Post-loan dynamic reminding method and device, equipment, storage medium and product

Abstract

The embodiment of the invention provides a post-loan dynamic reminding method and device, equipment, a storage medium and a product, and relates to the field of finance and related fields thereof. The method comprises the steps of obtaining image data of a loan approval document, and performing text extraction processing on the image data to obtain text information corresponding to the loan approval document; analyzing and processing the text information through a natural language technology, and determining loan information; according to the loan information, generating a repayment reminding task of the loan approval document; and based on historical behavior data of a customer corresponding to the loan approval document, dynamically adjusting the repayment reminding task to obtain a target repayment reminding task, and based on the target repayment reminding task, performing user reminding. According to the method, dynamic adjustment of post-loan reminding is realized, and post-loan management efficiency is improved.

Description

Technical Field

[0001] This application relates to the financial field and related fields, and in particular to a method, device, equipment, storage medium and product for post-loan dynamic reminders. Background Technology

[0002] With changing consumption concepts and the widespread availability of financial services, more and more people are choosing loans to meet their consumption needs. Loan customers should abide by the loan agreement and repay on time. Post-loan management is a crucial aspect of financial institutions' effective control of loan risk, and financial institutions should conduct thorough post-loan management to avoid financial losses.

[0003] Currently, the post-loan management process often relies on manual review of loan application materials and setting reminder tasks based on these materials to notify loan customers before the repayment deadline.

[0004] However, manually checking loan application materials is prone to omissions and delays. Large financial institutions have a large number of customers and complex and diverse documents, making it even more difficult to manually check loan application materials and resulting in extremely low manual management efficiency. Summary of the Invention

[0005] This application provides a method, device, equipment, storage medium, and product for post-loan dynamic reminders, addressing the problem that manually checking loan material information is more difficult and manual management is extremely inefficient.

[0006] Firstly, this application provides a method for post-loan dynamic notification, including:

[0007] Obtain image data of the loan approval document, perform text extraction processing on the image data, and obtain the text information corresponding to the loan approval document;

[0008] The text information is parsed and processed using natural language processing technology to determine loan information;

[0009] Based on the loan information, generate a repayment reminder task for the loan approval document;

[0010] Based on the customer's historical behavior data corresponding to the loan approval document, the repayment reminder task is dynamically adjusted to obtain the target repayment reminder task, and the user is reminded based on the target repayment reminder task.

[0011] Secondly, this application provides a post-loan dynamic reminder device, comprising:

[0012] The acquisition module is used to acquire image data of the loan approval document, perform text extraction processing on the image data, and obtain the text information corresponding to the loan approval document.

[0013] The processing module is used to parse and process the text information using natural language processing technology to determine loan information;

[0014] The processing module is also used to generate a repayment reminder task for the loan approval document based on the loan information;

[0015] The reminder module is used to dynamically adjust the repayment reminder task based on the customer's historical behavior data corresponding to the loan approval document, obtain the target repayment reminder task, and provide user reminders based on the target repayment reminder task.

[0016] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0017] The memory stores computer-executed instructions;

[0018] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0019] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0020] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0021] The post-loan dynamic reminder method, device, equipment, storage medium, and product provided in this application acquire image data of the loan approval document, extract text from the image data to obtain the corresponding text information of the loan approval document, converting physical data into digital data, realizing the extraction of text information from the loan approval document, facilitating the understanding of the text meaning in the image, reducing the complexity and error rate of manual operation, and improving the efficiency of text information extraction; through natural language processing, the text information is parsed to determine the loan information, eliminating redundant data in the text information; according to the loan information, a repayment reminder task for the loan approval document is generated, reducing repetitive work for account managers and improving the efficiency of post-loan management; based on the historical behavior data of the customer corresponding to the loan approval document, the repayment reminder task is dynamically adjusted to obtain the target repayment reminder task, and user reminders are given based on the target repayment reminder task, realizing intelligent prediction and adjustment based on the customer's historical behavior data, improving the customer experience. Attached Figure Description

[0022] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0023] Figure 1 A flowchart illustrating a post-loan dynamic reminder method provided in this application embodiment. Figure 1 ;

[0024] Figure 2 A flowchart illustrating a post-loan dynamic reminder method provided in this application embodiment. Figure 2 ;

[0025] Figure 3 A schematic diagram of the structure of a post-loan dynamic reminder device provided in an embodiment of this application;

[0026] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0027] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0028] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0029] It should be noted that the customer information (including but not limited to customer device information, customer personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the customer or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of the relevant data all comply with the relevant laws, regulations, and standards of the relevant countries and regions, and necessary confidentiality measures have been taken. They do not violate public order and good morals, and corresponding operation portals are provided for customers to choose to authorize or refuse.

