Credit card application information processing method and device, storage medium, electronic equipment and product

Abstract

The invention relates to the field of information processing, and discloses a credit card application information processing method and device, a storage medium, electronic equipment and a product. The method comprises the steps of performing statistical processing on multi-dimensional data based on statistical rules corresponding to a plurality of statistical indexes to obtain statistical data corresponding to the plurality of statistical indexes; performing classification processing on the credit card application objects based on text data in the multi-dimensional data to obtain categories of the credit card application objects; performing sentiment classification based on the credit card application text information, and determining a sentiment type corresponding to the credit card application text information; performing multi-dimensional prediction on the credit card application object based on one or more of the multi-dimensional data, the plurality of statistical data, the category and the emotion type to obtain a multi-dimensional prediction result of the credit card application object; and determining a processing result of the credit card application information based on the multi-dimensional prediction result of the credit card application object. The credit card application information is automatically processed, and the processing accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to a method, device, storage medium, electronic equipment and product for processing credit card application information. Background Art

[0002] The credit card approval process involves a large amount of data analysis. During the credit card approval process, manual review consumes a lot of time and labor costs, and the review efficiency is low.

[0003] In the automated implementation of credit card approval, approval rules can be set to process the credit card approval process to obtain approval results. However, the setting of approval rules is not flexible enough to flexibly process credit card application information. In addition, the data type of credit card application information lacks diversity, which affects the accuracy of the processing of credit card application information. Summary of the invention

[0004] The present invention provides a method, device, storage medium, electronic equipment and product for processing credit card application information, which improves the accuracy of the processing process of credit card application information.

[0005] According to one aspect of the present invention, a method for processing credit card application information is provided, comprising:

[0006] Acquire credit card application information, wherein the credit card application information includes multi-dimensional data of the credit card application object and credit card application text information;

[0007] Performing statistical processing on the multi-dimensional data based on statistical rules corresponding to the multiple statistical indicators respectively, to obtain statistical data corresponding to the multiple statistical indicators;

[0008] Classify the credit card application object based on the text data in the multi-dimensional data to obtain the category of the credit card application object;

[0009] Performing sentiment classification based on the credit card application text information to determine the sentiment type corresponding to the credit card application text information, wherein the sentiment type includes positive sentiment and negative sentiment;

[0010] Perform multi-dimensional prediction on the credit card application object based on the multi-dimensional data, the plurality of statistical data, one or more of the categories and the sentiment types, and obtain a multi-dimensional prediction result of the credit card application object;

[0011] The processing result of the credit card application information is determined based on the multi-dimensional prediction result of the credit card application object.

[0012] Optionally, the statistical processing of the multidimensional data is performed based on statistical rules corresponding to multiple statistical indicators, including: based on pre-set derivative calculation rules, performing derivative calculation on at least one item of the multidimensional data to determine derived indicator data; wherein the derived indicator data includes one or more of the following: trend information of any data, correlation information between at least two data, and mean and variance of any data.

[0013] Optionally, the statistical processing of the multidimensional data based on statistical rules corresponding to multiple statistical indicators includes: clustering the transaction data in the multidimensional data to obtain a target transaction cluster to which the transaction data belongs, and the target transaction cluster represents a transaction type corresponding to the transaction data.

[0014] Optionally, the clustering processing of the transaction data in the multi-dimensional data includes: obtaining historical transaction data of multiple objects, performing clustering processing based on the historical transaction data of the multiple objects to obtain multiple transaction clusters; obtaining the cluster center of each of the transaction clusters, and determining the target transaction cluster to which the transaction data in the multi-dimensional data belongs based on matching data between the transaction data in the multi-dimensional data and the cluster center of each of the transaction clusters.

[0015] Optionally, the statistical processing of the multi-dimensional data based on statistical rules corresponding to multiple statistical indicators includes: performing pattern recognition on financial behavior data in the multi-dimensional data to determine the credit behavior pattern corresponding to the financial behavior data.

[0016] Optionally, the financial behavior data includes repayment behavior data;

[0017] The method of performing pattern recognition on the financial behavior data in the multi-dimensional data to determine the credit behavior pattern corresponding to the financial behavior data includes: obtaining expected repayment behavior data corresponding to each repayment behavior data, determining overdue behavior data, and overdue duration and overdue repayment amount of each overdue behavior data based on the expected repayment behavior data corresponding to each repayment behavior data; and determining the credit behavior pattern based on one or more of the statistical number of overdue behavior data, overdue duration and overdue repayment amount of each overdue behavior data.

[0018] Optionally, the classifying the credit card application object based on the text data in the multi-dimensional data to obtain the category of the credit card application object includes: extracting key information from the text data in the multi-dimensional data based on a natural language processing model; performing entity recognition on the key information to obtain entity information; and classifying the credit card application object based on the entity information to obtain the category of the credit card application object.

[0019] Optionally, the sentiment classification based on the credit card application text information is performed to determine the sentiment type corresponding to the credit card application text information, including: performing word segmentation processing on the credit card application text information to determine multiple word segments to be processed; for each word segment to be processed, determining a context word segmentation corresponding to the word segment to be processed, the context word segmentation including at least one word before the word segment to be processed and at least one word after the word segment to be processed; determining a target vector of the word segment to be processed based on the word segment to be processed and the context word segmentation corresponding to the word segment to be processed; and classifying the target vector of each word segment to be processed based on a pre-trained sentiment classification model to obtain the sentiment type corresponding to the credit card application text information.

[0020] Optionally, the multi-dimensional prediction includes one or more of a credit risk prediction and an anti-fraud prediction, and correspondingly, the multi-dimensional prediction result includes one or more of a credit risk prediction result and an anti-fraud prediction result.

[0021] Optionally, performing a multi-dimensional prediction on the credit card application object based on the multi-dimensional data, the plurality of statistical data, one or more of the categories and the sentiment types to obtain a multi-dimensional prediction result of the credit card application object includes: predicting the multi-dimensional data through a pre-trained anti-fraud model to obtain the anti-fraud prediction result; and / or,

[0022] The multi-dimensional data, the plurality of statistical data, the categories and the sentiment types are predicted through a pre-trained risk prediction model to obtain the credit risk prediction result.

