Information determination method and device, equipment and storage medium

Abstract

The invention provides an information determination method and device, equipment and a storage medium, and can be applied to the technical field of artificial intelligence and finance. The method comprises: according to service information required by a user to handle a service, obtaining historical service information of N service products associated with the service information, the service information comprising initial asset information, and N being a positive integer; performing feature extraction on the business information and the historical business information to obtain business features and historical business features; determining a target business product matched with the business information from the N business products based on the business characteristics and historical business characteristics; and determining target asset information for the user based on the target business product, the attribute information of the user and the initial asset information.

Description

Technical Field

[0001] The present disclosure relates to the fields of artificial intelligence and financial technology, and in particular to information determination methods, devices, equipment, media and program products. Background Art

[0002] With the development of computer technology, more and more computer technologies are being applied in the financial field.

[0003] In the process of implementing the present disclosure, it is found that in the related art, in the process of determining information related to the financial field, due to the mismatch between business products and user needs, the determined business products cannot meet the needs of users, affecting the user experience. In addition, due to the increase in the types of business products, it is impossible to determine information for business products that meet user needs, resulting in low information processing efficiency and low accuracy. Summary of the invention

[0004] In view of the above problems, the present disclosure provides an information determination method, apparatus, device, storage medium and program product.

[0005] According to the first aspect of the present disclosure, there is provided an information determination method, comprising: obtaining historical business information of N business products associated with the business information according to business information required by a user for handling business, wherein the business information includes initial asset information and N is a positive integer; performing feature extraction on the business information and the historical business information respectively to obtain business features and historical business features; determining a target business product that matches the business information from the N business products based on the business features and the historical business features; and determining target asset information for the user based on the target business product, attribute information of the user, and the initial asset information.

[0006] According to an embodiment of the present disclosure, target asset information for a user is determined based on a target business product, attribute information of the user, and initial asset information, including: based on the target business product, obtaining product transaction information from multiple historical users associated with the target business product; based on the product transaction information, determining a risk assessment value of the target business product, wherein the risk assessment value is used to characterize the degree of risk of the target business product; and when it is determined that the risk assessment value meets a first threshold, determining the target asset information based on the attribute information and initial asset information of the user.

[0007] According to an embodiment of the present disclosure, when it is determined that the risk assessment value meets the first threshold, the target asset information is determined based on the user's attribute information and the initial asset information, including: extracting features from the user's attribute information to obtain user attribute features; inputting the user attribute features into an asset valuation model to output asset change information, wherein the asset valuation model is pre-trained based on sample attribute information of sample users and labels matching the sample asset change information; and determining the target asset information based on the asset change information and the initial asset information.

[0008] According to an embodiment of the present disclosure, product transaction information includes normal transaction information and abnormal transaction information; based on the product transaction information, a risk assessment value of a target business product is determined, including: parsing the abnormal transaction information to obtain the number of abnormal transactions; parsing the normal transaction information to obtain the number of normal transactions; and determining the risk assessment value of the target business product based on the number of abnormal transactions and the number of normal transactions.

[0009] According to an embodiment of the present disclosure, the information determination method also includes: extracting features from sample attribute information of sample users to obtain sample user attribute features, wherein the sample users are historical users associated with target business products; inputting the sample user attribute features into an initial asset valuation model and outputting sample asset change information; training the initial asset valuation model based on the sample asset change information and labels matching the sample asset change information to obtain an asset valuation model.

[0010] According to an embodiment of the present disclosure, the historical business features include multiple ones; based on the business features and the historical business features, a target business product matching the business information is determined from N business products, including: classifying the multiple historical business features to obtain the target historical business features corresponding to each business product; based on the target historical business features and the business features, a target business product matching the business information is determined.

[0011] According to an embodiment of the present disclosure, a target business product that matches the business information is determined based on the target historical business characteristics and the business characteristics, including: determining the feature similarity based on the business characteristics and the target historical business characteristics corresponding to each type of business product; and determining the target business product that matches the business information based on the business products corresponding to the target historical business characteristics when it is determined that the feature similarity is not less than a second threshold.

[0012] The second aspect of the present disclosure provides an information determination device, including: an acquisition module, used to acquire historical business information of N business products associated with the business information based on business information required by the user to handle the business, wherein the business information includes initial asset information and N is a positive integer; an extraction module, used to extract features from the business information and the historical business information respectively to obtain business features and historical business features; a first determination module, used to determine a target business product that matches the business information from N business products based on the business features and the historical business features; and a second determination module, used to determine target asset information for the user based on the target business product, the user's attribute information, and the initial asset information.

[0013] A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute the above-mentioned information determination method.