[0030] Furthermore, this application involves a technical solution that performs big data analysis on customer information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.), and uses artificial intelligence technology to make automated decisions. Based on the automated decision results, it makes decisions that have a significant impact on personal rights and interests, and provides customers with corresponding operation access points for customers to choose to agree to or reject the automated decision results; if the customer chooses to reject, the process will proceed to the expert decision-making process.

[0031] "Multiple" refers to two or more, and other quantifiers are similar. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following objects have an "or" relationship.

[0032] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, products, or apparatus.

[0033] It should be noted that, in the embodiments of this application, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0034] It should be noted that the post-loan dynamic reminder method, device, equipment, storage medium and product provided in this application can be used in the financial field, or in any field other than the financial field. The application field of the post-loan dynamic reminder method, device, equipment, storage medium and product in this application is not limited.

[0035] First, let me explain the terms used in this application:

[0036] Optical Character Recognition (OCR): A technology that converts text in an image into editable text.

[0037] Natural Language Processing (NLP) aims to enable computers to understand and process human language in order to perform a range of tasks.

[0038] With changing consumption concepts and the widespread availability of financial services, more and more people are choosing loans to meet their consumption needs. Loan customers should abide by the loan agreement and repay on time. Post-loan management is a crucial aspect of financial institutions' effective control of loan risk, and financial institutions should conduct thorough post-loan management to avoid financial losses.

[0039] Currently, post-loan management often relies on manual review of loan application materials and the setting of reminders based on this information to notify loan customers before the repayment deadline. However, manual review of loan application materials is prone to omissions and delays. Large financial institutions, with their vast customer base and complex and diverse documents, face even greater challenges in manually reviewing loan application materials, resulting in extremely low efficiency.

[0040] In existing technologies, Optical Character Recognition (OCR) and Natural Language Processing (NLP) have been applied to some document processing scenarios, but there is no mature solution on how to apply them to post-loan management and automate the process by combining them with a post-loan dynamic reminder system.

[0041] The post-loan dynamic reminder method provided in this application extracts text from the image data of the loan approval document, performs natural language processing on the extracted text information, determines the repayment reminder task of the loan approval document, and performs post-loan dynamic reminders, aiming to solve the above-mentioned technical problems of the prior art.

[0042] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0043] Figure 1 A flowchart illustrating a post-loan dynamic reminder method provided in this application embodiment. Figure 1 ,like Figure 1 As shown, the method includes:

[0044] S101. Obtain the image data of the loan approval document, perform text extraction processing on the image data, and obtain the text information corresponding to the loan approval document;

[0045] Image data refers to the digital form corresponding to the paper version of the loan approval document. For example, image data of the loan approval document can be obtained by scanning or photographing it. Text information is used to indicate visual information such as text and signatures in the loan approval document. Text extraction processing refers to the process of extracting text information from image data.

[0046] After obtaining the image data of the loan approval document, text extraction processing is performed on the image data. This can be done using techniques such as OCR, Optical Layout Recognition (OLR), or matching with a predefined loan approval document template. After text extraction processing of the image data, the corresponding text information of the loan approval document is obtained.

[0047] In one possible implementation, the text extraction processing of the image data described above to obtain the text information corresponding to the loan approval document is explained in detail, including:

[0048] The image data is preprocessed; the preprocessed image data is then input into a pre-trained detection model to obtain the text region of the preprocessed image data; the text region is then serialized to obtain the text information corresponding to the loan approval document.

[0049] The preprocessing includes smoothing, grayscale conversion, and binarization. The detection model is obtained by training a convolutional neural network based on historical image data.

[0050] The image data is smoothed using Gaussian blur to obtain a first preprocessed image. This first preprocessed image is then converted to grayscale, compressing color information into single brightness information to obtain a second preprocessed image. This second preprocessed image is then binarized to obtain a third preprocessed image. This third preprocessed image is input into a pre-trained detection model for feature extraction, detecting text regions. A pre-trained Bidirectional Long Short-Term Memory (BiLSTM) network is used to sequence the detected text regions, inferring blurred characters using bidirectional information, thus outputting the recognized text.