[0023] Optionally, the credit risk prediction result includes a credit risk probability, and the anti-fraud prediction result includes a fraud probability;

[0024] The determining the processing result of the credit card application information based on the multi-dimensional prediction result of the credit card application object includes: inputting one or more of the credit risk prediction result and the anti-fraud prediction result into a pre-trained decision model to obtain the processing result of the credit card application information; or

[0025] When the credit risk probability in the credit risk prediction result is greater than a first threshold, and / or the fraud probability in the anti-fraud prediction result is greater than a second threshold, the processing result of the credit card application information is determined to be application rejection; when the credit risk probability in the credit risk prediction result is less than the first threshold, and the fraud probability in the anti-fraud prediction result is less than the second threshold, the processing result of the credit card application information is determined to be application approval.

[0026] According to one aspect of the present invention, there is provided a device for processing credit card application information, comprising:

[0027] An information acquisition module, used to acquire credit card application information, wherein the credit card application information includes multi-dimensional data of the credit card application object and credit card application text information;

[0028] A statistical indicator determination module, used to perform statistical processing on the multi-dimensional data based on statistical rules corresponding to the multiple statistical indicators, to obtain statistical data corresponding to the multiple statistical indicators;

[0029] A classification module, used for classifying the credit card application object based on the text data in the multi-dimensional data to obtain the category of the credit card application object;

[0030] A sentiment classification module, used to perform sentiment classification based on the credit card application text information, and determine the sentiment type corresponding to the credit card application text information, wherein the sentiment type includes positive sentiment and negative sentiment;

[0031] A multi-dimensional prediction module, used for performing multi-dimensional prediction on the credit card application object based on the multi-dimensional data, a plurality of the statistical data, one or more of the categories and the sentiment types, to obtain a multi-dimensional prediction result of the credit card application object;

[0032] A processing result determination module is used to determine the processing result of the credit card application information based on the multi-dimensional prediction result of the credit card application object.

[0033] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0034] at least one processor; and

[0035] a memory communicatively connected to the at least one processor; wherein,

[0036] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for processing credit card application information described in any embodiment of the present invention.

[0037] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for processing credit card application information described in any embodiment of the present invention when executed.

[0038] According to another aspect of the present invention, a computer program product is provided, characterized in that the computer program product includes a computer program, and when the computer program is executed by a processor, it implements the method for processing credit card application information described in any embodiment of the present invention.

[0039] The technical solution of the embodiment of the present invention obtains multi-dimensional data of a credit card application object, performs statistical processing on the multi-dimensional data, and obtains multiple statistical data; classifies the credit card application object through the multi-dimensional data to obtain the category of the credit card application object; and performs sentiment classification on the credit card application text information to obtain the sentiment type corresponding to the credit card application text information. The multi-dimensional data, multiple statistical data, one or more of the categories and the sentiment type are used as input information for multi-dimensional prediction to improve the accuracy of multi-dimensional prediction, and the processing result of the credit card application information is determined based on the multi-dimensional prediction result, thereby further improving the accuracy of the processing result of the credit card application information.

[0040] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0042] Figure 1 is a flow chart of a method for processing credit card application information provided by Embodiment 1 of the present invention;

[0043] Figure 2 It is a structural schematic diagram of a device for processing credit card application information provided by Embodiment 3 of the present invention;

[0044] Figure 3 It is a structural schematic diagram of an electronic device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0045] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0046] It should be noted that the terms "first characteristic data", "second characteristic data", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0047] The acquisition, storage and / or processing of data in the technical solutions involved in this application shall comply with the relevant provisions of national laws and regulations. It should be noted that in the embodiments of this application, some existing solutions in the industry such as certain software, components, models, etc. may be mentioned, which should be considered as exemplary, and their purpose is only to illustrate the feasibility of the implementation of the technical solutions of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0048] Embodiment 1

[0049] Figure 1 This is a flowchart of a method for processing credit card application information provided in the first embodiment of the present invention. This embodiment can be applied to process the credit card application information during the credit card application process to obtain a processing result. The processing result can represent the approval recommendation information of the credit card application information. The method can be executed by a processing device for credit card application information. The processing device for credit card application information can be implemented in the form of hardware and / or software. The processing device for credit card application information can be configured in the electronic device provided in the embodiment of the present invention. The electronic device includes but is not limited to a mobile terminal, a server or a computer. Figure 1 As shown, the method includes:

[0050] S110. Obtain credit card application information, where the credit card application information includes multi-dimensional data of a credit card application object and credit card application text information.

[0051] S120, performing statistical processing on the multi-dimensional data based on statistical rules corresponding to the multiple statistical indicators, to obtain statistical data corresponding to the multiple statistical indicators.

[0052] S130. Classify the credit card application object based on the text data in the multi-dimensional data to obtain the category of the credit card application object.

[0053] S140. Perform sentiment classification based on the credit card application text information to determine the sentiment type corresponding to the credit card application text information, where the sentiment type includes positive sentiment and negative sentiment.

[0054] S150, performing multi-dimensional prediction on the credit card application object based on the multi-dimensional data, the plurality of statistical data, one or more of the categories and the sentiment types, to obtain a multi-dimensional prediction result of the credit card application object.

[0055] S160. Determine a processing result of the credit card application information based on the multi-dimensional prediction result of the credit card application object.

[0056] In this embodiment, the credit card application object may be a user who applies for a credit card. For example, the credit card application object may apply for a credit card through a mobile terminal or a self-service terminal or other device. The mobile terminal here may include but is not limited to a mobile phone or a tablet computer. The credit card application page is displayed in the display interface of the above-mentioned mobile terminal or self-service terminal, and the credit card application information is collected through the credit card application page. Corresponding to the application operation of the credit card application object on the credit card application page, the credit card application information is transmitted to the application information processing end. The application information processing end can execute the credit card application information processing method in this embodiment to obtain the processing result of the credit card application information, and display the processing result of the credit card application information on the credit card approval end for the credit card approver to refer to the information. The application information processing end and the credit card approval end here may be the same device or different devices, which is not limited here.