[0014] The fourth aspect of the present disclosure also provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the above-mentioned information determination method.

[0015] The fifth aspect of the present disclosure also provides a computer program product, including a computer program, which implements the above-mentioned information determination method when executed by a processor.

[0016] According to the embodiments of the present disclosure, based on the business characteristics of the business information required by the user to handle the business and the historical business characteristics of the historical business information, a target business product matching the business information is determined from multiple business products, and a target business product that meets the user's needs can be intelligently recommended to enhance the user experience. Furthermore, for the target business product, by combining the user's attribute information and initial asset information, the target asset information for the user can be accurately determined, thereby improving the efficiency of information processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above contents and other purposes, 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 The application scenario diagram of the information determination method, apparatus, device, storage medium and program product according to the embodiments of the present disclosure is schematically shown;

[0019] Figure 2 A flowchart of an information determination method according to an embodiment of the present disclosure is schematically shown;

[0020] Figure 3A flowchart of determining target asset information for a user according to an embodiment of the present disclosure is schematically shown;

[0021] Figure 4 A flowchart of an information determination method according to another embodiment of the present disclosure is schematically shown;

[0022] Figure 5 A structural block diagram of an information determination device according to an embodiment of the present disclosure is schematically shown; and

[0023] Figure 6 A block diagram of an electronic device suitable for implementing the information determination method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0024] 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.

[0025] 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 features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0026] 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.

[0027] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0028] In the technical solution of the present disclosure, the user information (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and information (including but not limited to information used for analysis, stored information, displayed information, etc.) involved are all information and information authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant information comply with the relevant laws, regulations and standards of relevant countries and regions, 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.

[0029] In the technical solution of the embodiment of the present disclosure, the user's authorization or consent is obtained before obtaining or collecting the user's personal information.

[0030] In the process of implementing the present disclosure, it is found that in the related art, in the process of determining information related to the financial field, due to the mismatch between business products and user needs, the determined business products cannot meet the needs of users, affecting the user experience. In addition, due to the increase in the types of business products, it is impossible to determine information for business products that meet user needs, resulting in low information processing efficiency and low accuracy.

[0031] The embodiments of the present disclosure provide an information determination method, apparatus, device, medium and program product. The method includes: according to the business information required by the user to handle the business, obtaining the historical business information of N business products associated with the business information, wherein the business information includes the initial asset information, and N is a positive integer; extracting features from the business information and the historical business information respectively to obtain business features and historical business features; based on the business features and the historical business features, determining the target business product that matches the business information from the N business products; and determining the target asset information for the user based on the target business product, the attribute information of the user, and the initial asset information.

[0032] Figure 1 The application scenario diagram of the information determination method, apparatus, device, storage medium and program product according to the embodiments of the present disclosure is schematically shown.

[0033] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0034] The user may use at least one of the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0035] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0036] 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 first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process received information such as user requests, and feed back processing results (such as web pages, information, or information obtained or generated according to user requests) to the terminal device.

[0037] It should be noted that the information determination method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the information determination device provided in the embodiment of the present disclosure can generally be set in the server 105. The information determination method 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 first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the information determination device 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 first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0038] 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.

[0039] The following will be based on Figure 1 The scene described by Figure 2 to Figure 5 The information determination method of the disclosed embodiment is described in detail.

[0040] Figure 2 The flowchart of the information determination method according to the embodiment of the present disclosure is schematically shown.

[0041] like Figure 2 As shown, the information determination method 200 of this embodiment includes operations S210 to S240.

[0042] In operation S210, historical business information of N business products associated with the business information is obtained according to the business information required by the user to handle the business.

[0043] According to an embodiment of the present disclosure, the business information may include initial asset information, where N is a positive integer.

[0044] According to an embodiment of the present disclosure, the historical business information may include historical transaction information of N business products associated with the initial asset information.

[0045] For example, historical transaction information may include at least one of the following: the purpose of the business product, the period of holding the business product, asset information associated with the business product, the type of business product, the user group to which the business product belongs, transaction qualifications of the business product, etc.

[0046] In operation S220, feature extraction is performed on the service information and the historical service information to obtain service features and historical service features.

[0047] According to the embodiments of the present disclosure, a feature extraction model can be used to extract features from business information to obtain business features. A feature extraction model can be used to extract features from historical business information to obtain historical business features. The feature extraction model can select any model that can achieve feature extraction, and is not specifically limited in the embodiments of the present disclosure.

[0048] In operation S230, based on the service characteristics and the historical service characteristics, a target service product matching the service information is determined from the N service products.

