

4. The method according to claim 1, wherein: The first category to which the first cardholder belongs is obtained in the following manner: Acquiring the transaction data of the first cardholder to obtain first transaction data after obtaining authorization from the first cardholder to use the transaction data; Filtering out value dimension transaction data and risk dimension transaction data from the first transaction data according to preset value dimension filtering conditions and risk dimension filtering conditions; Extracting features of the first cardholder from the value dimension transaction data and the risk dimension transaction data to obtain a first feature vector; The first feature vector is clustered by using a trained clustering model to obtain a first cluster to which the first cardholder belongs; wherein the clustering result of the clustering model is k clusters; The first category is determined based on the values ​​of the two dimensions of value and risk corresponding to the first cluster.

5. The method according to claim 4, wherein: The extracting the features of the first cardholder from the value dimension transaction data and the risk dimension transaction data to obtain a first feature vector includes: Convert the value dimension transaction data based on the coding rules corresponding to the value dimension screening conditions to obtain value features; Converting the risk dimension transaction data based on a coding rule corresponding to the risk dimension screening condition to obtain a risk feature; and The value feature and the risk feature are combined to obtain a first feature vector.

6. The method according to claim 4, wherein: The determining of the first category based on the values ​​of the two dimensions of value and risk corresponding to the first cluster includes: Obtain a list of values ​​of the two dimensions of value and risk corresponding to each of the k clusters, wherein the values ​​of the value dimension and the risk dimension corresponding to each cluster are obtained according to the centroid of each cluster according to a preset mapping relationship; The values ​​of the two dimensions of value and risk corresponding to the first cluster are read from the list to obtain the first category.

7. A credit card user interaction device based on digital human, comprising: A receiving module, used for receiving a first interaction operation of a first card-holding user; A first acquisition module, configured to acquire, in response to the first interactive operation, a first category to which the first cardholder belongs, wherein the first category is one of k categories, each category corresponds to a combination of values ​​of two dimensions, namely, value and risk, and k is an integer greater than 1; a second acquisition module, configured to acquire a first digital person corresponding to the first category, wherein the k categories correspond to k digital persons one by one, and the k digital persons are different from each other; and A user interaction module is used to interact with the first card-holding user through the first digital person.

8. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, The one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program or instruction stored thereon, wherein: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program or instructions, wherein: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images