[0057] Credit card application information is the information required to verify the credit card application object in the credit card application scenario, and may include multi-dimensional data of the credit card application object and credit card application text information, wherein the multi-dimensional data of the credit card application object can be understood as data that displays the credit card application object from multiple dimensions. Optionally, the multi-dimensional data of the credit card application object includes but is not limited to the basic information, transaction data and financial behavior data of the credit card application object. Among them, the basic information of the object can represent the information of the user's identity and social identity, and may include but is not limited to identity information, work information, etc. Transaction data is the data generated by the transaction operation of the credit card application object within a preset time period, and financial behavior data is the data generated by the financial behavior of the credit card application object within a preset time period, wherein financial behavior can be understood as interactive behavior with financial institutions, including but not limited to loans and repayments, etc. Credit card application text information may be text information submitted by the credit card application object to apply for a credit card.

[0058] It is understandable that the credit card application page may include a collection control for multi-dimensional data and a collection control for application text information. For example, the collection control may be an input control or a selection control, etc. Exemplarily, the collection control for application text information may be an input control that can accept input text information. Optionally, each of the above collection controls may be configured with description information to display the type of data and content collected by the collection control, so that the credit card applicant can accurately input the corresponding data based on the above description information, thereby improving the accuracy of credit card application information collection. It should be noted that the multi-dimensional information of the credit card application object may be obtained under the premise of authorization by the credit card application object, and is used to verify the information of the credit card application process. After the credit card application is completed, the multi-dimensional information of the credit card application object is deleted to avoid information leakage.

[0059] Optionally, data of at least one dimension of the credit card application object can be read from a preset database based on the identity information of the credit card application object, for example, transaction data and financial behavior data can be read from a corresponding database. A database interface can be configured in the credit card application page, and data can be matched and read from multiple databases based on the identity information of the credit card application object. Exemplarily, the credit card application page includes a financial database interface, and a database in a financial server can be accessed through the financial database interface. The database in the financial server can store historical financial behavior data of multiple objects. Based on the detection of the identity information of the credit card application object in the credit card application page, the financial behavior data corresponding to the credit card application object is read through the financial database interface based on the identity information of the credit card application object.

[0060] After acquiring the multi-dimensional data, the multi-dimensional data is cleaned to remove invalid data and interference data, and reduce the impact of invalid data and interference data in subsequent processing. For example, it can include removing missing values, abnormal values ​​and duplicate values ​​from the multi-dimensional data, unifying the data format of the multi-dimensional data, and standardizing the data.

[0061] It can be understood that the multidimensional data is the collected raw data. In order to improve the comprehensiveness and effectiveness of the data, the multidimensional data is statistically processed to obtain statistical data. The statistical data is calculated based on statistical rules for one or more data in the multidimensional data, and can characterize the statistical characteristics of one or more data in the multidimensional data.

[0062] In some embodiments, statistical processing is performed on the multi-dimensional data based on statistical rules corresponding to the plurality of statistical indicators, including: performing derivative calculation on at least one item of the multi-dimensional data based on a preset derivative calculation rule to determine derivative indicator data;

[0063] The derivative calculation rules can be understood as calculation rules for performing derivative calculations on one or more data in the multi-dimensional data, for example, including but not limited to one or more of data trend calculation rules, data association calculation rules, data operation rules, etc. Correspondingly, the derived indicator data obtained based on the above derivative calculation rules include one or more of the following: trend information of any data, association information between at least two data, and the sum, mean, and variance of any data.

[0064] Multidimensional data may include data that changes over time, such as transaction data, and the above data is configured with a timestamp. Data trend calculation rules can be used to determine the trend information formed over time for the value of the data to be processed (any data in the multidimensional data) in a unit time. The unit time can be at least one, such as a day or a month. The value of the data to be processed in a unit time can be one or more of the sum, mean, and variance in the unit time. The trend information of the data to be processed can be in the form of a data sequence, including multiple values ​​of unit time. For example, the trend information of the data to be processed can be expressed as Represents the data mean of the data to be processed k at the tth unit time, Characterizes the data variance of the to-be-processed data k at the t-th unit time. Here, the data sequence only includes the mean and variance, as an example, in other embodiments, other statistical values ​​can be set according to statistical requirements.

[0065] There may be associations between different statistics to be counted in the multi-dimensional data, and the data association calculation rules may be used to determine the association information between at least two items of data. The association information between different statistics to be counted may include: there is no association between the two items of data, there is a fuzzy association between the two items of data, and there is a strong association between the two items of data. Optionally, a correlation analysis is performed on at least two items of statistics to be counted in the multi-dimensional data using a pre-set correlation analysis algorithm to obtain association information between different items of data. For example, the correlation analysis algorithm includes but is not limited to the APRIORI algorithm, the FP-Growth algorithm, the ECLAT algorithm, the FPMax algorithm, and the gray correlation analysis algorithm. Optionally, a correlation analysis is performed on at least two items of statistics to be counted in the multi-dimensional data using a pre-set correlation analysis model, for example, at least two items of statistics to be counted are input into the above correlation analysis model, and the association information between at least two items of statistics to be counted is output.

[0066] The sum, mean and variance of any item of data may be a statistical group obtained by respectively performing sum calculation, mean calculation and variance calculation on the statistical data to be counted within a preset time period.

[0067] In some embodiments, the multi-dimensional data is statistically processed based on statistical rules corresponding to multiple statistical indicators, including: clustering the transaction data in the multi-dimensional data to obtain a target transaction cluster to which the transaction data belongs, and the target transaction cluster represents the transaction type corresponding to the transaction data.

[0068] Transaction data may be resource interaction data between a credit card applicant and a merchant, for example, it may include resource expenditure data. The resource here may be understood as monetary resources. The transaction data of different credit card applicants may correspond to the same or different transaction types, and different transaction types satisfy different transaction characteristics, including but not limited to transaction field preference characteristics, transaction volume characteristics, transaction frequency characteristics, etc.

[0069] In this embodiment, the transaction data of the credit card application object is classified by clustering to determine the target transaction cluster to which the transaction data of the credit card application object belongs. Optionally, the transaction data of the credit card application object and the transaction data of the applied object are clustered to obtain multiple clusters, the cluster to which the transaction data of the credit card application object belongs is the target transaction cluster, the transaction type of the transaction data of the credit card application object is the same as the transaction type corresponding to the target transaction cluster, and the transaction characteristics of the target transaction cluster can represent the transaction characteristics of the transaction data of the credit card application object.