[0049] According to the embodiment of the present disclosure, by determining the feature similarity between the business feature and the historical business feature of each business product, when it is determined that the feature similarity meets a threshold, the corresponding business product is determined as a target business product matching the business information.

[0050] According to the embodiment of the present disclosure, the business features and the historical business features of N business products can also be input into a pre-trained gradient boosting decision tree model to output a target business product that matches the business information. The gradient boosting decision tree model can be pre-trained based on sample feature variables and label variables of sample business products.

[0051] In operation S240, target asset information for the user is determined based on the target business product, attribute information of the user, and initial asset information.

[0052] According to the embodiments of the present disclosure, when a user handles a business, the user's attribute information can be obtained with the user's authorization or consent. For example, the user's attribute information includes at least one of the following: user age information, user occupation information, user annual income information, user level information in financial institutions, user credit level information in financial institutions, user debt information, user historical repayment information, user mortgage product asset information, business product information, business asset information, etc.

[0053] According to the embodiment of the present disclosure, when it is determined that the initial asset information is to be changed, the change ratio information can be determined according to the attribute information of the user, and the target asset information can be determined according to the change ratio information and the initial asset information.

[0054] For example, the weight values ​​of various attribute information can be determined according to the importance of various attribute information in the user's attribute information in the target business product. According to the weight values ​​of various attribute information and the coefficient values ​​of various attribute information, it is determined whether to change the initial asset information. In the case of determining to change the initial asset information, the product of the weight values ​​of various attribute information and the coefficient values ​​of various attribute information can be accumulated as the change ratio information. According to the product of the change ratio information and the initial asset information, the target asset information is determined.

[0055] According to the embodiments of the present disclosure, based on the business characteristics of the business information required by the user to handle the business and the historical business characteristics of the historical business information, a target business product matching the business information is determined from multiple business products, and a target business product that meets the user's needs can be intelligently recommended to enhance the user experience. Furthermore, for the target business product, by combining the user's attribute information and initial asset information, the target asset information for the user can be accurately determined, thereby improving the efficiency of information processing.

[0056] According to an embodiment of the present disclosure, the historical business features include multiple ones.

[0057] For example Figure 2 The operation S230 shown, based on the business characteristics and the historical business characteristics, determining the target business product matching the business information from the N business products, may include: classifying the multiple historical business characteristics to obtain the target historical business characteristics corresponding to each business product. Based on the target historical business characteristics and the business characteristics, determining the target business product matching the business information.

[0058] According to the embodiments of the present disclosure, multiple historical business features can be classified based on the tags that match the business products to obtain the target historical business features corresponding to each business product. The business features are matched with the target historical business features corresponding to each business product, and when it is determined that the match is successful, the matched business product is determined as the target business product that matches the business information.

[0059] According to the embodiments of the present disclosure, by classifying multiple historical business features, all target historical business features corresponding to the same business product can be integrated together, and then the business features can be matched with each business product, so that the target business product determined can meet the user's business needs and enhance the user experience. In addition, by classifying and integrating features, the computing resources for determining the target business product are reduced.

[0060] According to an embodiment of the present disclosure, determining a target business product that matches the business information based on target historical business features and business features may include: determining feature similarity based on the business features and the target historical business features corresponding to each type of business product; and determining a target business product that matches the business information based on the business products corresponding to the target historical business features when it is determined that the feature similarity is not less than a second threshold.

[0061] According to an embodiment of the present disclosure, the second threshold may be determined according to the degree of association between the desired user and the target business product in actual situations.

[0062] According to the embodiments of the present disclosure, the feature similarity between the business feature and the target historical business feature corresponding to each type of business product can be calculated according to a common feature similarity algorithm. By comparing the feature similarity with the second threshold, if it is determined that the feature similarity is not less than the second threshold, the corresponding business product is determined as the target business product matching the business information.

[0063] According to the embodiments of the present disclosure, by determining the feature similarity, all business products that meet the second threshold can be determined, thereby enhancing the user experience.

[0064] According to another embodiment of the present disclosure, the target business product may include multiple products. When it is determined that the target business product includes multiple products, when the target business product is sent to the user, it can be displayed from large to small according to feature similarity.

[0065] For example Figure 2 Operation S240 shown in the figure, based on the target business product, the attribute information of the user, and the initial asset information, determines the target asset information for the user, which can be specifically determined by the following steps: Figure 3 Shown.

[0066] Figure 3 A flowchart for determining target asset information for a user according to an embodiment of the present disclosure is schematically shown.

[0067] like Figure 3 As shown, the method 340 for determining target asset information for a user in this embodiment includes operations S341 to S343.

[0068] In operation S341 , based on the target business product, product transaction information from multiple historical users associated with the target business product is acquired.