[0051] By performing text extraction processing on image data, text information in loan approval documents can be extracted, making it easier to understand the meaning of the text in the image, reducing the complexity and error rate of manual operation, and improving the efficiency of text information extraction.

[0052] S102. Using natural language processing technology, the text information is parsed and processed to determine the loan information;

[0053] Among them, loan information is the key information in the loan approval document. Loan information refers to the loan information in the loan approval document, such as repayment date, loan amount, outstanding amount, repayment method, etc.

[0054] In one possible implementation, a detailed explanation is provided of the above-mentioned method of parsing and processing text information using natural language processing technology to determine loan information, including:

[0055] The text information is decomposed according to the decomposition unit; the decomposed word segments are embedded into the word vector space to obtain word segment vectors, and semantic capture processing is performed on the word segment vectors to obtain semantic word segment vectors; the semantic word segment vectors are labeled and classified to determine the loan information.

[0056] The decomposition units include paragraphs, sentences, words, and at least one of these. The word vector space refers to a continuous space model that transforms the decomposition units into word vectors.

[0057] Based on the decomposition units—paragraphs, sentences, words, or at least one of these—text information identified from image data is decomposed into multiple word segments. These word segments are then embedded into a word vector space to obtain a word vector corresponding to each segment. Semantic capture processing of the word vectors is performed using a multi-layer Transformer structure of a pre-trained language model (Bidirectional Encoder Representations from Transformers, BERT model). Each layer calculates semantic word vectors through a self-attention mechanism. Additional classification layers of the pre-trained language model then identify loan information, such as repayment date and loan amount.

[0058] For example, if the text information includes "repayment before year M month D day", the pre-trained language model will annotate and recognize the date in the text.

[0059] By extracting text using natural language processing technology, redundant data in the text information was removed, and loan information was determined.

[0060] S103. Generate a repayment reminder task for the loan approval document based on the loan information;

[0061] The repayment reminder task includes the reminder time and reminder content, which includes the repayment amount and the final repayment time.

[0062] Based on loan information such as repayment date, loan amount, and outstanding amount, repayment reminder tasks are determined.

[0063] S104. Based on the customer's historical behavior data corresponding to the loan approval document, dynamically adjust the repayment reminder task to obtain the target repayment reminder task, and remind the user based on the target repayment reminder task.

[0064] Based on the customer's historical behavior data corresponding to the loan approval document, the repayment reminder task is dynamically adjusted to generate a personalized target repayment reminder task, and the user is reminded according to the target repayment reminder task.

[0065] In one possible implementation, the above-mentioned user reminder based on the target repayment reminder task is described in detail, including:

[0066] Generate user reminder messages based on the target repayment reminder task; send the user reminder messages to the target users.

[0067] The user reminder information includes the current repayment amount, the latest repayment date, the minimum repayment amount for the current period, and the total outstanding principal and interest. Target users include: account managers and / or loan customers.

[0068] Based on the target repayment reminder task, generate user reminder information including the current repayment amount, the latest repayment time, the current minimum repayment amount, and the total outstanding principal and interest. Send the user reminder information to the target user via email, SMS, or other means. The reminder methods include instant reminders and scheduled reminders.

[0069] Dynamic post-loan reminders were achieved by sending reminder messages to customers; and post-loan management efficiency was improved by sending reminder messages to account managers.

[0070] This application provides a post-loan dynamic reminder method. It acquires image data of a loan approval document, performs text extraction processing on the image data to obtain the corresponding text information, converting physical data into digital data. This enables the extraction of text information from the loan approval document, facilitating the understanding of the text meaning in the image, reducing the complexity and error rate of manual operations, and improving the efficiency of text information extraction. Natural language processing is used to parse the text information, determining the loan information and eliminating redundant data. Based on the loan information, a repayment reminder task for the loan approval document is generated, reducing repetitive work for account managers and improving the efficiency of post-loan management. Based on the historical behavior data of the customer corresponding to the loan approval document, the repayment reminder task is dynamically adjusted to obtain a target repayment reminder task. Based on the target repayment reminder task, user reminders are sent, achieving intelligent prediction and adjustment based on the customer's historical behavior data, thus improving the customer experience.