[0070] Optionally, clustering is performed on the transaction data in the multi-dimensional data, including: obtaining historical transaction data of multiple objects, performing clustering based on the historical transaction data of the multiple objects to obtain multiple transaction clusters; obtaining the cluster center of each of the transaction clusters, and determining the target transaction cluster to which the transaction data in the multi-dimensional data belongs based on the matching data between the transaction data in the multi-dimensional data and the cluster center of each of the transaction clusters. The historical transaction data of multiple objects here may be the historical transaction data of the applied objects. Multiple transaction clusters are obtained by performing clustering on the historical transaction data of multiple objects, and different transaction clusters correspond to different transaction types. The cluster center of each transaction cluster can characterize the transaction type of the transaction cluster.

[0071] The transaction data of the credit card application object is matched with the cluster center of each of the transaction clusters to obtain matching data. For example, the matching data can be determined by calculating the similarity between the transaction data of the credit card application object and the cluster center of the transaction cluster. The transaction cluster corresponding to the maximum matching data is determined as the target transaction cluster to which the transaction data of the credit card application object belongs.

[0072] Optionally, in order to avoid multiple reliance on cluster centers, the cluster center of each transaction cluster and random transaction data in the transaction cluster can be obtained, and the first similarity between the transaction data of the credit card application object and the cluster center of the transaction cluster and the second similarity between the transaction data of the credit card application object and the random transaction data of the transaction cluster can be determined. The matching data between the transaction data of the credit card application object and the transaction cluster can be calculated based on the mean or weighted value of the first similarity and the second similarity, and the target transaction cluster to which the transaction data of the credit card application object belongs can be determined based on the matching data between the transaction data of the credit card application object and each transaction cluster. The transaction type corresponding to the transaction data of the credit card application object is characterized by the cluster center of the target transaction cluster.

[0073] In some embodiments, the multi-dimensional data is statistically processed based on statistical rules corresponding to multiple statistical indicators, including: pattern recognition of financial behavior data in the multi-dimensional data to determine the credit behavior pattern corresponding to the financial behavior data. The credit behavior pattern can be understood as a credit pattern of the interaction between the credit card applicant and the financial institution, including but not limited to a good credit model, a medium credit model, or a no credit model. Different credit behavior patterns can characterize the credit level of the credit card applicant.

[0074] The financial behavior data includes repayment behavior data; the repayment behavior data may include actual repayment time information and actual repayment amount.

[0075] Optionally, pattern recognition is performed on the financial behavior data in the multi-dimensional data to determine the credit behavior pattern corresponding to the financial behavior data, including: obtaining expected repayment behavior data corresponding to each repayment behavior data, and determining overdue behavior data, as well as the overdue duration and overdue repayment amount of each overdue behavior data based on the expected repayment behavior data corresponding to each repayment behavior data; and determining the credit behavior pattern based on one or more of the statistical number of overdue behavior data, the overdue duration and overdue repayment amount of each overdue behavior data.

[0076] The expected repayment behavior data includes expected repayment time information and expected repayment amount. The repayment behavior data is compared with the corresponding expected repayment behavior data. When the actual repayment time information is earlier than the expected repayment time information and the actual repayment amount is equal to the expected repayment amount, it is determined that the repayment behavior data does not belong to overdue behavior data. When the actual repayment time information is later than the expected repayment time information, and / or the actual repayment amount is less than the expected repayment amount, it is determined that the repayment behavior data belongs to overdue behavior data. The overdue duration of the overdue behavior data can be the time difference between the actual repayment time information and the expected repayment time information when the actual repayment time information is later than the expected repayment time information, and the overdue repayment amount is the difference between the actual repayment amount and the expected repayment amount when the actual repayment amount is less than the expected repayment amount.

[0077] Among them, the greater the statistical number of overdue behavior data, the longer the overdue duration, and the larger the overdue repayment amount, the lower the credit level of the credit card applicant. Optionally, the overdue behavior data corresponding to each credit behavior pattern is pre-set in the data range of statistical number, overdue duration, and overdue repayment amount, and the overdue behavior data corresponding to the financial behavior data of the credit card applicant is matched with the above credit behavior patterns to determine the credit behavior pattern corresponding to the financial behavior data of the credit card applicant. Optionally, the statistical number, overdue duration, and overdue repayment amount of the overdue behavior data corresponding to the financial behavior data of the credit card applicant are weightedly calculated to obtain expected indicator data. The overdue indicator range corresponding to each credit behavior pattern is pre-set, and the expected indicator data corresponding to the financial behavior data of the credit card applicant is matched with the overdue indicator range corresponding to each credit behavior pattern to determine the credit behavior pattern corresponding to the financial behavior data of the credit card applicant.

[0078] On the basis of the above-mentioned embodiment, the credit card application object is classified by multi-dimensional data to obtain the category of the credit card application object, and different categories of the credit card application object may correspond to different risk resistance capabilities. Optionally, the multi-dimensional data may include income information and liability information, and the credit card application object is classified by the income information and liability information, and the surplus stability corresponding to the income information and liability information. Among them, the higher the surplus corresponding to the income information and liability information, the higher the surplus stability, and the higher the risk resistance capability. Multiple categories and the surplus corresponding to each category and the surplus stability conditions can be pre-set, and the category of the credit card application object is determined based on the matching of the income information and liability information in the multi-dimensional data with each category. The category of the credit card application object may include multiple level categories, such as n categories such as level 1, level 2 and level 3, and the number of categories here can be set according to demand.

[0079] Optionally, the credit card application object is classified based on the text data in the multi-dimensional data to obtain the category of the credit card application object, including: extracting key information from the text data in the multi-dimensional data based on a natural language processing model; performing entity recognition on the key information to obtain entity information; and classifying the credit card application object based on the entity information to obtain the category of the credit card application object.

[0080] The text data in the multi-dimensional data includes but is not limited to documents such as work certificates, income certificates, and identity certificates. Through the natural language processing model, key information is extracted from the above text data, and entity information is extracted. For example, the entity information includes but is not limited to entity words such as formula names and job titles that can represent job stability or income information.