[0069] According to the embodiment of the present disclosure, when the target business product is determined, product transaction information of the target business product from multiple historical users can be obtained from the database. The product transaction information may include product transaction information of each of the multiple historical users, for example, it may include but is not limited to asset information associated with the transaction of the target business product during the period in which the historical user holds the target business product.

[0070] In operation S342, a risk assessment value of the target business product is determined based on the product transaction information.

[0071] According to the embodiment of the present disclosure, the risk assessment value is used to characterize the degree of risk of the target business product.

[0072] According to the embodiment of the present disclosure, product transaction information can be input into a risk assessment algorithm to output a risk assessment value of a target business product.

[0073] For example, normal evaluation values ​​and abnormal evaluation values ​​can be set in advance for normal transaction information and abnormal transaction information, respectively. According to the number of times normal transaction information appears and the number of times abnormal transaction information appears in the product transaction information of each historical user, the evaluation value of the target business product for each historical user is counted. The total evaluation value of all historical users can be counted, and then the average evaluation value is calculated to obtain the risk evaluation value.

[0074] In operation S343, in a case where it is determined that the risk assessment value satisfies the first threshold, target asset information is determined based on the attribute information of the user and the initial asset information.

[0075] According to an embodiment of the present disclosure, the first threshold may be determined according to the degree of risk control in actual situations.

[0076] According to an embodiment of the present disclosure, when it is determined that the risk assessment value is less than the first threshold, it is determined that the risk assessment value meets the first threshold. When it is determined that the risk assessment value is not less than the first threshold, it is determined that the risk assessment value does not meet the first threshold, and the initial asset information is determined as the target asset information.

[0077] According to an embodiment of the present disclosure, the weight values ​​of various attribute information can be determined according to the importance of various attribute information in the user's attribute information in the target business product. According to the weight values ​​of various attribute information and the coefficient values ​​of various attribute information, it is determined whether to change the initial asset information. In the case of determining to change the initial asset information, the product of the weight values ​​of various attribute information and the coefficient values ​​of various attribute information can be accumulated as the change ratio information. According to the product of the change ratio information and the initial asset information, the target asset information is determined.

[0078] According to the embodiments of the present disclosure, the risk assessment value of the target business product is determined based on the product transaction information from multiple historical users associated with the target business product. Then, when the risk assessment value meets the first threshold, the target asset information is determined, which can reasonably allocate assets to users and enhance user experience under the premise of reducing the transaction abnormality rate, which is conducive to the development of the business scale of financial institutions, provides convenience for users, and enhances information processing efficiency.

[0079] According to an embodiment of the present disclosure, product transaction information includes normal transaction information and abnormal transaction information.

[0080] According to the embodiments of the present disclosure, Figure 3 The operation S342 shown, determining the risk assessment value of the target business product based on the product transaction information, may include: parsing the abnormal transaction information to obtain the number of abnormal transactions. Parsing the normal transaction information to obtain the number of normal transactions. Determining the risk assessment value of the target business product based on the number of abnormal transactions and the number of normal transactions.

[0081] According to the embodiment of the present disclosure, the abnormal evaluation value under the abnormal transaction number and the normal evaluation value under the normal transaction number can be counted respectively according to the risk evaluation rule, and the risk evaluation value is determined according to the difference between the abnormal evaluation value and the normal evaluation value.

[0082] According to an embodiment of the present disclosure, the risk assessment value of the target business product can also be determined by comparing the number of abnormal transactions with the number of normal transactions.

[0083] For example, when the number of abnormal transactions is much smaller than the number of normal transactions, the risk assessment value is determined to be 0, otherwise the risk assessment value is determined to be 1. When the risk assessment value is 0, it is determined that the risk assessment value meets the first threshold. When the risk assessment value is 1, it is determined that the risk assessment value does not meet the first threshold.

[0084] According to an embodiment of the present disclosure, a risk assessment value is determined based on the number of normal transactions and abnormal transactions, which is conducive to reducing the abnormal transaction rate.

[0085] According to the embodiments of the present disclosure, Figure 3The operation S343 shown, when it is determined that the risk assessment value meets the first threshold, determines the target asset information based on the user's attribute information and the initial asset information, which may include: extracting features from the user's attribute information to obtain user attribute features. Inputting the user attribute features into the asset assessment model, outputting asset change information. Determining the target asset information based on the asset change information and the initial asset information.

[0086] According to an embodiment of the present disclosure, the asset evaluation model is pre-trained based on sample attribute information of sample users and labels matching sample asset change information.

[0087] According to an embodiment of the present disclosure, the asset change information can be used to characterize the proportion of the changed assets. The target asset information can be determined based on the initial asset information and combined with the asset change information.