[0071] Figure 2A flowchart illustrating a post-loan dynamic reminder method provided in this application embodiment. Figure 2 In this embodiment, the repayment reminder task includes the number of reminders, the next reminder time, and the reminder content. Figure 1 Based on the embodiments, a post-loan dynamic reminder method is described in detail, such as... Figure 2 As shown, the method includes:

[0072] S201. Obtain the historical behavior data of the customer corresponding to the loan approval document;

[0073] Historical behavioral data is used to indicate a user's willingness to repay. This data can include, for example, historical repayment records, current debt amount, repayment schedule patterns, and repayment methods.

[0074] Based on the customer's historical repayment records corresponding to the loan approval document, determine the customer's historical behavioral data.

[0075] S202. Use historical behavior data as input to the random forest model, classify the historical behavior data, and obtain the classification results of the random forest model.

[0076] The random forest model is trained based on multiple sample behavioral data.

[0077] Random Forest is a powerful ensemble learning algorithm widely used in tasks such as classification, regression, and feature importance evaluation. It reduces the risk of overfitting and improves the generalization ability of the random forest model by combining multiple decision trees.

[0078] Customers' historical behavior data is input into a random forest model trained based on multiple sample behavior data. The historical behavior data is then classified to obtain the classification results of the random forest model.

[0079]

[0080] in, This is the classification result; It is the first The output of each decision tree It is the number of decision trees.

[0081] S203. Based on the classification results, determine the corresponding customer alert level;

[0082] The classification result can be, for example, a label, or a probability.

[0083] When the classification result is a label, the alert level that maps to the classification result from multiple preset alert levels is selected as the customer's alert level; when the classification result is a probability, the customer's alert level is determined based on the interval of the alert level to which the probability belongs. Different alert levels correspond to different intervals. The customer alert level is determined based on the inclusion relationship between the interval and the classification results.

[0084] For example: The preset reminder levels include: a first reminder level, a second reminder level, and a third reminder level. When the classification result is a label, the classification result can be 1, 2, or 3. When the classification result is 1, the customer's reminder level is determined to be the first reminder level; when the classification result is 2, the customer's reminder level is determined to be the second reminder level; and when the classification result is 3, the customer's reminder level is determined to be the third reminder level.

[0085] S204. Based on the customer reminder level, dynamically adjust the reminder count and reminder content to obtain personalized reminder count and personalized reminder content.

[0086] When a customer's reminder level is Level 3, it indicates that the customer's historical repayment behavior is poor, so the reminder frequency is determined to be multiple times, for example, 3 times; when a customer's reminder level is Level 2, it indicates that the customer's historical repayment behavior is normal, so the reminder frequency can be 1 time, and the reminder content can be a standard reminder, such as: Please repay on time; when a customer's reminder level is Level 1, it indicates that the customer's historical repayment behavior is excellent, so the reminder frequency can be 1 time, and the reminder content is a mild reminder, such as: Friendly reminder.

[0087] S205. Map the next reminder time according to the customer's reminder level and the corresponding personalized adjustment parameters to obtain the personalized reminder time;

[0088] The personalized adjustment parameters are used to dynamically adjust the next reminder time and reminder content.

[0089] Based on the personalized adjustment parameters corresponding to the customer's reminder level, the next reminder time is mapped to obtain the personalized reminder time.

[0090]

[0091] in, For personalized reminder times, For the next reminder time, Personalized adjustment parameters corresponding to the customer's alert level.

[0092] Understandably, if a customer's reminder level corresponds to multiple reminders, there are multiple personalized reminder parameters. The personalized reminder time for each reminder is determined according to these multiple personalized reminder parameters.

[0093] In one possible implementation, this embodiment includes loan start date and reminder period. The repayment reminder task for generating a loan approval document based on the loan information is described in detail, including:

[0094] Based on the reminder cycle and loan start date, determine the next reminder time; based on the loan information, determine the reminder content; based on the next reminder time and reminder content, generate a repayment reminder task for the loan approval document.

[0095] The following formula is used to determine the next reminder time:

[0096]

[0097] in, Set the time for the next reminder; For reminder cycles, The loan start date. This represents the number of periods that have been repaid.

[0098] Based on loan information, such as loan amount and installment period, determine the amount due and the reminder content.

[0099] By using loan information, the content of the reminder and the next reminder time were determined to support personalized adjustments, ensuring the accuracy and interpretability of personalized reminder times and content.

[0100] S206. Based on the number of personalized reminders, the content of personalized reminders, and the time of personalized reminders, determine the target repayment reminder task.