[0081] The extracted entity information is classified by a pre-trained object classification model to obtain the category of the credit card application object. The object classification model is trained by the entity information and classification labels of the sample objects.

[0082] On the basis of the above-mentioned embodiment, the credit card application text information can be understood as the text information entered by the credit card applicant in order to apply for a credit card, which may include but is not limited to the purpose of the credit card application, the demand and the repayment plan. The attitude of the credit card applicant towards the credit card application can be determined through the credit card application text information. If the self-description of the credit card applicant is positive and sincere, it may indicate that he has a clear plan for the use of the credit card and a good willingness to repay; on the contrary, if the self-description of the credit card applicant is negative and perfunctory, further information verification is required. By performing sentiment classification on the credit card application text information, the sentiment type corresponding to the credit card application text information is obtained. The sentiment type corresponding to the credit card application text information is positive sentiment, which indicates that the attitude of the credit card applicant is extremely positive. The sentiment type corresponding to the credit card application text information is negative sentiment, which indicates that the attitude of the credit card applicant is extremely negative.

[0083] For example, the credit card application text information can be classified and processed by a sentiment classification model. Optionally, sentiment classification is performed based on the credit card application text information to determine the sentiment type corresponding to the credit card application text information, including: performing word segmentation processing on the credit card application text information to determine multiple to-be-processed word segments; for each to-be-processed word segment, determining the context word segment corresponding to the to-be-processed word segment, the context word segment includes at least one word segment before the to-be-processed word segment and at least one word segment after the to-be-processed word segment; determining the target vector of the to-be-processed word segment based on the to-be-processed word segment and the context word segment corresponding to the to-be-processed word segment; and classifying the target vectors of each to-be-processed word segment based on a pre-trained sentiment classification model to obtain the sentiment type corresponding to the credit card application text information.

[0084] Among them, through word segmentation processing, multiple word segments corresponding to the credit card application text information are obtained, stop words and prohibited words are removed from the multiple word segments, and the remaining word segments are determined as the word segments to be processed. The word segments to be processed have a temporal relationship in the credit card application text information, and based on the temporal relationship of the word segments to be processed, at least one word before the word to be processed and at least one word after the word to be processed are determined as context word segments of the word to be processed.

[0085] Optionally, the context segmentation and the segmentation to be processed are converted into word vectors based on the word vector model to obtain the word vector of the context segmentation and the word vector of the segmentation to be processed. The word vector of the segmentation to be processed and the word vector of the context segmentation corresponding to the segmentation to be processed can form a target vector of the segmentation to be processed. The target vector of each segmentation to be processed can form a matrix to be processed, and the target vector of each segmentation to be processed or the matrix to be processed is input into a pre-trained sentiment classification model to obtain the sentiment type corresponding to the credit card application text information.

[0086] Optionally, a preset part-of-speech database is pre-constructed, and the part-of-speech feature library is used to record the different part-of-speech features of each segmentation. Multiple part-of-speech features corresponding to the segmentation to be processed are found in the part-of-speech feature library, and the degree of match between each part-of-speech feature and the word vector of the context segmentation corresponding to the segmentation to be processed is determined, and the part-of-speech feature with the highest matching degree is determined as the part-of-speech feature matched by the context feature. The part-of-speech features matched by the context feature and the word vector of the segmentation to be processed can form a target vector for the segmentation to be processed. The target vectors of each segmentation to be processed can form a matrix to be processed, and the target vectors or the matrix to be processed of each segmentation to be processed are input into the pre-trained sentiment classification model to obtain the sentiment type corresponding to the credit card application text information.

[0087] In this embodiment, by determining statistical data, categories of credit card application objects, and sentiment types corresponding to credit card application text information, more comprehensive and effective features are provided for processing credit card application information, further improving the accuracy of credit card application information processing.

[0088] On the basis of the above embodiment, a multi-dimensional prediction is performed on the credit card application object based on the multi-dimensional data, multiple statistical data, one or more of the categories and the sentiment types to obtain a multi-dimensional prediction result, providing a reference for determining the processing result of the credit card application information.

[0089] Optionally, the multi-dimensional prediction includes one or more of a credit risk prediction and an anti-fraud prediction, and accordingly, the multi-dimensional prediction result includes one or more of a credit risk prediction result and an anti-fraud prediction result. The credit risk prediction result includes a credit risk probability. The higher the credit risk probability, the higher the credit risk of the credit card applied for by the credit card applicant during use, and the greater the probability of being unable to repay. The anti-fraud prediction result includes a fraud probability; the higher the fraud probability, the higher the possibility of fraudulent behavior by the credit card applicant.

[0090] Optionally, the multi-dimensional data is predicted by a pre-trained anti-fraud model to obtain the anti-fraud prediction result. The anti-fraud model may be one or more of a regression model, a neural network model, a decision tree model, etc. For example, different types of anti-fraud models are pre-set, and multi-dimensional data is predicted by different types of anti-fraud models to obtain multiple anti-fraud prediction results, and the target anti-fraud prediction result is determined based on the mean, maximum value or weighted value of the multiple anti-fraud prediction results.

[0091] Optionally, the multi-dimensional data, the plurality of statistical data, the categories and the sentiment types are predicted by a pre-trained risk prediction model to obtain the credit risk prediction result. For example, the risk prediction model may be one or more of a regression model, a neural network model, a decision tree model, etc. The multi-dimensional data, the plurality of statistical data, the categories and the sentiment types are used as input information of the risk prediction model to provide comprehensive input information, thereby facilitating the accuracy of providing the credit risk prediction result.

[0092] The structures of the anti-fraud model and the risk prediction model are not limited in this embodiment. The anti-fraud model and the risk prediction model can be obtained by evaluating and screening multiple machine learning models with different structures. For example, taking the risk prediction model as an example, multiple initial risk prediction models with different structures are set, and multiple initial risk prediction models are trained to obtain multiple risk prediction models. Multiple risk prediction models are tested, and the performance of multiple risk prediction models is evaluated by one or more indicators such as accuracy, recall rate, precision, F1 score, ROC curve and AUC corresponding to the multiple risk prediction models, and the risk prediction model with the best performance is determined as the target risk prediction model.