[0088] According to an embodiment of the present disclosure, when the risk assessment value meets the first threshold, the target asset information is determined through the asset assessment model, which reduces the transaction abnormality rate, enhances the user experience, and enhances the information processing efficiency.

[0089] According to another embodiment of the present disclosure, in addition to the above-mentioned operations S210 to S240, the information determination method may also include extracting features from sample attribute information of sample users to obtain sample user attribute features. The sample user attribute features are input into the initial asset valuation model, and sample asset change information is output. The sample user is a historical user associated with the target business product. Based on the sample asset change information and the label matching the sample asset change information, the initial asset valuation model is trained to obtain an asset valuation model.

[0090] According to an embodiment of the present disclosure, the initial asset evaluation model may include a gradient boosting decision tree (lightGBM) model. A distributed gradient boosting tree (Gradient Boosting Decision Tree, GBDT) may be used. The basic principles of the LightGBM model mainly include four algorithms: XGBoost, an improved algorithm of the gradient boosting tree, Histogram algorithm, Gradient-based One-Side Sampling (GOSS), and Exclusive Feature Bundling (EFB).

[0091] XGBoost is a decision tree growth strategy based on leaf-wise growth. Traditional level-wise growth means that the leaf nodes of the same level are split together each time, but in fact, the split gain of some leaf nodes is low. The advantage of the leaf-wise algorithm is that each time in the current leaf node, the leaf node with the largest split gain is found for splitting, instead of splitting all nodes, which can improve accuracy. The disadvantage of the leaf-wise algorithm is that it may grow a relatively deep decision tree and cause overfitting. Therefore, the XGBoost algorithm will add a maximum depth limit on top of the leaf-wise algorithm to prevent overfitting while ensuring high efficiency.

[0092] Histogram can first perform boxing on the eigenvalues, discretize the continuous floating-point eigenvalues ​​into multiple integers to form boxes, and then construct a histogram with a width equal to the number of integers. By traversing the data, the statistics are accumulated in the histogram based on the discretized values ​​as indexes, and then the optimal split point is traversed based on the discrete values ​​of the histogram.

[0093] GOSS reduces the complexity of calculating the objective function when calculating the gain by sampling samples. Sample points with larger gradients play a more important role in calculating information gain. Therefore, when calculating information gain during training, the lightGBM model retains samples with large gradients and randomly collects some samples with small gradients (to ensure the original data distribution). This provides a more accurate evaluation than random sampling with a uniform sampling ratio. For example, you can sort the samples by gradient, select a% of the samples with the largest gradients, and then randomly select b% of the samples from the remaining small gradients. When calculating the information gain, the information gain of the selected b% small gradient samples is increased by a multiple of 1-a / b.

[0094] EFB can bundle mutually exclusive features together to reduce feature dimensions, which can effectively reduce the number of features used to construct histograms, thereby reducing computational complexity. In addition, a bias can be added to ensure the feasibility of feature bundling. For example, to bind two features A and B, the original value of feature A is in the interval [0,40), and the original value of feature B is in the interval [0,60). A bias constant 40 can be added to the value of feature B to change its value range to [40,100). The feature value range after A and B features are bound is [0,100), thereby realizing the fusion of features A and B.

[0095] The lightGBM model can be used to process the sample user attribute features, and the mutually exclusive feature bundling algorithm and the histogram algorithm are used. The mutually exclusive feature bundling algorithm bundles the high-dimensional and sparse features of the data, which can reduce the complexity of the model, reduce redundant information, and improve the efficiency of the model. The histogram algorithm is used to construct a discretized feature histogram after the mutually exclusive feature bundling algorithm, which can perform parallel calculations and make full use of the advantages of multi-core processors. Then, the improved algorithm of the gradient boosting tree is used to generate a decision tree and control the splitting of leaf nodes and the depth of nodes to reduce errors and improve accuracy. After that, gradient-based unilateral sampling can be used for sampling and weighted training to improve the effect of model training.

[0096] Exemplarily, the main parameters of the lightGBM model may include at least one of the following: parameters specifying the gradient boosting tree, parameters of the objective function or loss function, parameters controlling the depth of the training tree, parameters controlling the number of leaf nodes on each tree, learning rate, parameters controlling the proportion of features used in each iteration, parameters controlling the proportion of data samples used in each iteration, etc.