[0101] This embodiment provides a post-loan dynamic reminder method. It acquires historical behavioral data of customers corresponding to loan approval documents, uses this data as input to a random forest model, classifies the historical behavioral data, and obtains the classification results. Based on the customer's historical behavioral data, it predicts the customer's future repayment probability. Based on the classification results, it determines the corresponding customer reminder level, dividing it into different levels. This facilitates personalized adjustments to repayment reminder tasks and provides a basis for such adjustments. Based on the customer reminder level, it dynamically adjusts the reminder frequency and content to obtain personalized reminder frequency and content. According to the customer reminder level and corresponding personalized adjustment parameters, it maps the next reminder time to obtain a personalized reminder time, improving customer experience and reminder efficiency. Based on the personalized reminder frequency, personalized reminder content, and personalized reminder time, it determines the target repayment reminder task, improving post-loan management efficiency.

[0102] Figure 3 This is a schematic diagram of the structure of a post-loan dynamic reminder device provided in an embodiment of this application, as shown below. Figure 3 As shown, the post-loan dynamic reminder device 30 provided in this embodiment includes:

[0103] The acquisition module 301 is used to acquire image data of the loan approval document, perform text extraction processing on the image data, and obtain text information corresponding to the loan approval document.

[0104] The processing module 302 is used to parse and process the text information using natural language processing technology to determine loan information;

[0105] The processing module 302 is also used to generate a repayment reminder task for the loan approval document according to the loan information;

[0106] The reminder module 303 is used to dynamically adjust the repayment reminder task based on the customer's historical behavior data corresponding to the loan approval document, obtain the target repayment reminder task, and provide user reminders based on the target repayment reminder task.

[0107] In one possible implementation, the reminder module 303 is further configured to determine the corresponding customer reminder level based on the customer's historical behavior data corresponding to the loan approval document; and dynamically adjust the repayment reminder task based on the customer reminder level to obtain the target repayment reminder task.

[0108] In one possible implementation, the reminder module 303 is further configured to acquire historical behavior data of the customer corresponding to the loan approval document; use the historical behavior data as input to a random forest model to classify the historical behavior data and obtain the classification result of the random forest model, wherein the random forest model is trained based on multiple sample behavior data; and determine the corresponding customer reminder level based on the classification result.

[0109] In one possible implementation, the repayment reminder task includes the number of reminders, the next reminder time, and the reminder content. The reminder module 303 is further configured to dynamically adjust the number of reminders and the reminder content based on the customer's reminder level to obtain personalized reminders and personalized reminder content; map the next reminder time according to the customer's reminder level and the corresponding personalized adjustment parameters to obtain a personalized reminder time; and determine the target repayment reminder task based on the personalized reminder number, personalized reminder content, and personalized reminder time.

[0110] In one possible implementation, the loan information includes the loan start date and the reminder period. The processing module 302 is further configured to determine the next reminder time based on the reminder period and the loan start date; determine the reminder content based on the loan information; and generate a repayment reminder task for the loan approval document based on the next reminder time and the reminder content.

[0111] In one possible implementation, the acquisition module 301 is further configured to preprocess the image data, the preprocessing including smoothing, grayscale conversion, and binarization; input the preprocessed image data into a pre-trained detection model to obtain the text region of the preprocessed image data, the detection model being trained on a convolutional neural network based on historical image data; and perform serialization processing on the text region to obtain the text information corresponding to the loan approval document.

[0112] In one possible implementation, the processing module 302 is further configured to decompose the text information according to the decomposition units, wherein the decomposition units include paragraphs, sentences, words, and at least one of paragraphs, sentences, and words; embed the decomposed word segments into the word vector space to obtain word segment vectors, and perform semantic capture processing on the word segment vectors to obtain semantic word segment vectors; and perform annotation and classification processing on the semantic word segment vectors to determine loan information.

[0113] In one possible implementation, the reminder module 303 is further configured to generate user reminder information according to the target repayment reminder task; and send the user reminder information to the target user, including: account manager and / or loan customer.

[0114] This embodiment provides a post-loan dynamic reminder device that can execute the method provided in the above-described method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0115] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device 40 provided in this embodiment includes at least one processor 401 and a memory 402. Optionally, the device 40 further includes a communication component 403. The processor 401, memory 402, and communication component 403 are connected via a bus 404.

[0116] In a specific implementation, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to perform the above-described method.