[0093] On the basis of the above-mentioned embodiment, the processing result of the credit card application information is determined by the multi-dimensional prediction result. For example, the processing result of the credit card application information can be determined based on the multi-dimensional prediction result through a pre-set decision model. Specifically, one or more of the credit risk prediction result and the anti-fraud prediction result are input into the pre-trained decision model to obtain the processing result of the credit card application information; the processing result of the credit card application information can be the application passed or the application failed. The processing result of the credit card application information is displayed to the credit card approval end to provide a reference for the credit card approval end to approve the credit card application information. Optionally, the decision model can be updated regularly or in real time, and is obtained based on the multi-dimensional prediction results and actual approval result information corresponding to the historical credit card application object. It can be understood that the approval result of the historical credit card application object is in line with market changes and changes in laws and regulations. Correspondingly, the processing result of the regularly updated decision model is also in line with market changes and changes in laws and regulations, and is highly flexible.

[0094] Optionally, the processing result of the credit card application information is determined based on the multi-dimensional prediction result of the credit card application object, including: when the credit risk probability in the credit risk prediction result is greater than a first threshold, and / or the fraud probability in the anti-fraud prediction result is greater than a second threshold, the processing result of the credit card application information is determined to be application rejection; when the credit risk probability in the credit risk prediction result is less than the first threshold, and the fraud probability in the anti-fraud prediction result is less than the second threshold, the processing result of the credit card application information is determined to be application approval. Wherein, the first threshold and the second threshold are respectively preset and can be determined based on the credit card approval scale. Optionally, the first threshold and the second threshold can be adjusted according to the approval requirements.

[0095] It is understandable that the credit card application process may include multiple nodes, such as data processing nodes, classification nodes for credit card application objects, sentiment classification nodes for credit card application text information, prediction nodes, and decision nodes. The above multiple nodes may be executed in parallel or in series, and in response to progress query conditions, the processed nodes and unprocessed nodes are determined and progress information is fed back, which may be the number of processed nodes, the proportion of processed nodes, or the estimated processing time of unprocessed nodes.

[0096] The technical solution provided in this embodiment obtains multi-dimensional data of credit card application objects, performs statistical processing on the multi-dimensional data, and obtains multiple statistical data; classifies the credit card application objects through the multi-dimensional data to obtain the categories of the credit card application objects; and performs sentiment classification on the credit card application text information to obtain the sentiment type corresponding to the credit card application text information. The multi-dimensional data, multiple statistical data, one or more of the categories and the sentiment type are used as input information for multi-dimensional prediction to improve the accuracy of multi-dimensional prediction, and the processing result of the credit card application information is determined based on the multi-dimensional prediction result, further improving the accuracy of the processing result of the credit card application information. Compared with the manual processing process, it saves labor costs and time costs and improves processing efficiency.

[0097] Embodiment 2

[0098] Based on the above embodiments, an embodiment of the present invention provides an optional example of a method for processing credit card application information. Specifically, multi-dimensional data of a credit card application object is obtained, wherein the multi-dimensional data includes basic information of the object, transaction data, and financial behavior data, wherein the multi-dimensional data may be structured data and / or unstructured data. For example, basic information of the object includes but is not limited to occupation, educational background, residence information, income information, debt information, etc.; financial behavior data includes but is not limited to credit card usage information, repayment behavior data, number of credit inquiries, etc. Optionally, the multi-dimensional data may also include credit reports, public records (such as court judgments and bankruptcy records, etc.) and social media data of the credit card application object.

[0099] Perform missing value processing on multi-dimensional data, i.e. fill missing values ​​or delete missing values, etc. Perform data cleaning on multi-dimensional data, including but not limited to removing duplicate data, correcting erroneous data, standardizing data formats, etc.

[0100] Data mining is performed on multi-dimensional data to obtain statistical data corresponding to multiple statistical indicators, such as derivative calculation data, transaction types corresponding to transaction data, and credit behavior patterns corresponding to financial behavior data.

[0101] Create object classification models, sentiment classification models, credit risk prediction models, anti-fraud models and decision models. Each of the above models is a machine learning model, and any model can be one or more of a logistic regression model, a decision tree, a random forest, a support vector machine and a deep learning model. Among them, the deep learning model can be a fully connected neural network model, a convolutional neural network model, a recurrent neural network model, a graph neural network model, a long short-term memory network model and a gated recurrent unit model. Each of the above models can be updated regularly to ensure the processing performance of the model.

[0102] Through the object classification model, the credit card application object is classified based on the text data in the multi-dimensional data to obtain the category of the credit card application object; through the sentiment classification model, the credit card application text information is sentimentally classified to determine the sentiment type corresponding to the credit card application text information; through the credit risk prediction model, the multi-dimensional data, the multiple statistical data, the category and the sentiment type are used to predict the credit risk to obtain the credit risk prediction result; based on the anti-fraud model, the multi-dimensional data is used to perform anti-fraud prediction to obtain the anti-fraud prediction result; the credit risk prediction result and the anti-fraud prediction result are processed through the decision model to obtain the processing result of the credit card application information.

[0103] Embodiment 3

[0104] Figure 2 Schematic diagram of a structure of a credit card application information processing device provided by Embodiment 3 of the present invention. Figure 2 As shown, the device comprises:

[0105] An information acquisition module 210 is used to acquire credit card application information, wherein the credit card application information includes multi-dimensional data of a credit card application object and credit card application text information;

[0106] A statistical indicator determination module 220 is used to perform statistical processing on the multi-dimensional data based on statistical rules corresponding to the multiple statistical indicators, so as to obtain statistical data corresponding to the multiple statistical indicators;

[0107] A classification module 230, configured to classify the credit card application object based on the text data in the multi-dimensional data to obtain a category of the credit card application object;

[0108] The sentiment classification module 240 is used to perform sentiment classification based on the credit card application text information, and determine the sentiment type corresponding to the credit card application text information, wherein the sentiment type includes positive sentiment and negative sentiment;

[0109] A multi-dimensional prediction module 250, configured to perform a multi-dimensional prediction on the credit card application object based on the multi-dimensional data, the plurality of statistical data, one or more of the categories and the sentiment types, and obtain a multi-dimensional prediction result of the credit card application object;

[0110] The processing result determination module 260 is used to determine the processing result of the credit card application information based on the multi-dimensional prediction result of the credit card application object.