[0097] For example, the depth parameter of the training tree can generally be set to about 5 to prevent overfitting. The depth parameter of the training tree will affect the parameters of the number of leaf nodes on each tree, the model performance and the training time, and play a role in the performance and generalization ability of the model. The parameter for controlling the feature ratio used in each iteration can be set to 0.9, and 90% of the features are randomly selected in each iteration, which can reduce overfitting and increase the generalization ability of the model. The parameter for controlling the data sample ratio used in each iteration can be set to 0.8, and 80% of the training data are randomly selected in each iteration, which can reduce overfitting and increase the generalization ability of the model. Preferably, the parameters for the number of leaf nodes on each tree and the depth parameter of the training tree can be set to smaller values, such as 10 leaf nodes on each tree, and the depth of the training tree is controlled to be 3 to avoid overfitting. The parameter for controlling the data sample ratio used in each iteration can also select a smaller value, such as 0.5. The parameter for controlling the feature ratio used in each iteration can be set to a smaller value, such as 0.5, which can speed up training.

[0098] According to the embodiments of the present disclosure, since the selection of lightGBM model parameters depends on the specific data set, a cross-validation method can be used to determine the most appropriate parameter combination. For example, through cross-validation, the optimal learning rate, tree depth, etc. are determined, and each round of training predicts and reduces the residual of the previous round.

[0099] According to another embodiment of the present disclosure, a test set can also be used for model evaluation, and evaluation functions such as root mean square error, mean absolute error, etc. can be used. If the model performs poorly, methods such as network search and Bayesian optimization can be used to automatically search for the optimal value in the parameter control to improve the accuracy and performance of the model. The test set can be a data set split according to the attribute characteristics of the sample user, and the training set and the test set are split in a ratio of 7:3.

[0100] According to an embodiment of the present disclosure, based on sample asset change information and labels matching the sample asset change information, the asset evaluation model trained has high accuracy and can accurately determine the sample asset change information, which is conducive to accurately determining the target asset information for the user.

[0101] Figure 4 The flowchart of the information determination method according to another embodiment of the present disclosure is schematically shown.

[0102] like Figure 4 As shown, the information determination method 400 of this embodiment includes operations S410 to S460.

[0103] In operation S410, the user may input business information required for handling business on the business application.

[0104] In operation S420, when it is determined that the business application detects the business information, the business information is sent to the server.

[0105] In operation S430, when it is determined that the server receives the business information, historical business information of N business products associated with the business information is acquired from a database.

[0106] In operation S440, based on the business information and the historical business information, a target business product matching the business information is determined from the N business products.

[0107] In operation S450, based on the target business product, matching is performed according to the attribute information of the user and the product transaction information from multiple historical users associated with the target business product, and asset change information is obtained based on the asset evaluation model.

[0108] In operation S460, the target business product and asset change information are sent to a display interface of the business application.

[0109] According to an embodiment of the present disclosure, the business application program may be a program installed on a mobile phone.

[0110] According to an embodiment of the present disclosure, the business information may include at least one of the following: deadline information for handling the business, initial asset information, purpose information for handling the business, and fixed asset information associated with the user.

[0111] According to the embodiments of the present disclosure, feature extraction can be performed on business information and historical business information respectively to obtain business features and historical business features; based on the business features and historical business features, a target business product matching the business information is determined from N business products.

[0112] According to the embodiments of the present disclosure, business features and historical business features can be input into a product prediction model to output a target business product. The product prediction model can be obtained by pre-training a lightGBM model using sample historical business features as feature variables and sample business products matching the sample historical business features as label variables.

[0113] According to the embodiments of the present disclosure, the user's attribute information can be feature extracted to obtain the user's attribute features. The user's attribute features are input into the asset evaluation model to output asset change information. The asset change information can include the proportion information of the changed assets.

[0114] For example, the target business product is product A, and the asset change information is information that increases by X%. Based on the initial asset information, the asset information can be increased by X% to obtain the target asset information. When determining X%, the asset evaluation model can be shown as follows (1):

[0115] X=a*λ1+b*λ2+c*λ3+d*λ4(1)

[0116] Among them, a can represent the ratio of the original holding period of the business product by the user to the actual holding period of the business product. The original holding period of the business product by the user can refer to the longest holding period agreed upon by the user and the business product holder. The actual holding period of the business product can refer to the actual holding period of the business product by the user. b can represent the credit rating coefficient of the user in the business product holder. The credit rating coefficient can be determined according to the existing grade classification rules. c can represent the user age range coefficient. For example, the age range can be divided into under 25 years old, 25 to 60 years old, and over 60 years old according to the user's historical repayment information and the user's historical repayment period, and then the user age range coefficient is determined. d can be used to represent the asset growth rate of the mortgage product. For example, it can be determined according to the ratio of the actual assets of the current mortgage product to the historical assets of the historical mortgage product. λ1, λ2, λ3, and λ4 are respectively represented as the weight values ​​of a, b, c, and d.