[0117] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0118] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0119] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0120] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0121] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0122] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0123] When integrated units / modules are implemented in hardware, the hardware can be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor can be any suitable hardware processor, such as a CPU, GPU, FPGA, DSP, and ASIC, etc. Unless otherwise specified, the storage unit can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc.

[0124] If the integrated unit / module is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application.

[0125] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application.

[0126] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0127] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0128] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.

[0129] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0130] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0131] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for post-loan dynamic reminders, characterized in that, include: Obtain image data of the loan approval document, perform text extraction processing on the image data, and obtain the text information corresponding to the loan approval document; The text information is parsed and processed using natural language processing technology to determine loan information; Based on the loan information, generate a repayment reminder task for the loan approval document; Based on the customer's historical behavior data corresponding to the loan approval document, the repayment reminder task is dynamically adjusted to obtain the target repayment reminder task, and the user is reminded based on the target repayment reminder task.

2. The method according to claim 1, characterized in that, The step of dynamically adjusting the repayment reminder task based on the customer's historical behavior data corresponding to the loan approval document to obtain the target repayment reminder task includes: Based on the historical behavior data of the customer corresponding to the loan approval document, the corresponding customer alert level is determined; Based on the customer's reminder level, the repayment reminder task is dynamically adjusted to obtain the target repayment reminder task.

3. The method according to claim 2, characterized in that, The step of determining the corresponding customer alert level based on the customer's historical behavior data corresponding to the loan approval document includes: Obtain the historical behavior data of the customer corresponding to the loan approval document; The historical behavior data is used as input to the random forest model to classify the historical behavior data and obtain the classification result of the random forest model. The random forest model is trained based on multiple sample behavior data. Based on the classification results, the corresponding customer alert level is determined.

4. The method according to claim 2, characterized in that, The repayment reminder task includes the number of reminders, the next reminder time, and the reminder content; the dynamic adjustment of the repayment reminder task based on the customer's reminder level to obtain the target repayment reminder task includes: Based on the customer's reminder level, the number of reminders and the reminder content are dynamically adjusted to obtain personalized reminders and personalized reminder content. The next reminder time is mapped according to the customer reminder level and the corresponding personalized adjustment parameters to obtain the personalized reminder time; The target repayment reminder task is determined based on the number of personalized reminders, the personalized reminder content, and the personalized reminder time.

5. The method according to claim 4, characterized in that, The loan information includes the loan start date and reminder period. The task of generating a repayment reminder for the loan approval document based on the loan information includes: The next reminder time will be determined based on the reminder cycle and the loan start date. Based on the loan information, determine the content of the reminder; Based on the next reminder time and reminder content, a repayment reminder task for the loan approval document is generated.

6. The method according to claim 1, characterized in that, The step of performing text extraction processing on the image data to obtain the text information corresponding to the loan approval document includes: The image data is preprocessed, including smoothing, grayscale conversion, and binarization. The preprocessed image data is input into a pre-trained detection model to obtain the text region of the preprocessed image data. The detection model is obtained by training a convolutional neural network based on historical image data. The text region is serialized to obtain the text information corresponding to the loan approval document.

7. The method according to claim 1, characterized in that, The step of parsing and processing the text information using natural language processing technology to determine loan information includes: The text information is decomposed according to the decomposition units, and the decomposition units include paragraphs, sentences, words, and at least one of paragraphs, sentences, and words. The decomposed word segments are embedded into the word vector space to obtain word segmentation vectors, and semantic capture processing is performed on the word segmentation vectors to obtain semantic word segmentation vectors. The semantic word segmentation vectors are labeled and classified to determine loan information.

8. The method according to claim 1, characterized in that, The step of reminding users based on the target repayment reminder task includes: Generate user reminder information according to the stated target repayment reminder task; The user reminder information is sent to the target users, including account managers and / or loan customers.

9. A post-loan dynamic reminder device, characterized in that, include: The acquisition module is used to acquire image data of the loan approval document, perform text extraction processing on the image data, and obtain the text information corresponding to the loan approval document. The processing module is used to parse and process the text information using natural language processing technology to determine loan information; The processing module is also used to generate a repayment reminder task for the loan approval document based on the loan information; The reminder module is used to dynamically adjust the repayment reminder task based on the customer's historical behavior data corresponding to the loan approval document, obtain the target repayment reminder task, and provide user reminders based on the target repayment reminder task.

10. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1 to 8.

12. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 8.

Images