[0111] On the basis of the above embodiment, optionally, the statistical indicator determination module 220 is used to: perform derivative calculation on at least one item of the multi-dimensional data based on a pre-set derivative calculation rule to determine the derived indicator data; wherein the derived indicator data includes one or more of the following: trend information of any data, association information between at least two data, and mean and variance of any data

[0112] Optionally, the statistical indicator determination module 220 is further used to: perform clustering processing on the transaction data in the multi-dimensional data to obtain a target transaction cluster to which the transaction data belongs, wherein the target transaction cluster represents a transaction type corresponding to the transaction data.

[0113] Optionally, the statistical indicator determination module 220 is also used to: obtain historical transaction data of multiple objects, perform clustering processing based on the historical transaction data of the multiple objects to obtain multiple transaction clusters; obtain the cluster center of each of the transaction clusters, and determine the target transaction cluster to which the transaction data in the multi-dimensional data belongs based on the matching data between the transaction data in the multi-dimensional data and the cluster center of each of the transaction clusters.

[0114] Optionally, the statistical indicator determination module 220 is further used to: perform pattern recognition on the financial behavior data in the multi-dimensional data to determine the credit behavior pattern corresponding to the financial behavior data.

[0115] Optionally, the financial behavior data includes repayment behavior data;

[0116] Optionally, the statistical indicator determination module 220 is also used to: obtain expected repayment behavior data corresponding to each repayment behavior data, determine overdue behavior data, and the overdue duration and overdue repayment amount of each overdue behavior data based on the expected repayment behavior data corresponding to each repayment behavior data; determine the credit behavior pattern based on one or more of the statistical number of overdue behavior data, the overdue duration and overdue repayment amount of each overdue behavior data.

[0117] Optionally, the classification module 230 is also used to: extract key information from the text data in the multi-dimensional data based on a natural language processing model; perform entity recognition on the key information to obtain entity information; and classify the credit card application object based on the entity information to obtain the category of the credit card application object.

[0118] Optionally, the sentiment classification module 240 is also used to: perform word segmentation processing on the credit card application text information to determine multiple word segmentations to be processed; for each word segmentation to be processed, determine the context word segmentation corresponding to the word segmentation to be processed, the context word segmentation includes at least one word segmentation located before the word segmentation to be processed and at least one word segmentation located after the word segmentation to be processed; determine the target vector of the word segmentation to be processed based on the word segmentation to be processed and the context word segmentation corresponding to the word segmentation to be processed; classify the target vector of each word segmentation to be processed based on a pre-trained sentiment classification model to obtain the sentiment type corresponding to the credit card application text information.

[0119] Optionally, the multi-dimensional prediction includes one or more of a credit risk prediction and an anti-fraud prediction, and correspondingly, the multi-dimensional prediction result includes one or more of a credit risk prediction result and an anti-fraud prediction result.

[0120] Optionally, the multidimensional prediction module 250 is used to: predict the multidimensional data through a pre-trained anti-fraud model to obtain the anti-fraud prediction result; and / or predict the multidimensional data, the multiple statistical data, the categories and the sentiment types through a pre-trained risk prediction model to obtain the credit risk prediction result.

[0121] Optionally, the credit risk prediction result includes a credit risk probability, and the anti-fraud prediction result includes a fraud probability;

[0122] The processing result determination module 260 is further used to: input one or more of the credit risk prediction result and the anti-fraud prediction result into a pre-trained decision model to obtain the processing result of the credit card application information; or,

[0123] When the credit risk probability in the credit risk prediction result is greater than a first threshold, and / or the fraud probability in the anti-fraud prediction result is greater than a second threshold, the processing result of the credit card application information is determined to be application rejection; when the credit risk probability in the credit risk prediction result is less than the first threshold, and the fraud probability in the anti-fraud prediction result is less than the second threshold, the processing result of the credit card application information is determined to be application approval.

[0124] The credit card application information processing device provided in the embodiment of the present invention can execute the credit card application information processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0125] Embodiment 4

[0126] Figure 3 1 is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0127] like Figure 3 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0128] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0129] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as a method for processing credit card application information.

[0130] In some embodiments, the method for processing credit card application information may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for processing credit card application information described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the method for processing credit card application information in any other appropriate manner (e.g., by means of firmware).

[0131] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0132] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0133] Embodiment 5

[0134] Embodiment 5 of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to execute a method for processing credit card application information, the method comprising:

[0135] Acquire credit card application information, the credit card application information includes multi-dimensional data of a credit card application object and credit card application text information; perform statistical processing on the multi-dimensional data based on statistical rules corresponding to multiple statistical indicators, and obtain statistical data corresponding to multiple statistical indicators; perform classification processing on the credit card application object based on text data in the multi-dimensional data, and obtain the category of the credit card application object; perform sentiment classification based on the credit card application text information, and determine the sentiment type corresponding to the credit card application text information, and the sentiment type includes positive sentiment and negative sentiment; perform multi-dimensional prediction on the credit card application object based on the multi-dimensional data, multiple statistical data, one or more of the categories and the sentiment types, and obtain the multi-dimensional prediction result of the credit card application object; determine the processing result of the credit card application information based on the multi-dimensional prediction result of the credit card application object. .

[0136] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0137] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0138] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0139] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0140] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0141] Embodiment 6

[0142] Embodiment 6 of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the method for processing credit card application information according to any embodiment of the present invention.

[0143] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for processing credit card application information, characterized in that: include: Acquire credit card application information, wherein the credit card application information includes multi-dimensional data of the credit card application object and credit card application text information; Performing statistical processing on the multi-dimensional data based on statistical rules corresponding to the multiple statistical indicators respectively, to obtain statistical data corresponding to the multiple statistical indicators; Classify the credit card application object based on the text data in the multi-dimensional data to obtain the category of the credit card application object; Performing sentiment classification based on the credit card application text information to determine the sentiment type corresponding to the credit card application text information, wherein the sentiment type includes positive sentiment and negative sentiment; Perform multi-dimensional prediction on the credit card application object based on the multi-dimensional data, the plurality of statistical data, one or more of the categories and the sentiment types, and obtain a multi-dimensional prediction result of the credit card application object; The processing result of the credit card application information is determined based on the multi-dimensional prediction result of the credit card application object.