[0117] According to the embodiments of the present disclosure, it is possible to intelligently recommend target business products that meet user needs and enhance user experience. Furthermore, for target business products, by combining user attribute information and initial asset information, it is possible to accurately determine target asset information for users, improve information processing efficiency, and facilitate financial institutions to efficiently and reasonably allocate assets for different users.

[0118] Based on the above information determination method, the present disclosure also provides an information determination device. Figure 5 The device is described in detail.

[0119] Figure 5 The structural block diagram of the information determination device according to an embodiment of the present disclosure is schematically shown.

[0120] like Figure 5 As shown, the information determination device 500 of this embodiment includes an acquisition module 510 , an extraction module 520 , a first determination module 530 and a second determination module 540 .

[0121] The acquisition module 510 is used to acquire historical business information of N business products associated with the business information according to the business information required by the user to handle the business, wherein the business information includes initial asset information, and N is a positive integer. In one embodiment, the acquisition module 510 can be used to perform the operation S210 described above, which will not be repeated here.

[0122] The extraction module 520 is used to extract features from the service information and the historical service information respectively to obtain service features and historical service features. In one embodiment, the extraction module 520 can be used to perform the operation S220 described above, which will not be described in detail here.

[0123] The first determination module 530 is used to determine a target service product matching the service information from the N service products based on the service characteristics and historical service characteristics. In one embodiment, the first determination module 530 can be used to perform the operation S230 described above, which will not be described in detail here.

[0124] The second determination module 540 is used to determine the target asset information for the user based on the target business product, the attribute information of the user, and the initial asset information. In one embodiment, the second determination module 540 can be used to perform the operation S240 described above, which will not be repeated here.

[0125] According to an embodiment of the present disclosure, the second determination module 540 may include: an acquisition subunit, a first determination subunit, and a second determination subunit.

[0126] The acquisition subunit is used to acquire product transaction information from multiple historical users associated with the target business product based on the target business product.

[0127] The first determination subunit is used to determine a risk assessment value of a target business product based on product transaction information, wherein the risk assessment value is used to characterize the degree of risk of the target business product.

[0128] The second determining subunit is used to determine the target asset information based on the user's attribute information and the initial asset information when it is determined that the risk assessment value meets the first threshold.

[0129] According to an embodiment of the present disclosure, when it is determined that the risk assessment value meets the first threshold, determining the target asset information based on the user's attribute information and the initial asset information may include: extracting features from the user's attribute information to obtain user attribute features. Inputting the user attribute features into an asset assessment model, and outputting asset change information. The asset assessment model is pre-trained based on sample attribute information of sample users and labels matching the sample asset change information. Determining the target asset information based on the asset change information and the initial asset information.

[0130] According to an embodiment of the present disclosure, product transaction information may include normal transaction information and abnormal transaction information. Based on the product transaction information, determining the risk assessment value of the target business product may include: parsing the abnormal transaction information to obtain the number of abnormal transactions. Parsing the normal transaction information to obtain the number of normal transactions. Determining the risk assessment value of the target business product based on the number of abnormal transactions and the number of normal transactions.

[0131] According to an embodiment of the present disclosure, the information determination device 500 may further include: a sample extraction module, a sample processing module and a training module.

[0132] The sample extraction module is used to extract features from sample attribute information of sample users to obtain attribute features of sample users. Sample users are historical users associated with target business products.

[0133] The sample processing module is used to input sample user attribute characteristics into the initial asset evaluation model and output sample asset change information.

[0134] The training module is used to train the initial asset valuation model based on the sample asset change information and the labels matching the sample asset change information to obtain the asset valuation model.

[0135] According to an embodiment of the present disclosure, the historical service feature may include multiple features. The first determination module 530 may include: a classification unit and a matching unit.

[0136] The classification unit is used to classify multiple historical business features to obtain the target historical business features corresponding to each business product.

[0137] The matching unit is used to determine the target business product that matches the business information based on the target historical business characteristics and business characteristics.

[0138] According to an embodiment of the present disclosure, determining a target business product that matches the business information based on the target historical business feature and the business feature may include: determining a feature similarity based on the business feature and the target historical business feature corresponding to each type of business product. When it is determined that the feature similarity is not less than a second threshold, determining a target business product that matches the business information based on the business product corresponding to the target historical business feature.

[0139] According to an embodiment of the present disclosure, any multiple modules of the acquisition module 510, the extraction module 520, the first determination module 530, and the second determination module 540 can be combined in 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 acquisition module 510, the extraction module 520, the first determination module 530, and the second determination module 540 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 a suitable combination of any of them. Alternatively, at least one of the acquisition module 510 , the extraction module 520 , the first determination module 530 , and the second determination module 540 may be at least partially implemented as a computer program module, and when the computer program module is executed, a corresponding function may be performed.

[0140] Figure 6 A block diagram of an electronic device suitable for implementing the information determination method according to an embodiment of the present disclosure is schematically shown.

[0141] like Figure 6 As shown, the electronic device 600 according to an embodiment of the present disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 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.

[0142] In RAM 603, various programs and information required for the operation of electronic device 600 are stored. Processor 601, ROM 602 and RAM 603 are connected to each other via bus 604. Processor 601 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in one or more memories.

[0143] According to an embodiment of the present disclosure, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the I / O interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0144] 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.

[0145] According to an embodiment of the present disclosure, the 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 602 and / or RAM 603 described above and / or one or more memories other than ROM 602 and RAM 603.

[0146] 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 method provided by the embodiment of the present disclosure.

[0147] 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 601. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0148] 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 609, and / or installed from a removable medium 611. 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.

[0149] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, 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, apparatus, module, unit, etc. described above can be implemented by a computer program module.

[0150] 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).

[0151] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0152] 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.

[0153] 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 method for determining information, comprising: According to the business information required by the user to handle the business, obtain the historical business information of N business products associated with the business information, wherein the business information includes initial asset information, and N is a positive integer; Extracting features from the service information and the historical service information respectively to obtain service features and historical service features; Based on the service characteristics and the historical service characteristics, determining a target service product that matches the service information from the N service products; and Target asset information for the user is determined based on the target business product, the attribute information of the user, and the initial asset information.

2. The method according to claim 1, wherein: The determining target asset information for the user based on the target business product, the attribute information of the user, and the initial asset information includes: Based on the target business product, obtaining product transaction information from multiple historical users associated with the target business product; Based on the product transaction information, determining a risk assessment value of the target business product, wherein the risk assessment value is used to characterize the degree of risk of the target business product; In a case where it is determined that the risk assessment value satisfies a first threshold, the target asset information is determined based on the attribute information of the user and the initial asset information.

3. The method according to claim 2, wherein: The step of determining the target asset information based on the attribute information of the user and the initial asset information when it is determined that the risk assessment value meets the first threshold comprises: Extracting features from the attribute information of the user to obtain user attribute features; Inputting the user attribute features into an asset evaluation model and outputting asset change information, wherein the asset evaluation model is pre-trained based on sample attribute information of sample users and labels matching the sample asset change information; The target asset information is determined according to the asset change information and the initial asset information.

4. The method according to claim 2, wherein: The product transaction information includes normal transaction information and abnormal transaction information; Determining the risk assessment value of the target business product based on the product transaction information includes: Analyze the abnormal transaction information to obtain the number of abnormal transactions; Parsing the normal transaction information to obtain the normal transaction times; A risk assessment value of the target business product is determined based on the number of abnormal transactions and the number of normal transactions.

5. The method according to claim 3, further comprising: Extracting features from sample attribute information of the sample users to obtain sample user attribute features, wherein the sample users are historical users associated with the target business product; Input the sample user attribute characteristics into the initial asset evaluation model, and output the sample asset change information; Based on the sample asset change information and the label matching the sample asset change information, the initial asset evaluation model is trained to obtain the asset evaluation model.

6. The method according to any one of claims 1 to 5, wherein: The historical business characteristics include multiple; The determining, based on the service characteristics and the historical service characteristics, a target service product matching the service information from the N service products includes: Classifying the plurality of historical business features to obtain a target historical business feature corresponding to each business product; Based on the target historical business characteristics and the business characteristics, the target business product matching the business information is determined.

7. The method according to claim 6, wherein: The determining, based on the target historical service characteristics and the service characteristics, the target service product matching the service information includes: Determining feature similarity based on the business features and target historical business features corresponding to each type of business product; When it is determined that the feature similarity is not less than a second threshold, the target business product that matches the business information is determined according to the business product corresponding to the target historical business feature.

8. An information determination device, comprising: An acquisition module, used to acquire historical business information of N business products associated with the business information according to the business information required by the user to handle the business, wherein the business information includes initial asset information, and N is a positive integer; An extraction module, used to extract features from the service information and the historical service information respectively to obtain service features and historical service features; A first determination module is configured to determine a target service product matching the service information from the N service products based on the service characteristics and the historical service characteristics; and The second determination module is used to determine target asset information for the user based on the target business product, attribute information of the user, and the initial asset information.

9. An electronic device, comprising: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors are enabled to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images