2. The method according to claim 1, characterized in that The statistical processing of the multi-dimensional data based on statistical rules corresponding to the multiple statistical indicators respectively includes: Based on a preset derivative calculation rule, performing derivative calculation on at least one item of the multi-dimensional data to determine derivative indicator data; The derived indicator data includes one or more of the following: trend information of any data, correlation information between at least two data, and mean and variance of any data.

3. The method according to claim 1, characterized in that The statistical processing of the multi-dimensional data based on statistical rules corresponding to the multiple statistical indicators respectively includes: Clustering is performed on the transaction data in the multi-dimensional data to obtain a target transaction cluster to which the transaction data belongs, wherein the target transaction cluster represents a transaction type corresponding to the transaction data.

4. The method according to claim 2, characterized in that: The clustering of the transaction data in the multi-dimensional data includes: Acquire historical transaction data of multiple objects, perform clustering based on the historical transaction data of multiple objects, and obtain multiple transaction clusters; The cluster center of each of the transaction clusters is obtained, and based on the matching data between the transaction data in the multi-dimensional data and the cluster center of each of the transaction clusters, the target transaction cluster to which the transaction data in the multi-dimensional data belongs is determined.

5. The method according to claim 1, characterized in that The statistical processing of the multi-dimensional data based on statistical rules corresponding to the multiple statistical indicators respectively includes: Perform pattern recognition on the financial behavior data in the multi-dimensional data to determine the credit behavior pattern corresponding to the financial behavior data.

6. The method according to claim 5, characterized in that The financial behavior data includes repayment behavior data; The performing pattern recognition on the financial behavior data in the multi-dimensional data to determine the credit behavior pattern corresponding to the financial behavior data includes: Obtaining expected repayment behavior data corresponding to each repayment behavior data, and determining overdue behavior data, as well as overdue duration and overdue repayment amount of each overdue behavior data based on the expected repayment behavior data corresponding to each repayment behavior data; The credit behavior pattern is determined based on one or more of the statistical number of the overdue behavior data, the overdue duration of each overdue behavior data, and the overdue repayment amount.

7. The method according to claim 1, characterized in that The classifying the credit card application object based on the text data in the multi-dimensional data to obtain the category of the credit card application object includes: Extracting key information from text data in the multi-dimensional data based on a natural language processing model; Performing entity recognition on the key information to obtain entity information; The credit card application object is classified based on the entity information to obtain the category of the credit card application object.

8. The method according to claim 1, characterized in that The performing sentiment classification based on the credit card application text information to determine the sentiment type corresponding to the credit card application text information includes: Performing word segmentation processing on the credit card application text information to determine a plurality of word segments to be processed; For each of the to-be-processed participles, determining a context participle corresponding to the to-be-processed participle, the context participles comprising at least one participle before the to-be-processed participle and at least one participle after the to-be-processed participle; Determine a target vector of the to-be-processed segmented word based on the to-be-processed segmented word and the context segmented word corresponding to the to-be-processed segmented word; Based on the pre-trained sentiment classification model, the target vector of each to-be-processed word segment is classified and processed to obtain the sentiment type corresponding to the credit card application text information.

9. The method according to claim 1, characterized in that: The multi-dimensional prediction includes one or more of a credit risk prediction and an anti-fraud prediction, and correspondingly, the multi-dimensional prediction result includes one or more of a credit risk prediction result and an anti-fraud prediction result.

10. The method according to claim 9, characterized in that The performing multi-dimensional prediction on the credit card application object based on the multi-dimensional data, the plurality of statistical data, one or more of the categories and the sentiment types to obtain the multi-dimensional prediction result of the credit card application object includes: Predicting the multi-dimensional data using a pre-trained anti-fraud model to obtain the anti-fraud prediction result; and / or, The multi-dimensional data, the plurality of statistical data, the categories and the sentiment types are predicted through a pre-trained risk prediction model to obtain the credit risk prediction result.

11. The method according to claim 9, characterized in that The credit risk prediction result includes the credit risk probability, and the anti-fraud prediction result includes the fraud probability; The step of determining the processing result of the credit card application information based on the multi-dimensional prediction result of the credit card application object includes: Inputting one or more of the credit risk prediction result and the anti-fraud prediction result into a pre-trained decision model to obtain a processing result of the credit card application information; or, When the credit risk probability in the credit risk prediction result is greater than a first threshold, and / or the fraud probability in the anti-fraud prediction result is greater than a second threshold, the processing result of the credit card application information is determined to be application rejection; when the credit risk probability in the credit risk prediction result is less than the first threshold, and the fraud probability in the anti-fraud prediction result is less than the second threshold, the processing result of the credit card application information is determined to be application approval.

12. A device for processing credit card application information, characterized in that: include: An information acquisition module, used to acquire credit card application information, wherein the credit card application information includes multi-dimensional data of the credit card application object and credit card application text information; A statistical indicator determination module, used to perform statistical processing on the multi-dimensional data based on statistical rules corresponding to the multiple statistical indicators, to obtain statistical data corresponding to the multiple statistical indicators; A classification module, used for classifying the credit card application object based on the text data in the multi-dimensional data to obtain the category of the credit card application object; A sentiment classification module, used to perform sentiment classification based on the credit card application text information, and determine the sentiment type corresponding to the credit card application text information, wherein the sentiment type includes positive sentiment and negative sentiment; A multi-dimensional prediction module, used for performing multi-dimensional prediction on the credit card application object based on the multi-dimensional data, a plurality of the statistical data, one or more of the categories and the sentiment types, to obtain a multi-dimensional prediction result of the credit card application object; A processing result determination module is used to determine the processing result of the credit card application information based on the multi-dimensional prediction result of the credit card application object.

13. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for processing credit card application information according to any one of claims 1-11.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for processing credit card application information according to any one of claims 1-11 when executed.

15. A computer program product, characterized in that The computer program product comprises a computer program, which, when executed by a processor, implements the method for processing credit card application information according to any one of claims 1-11.

Images