CN117635153A - Client financial transaction risk report generation method and device - Google Patents
Client financial transaction risk report generation method and device Download PDFInfo
- Publication number
- CN117635153A CN117635153A CN202311636141.8A CN202311636141A CN117635153A CN 117635153 A CN117635153 A CN 117635153A CN 202311636141 A CN202311636141 A CN 202311636141A CN 117635153 A CN117635153 A CN 117635153A
- Authority
- CN
- China
- Prior art keywords
- data
- report
- client
- risk
- financial transaction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 84
- 238000004458 analytical method Methods 0.000 claims abstract description 103
- 238000011835 investigation Methods 0.000 claims description 30
- 230000015654 memory Effects 0.000 claims description 26
- 238000004590 computer program Methods 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 24
- 238000003860 storage Methods 0.000 claims description 13
- 230000009191 jumping Effects 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 abstract description 9
- 238000012546 transfer Methods 0.000 description 55
- 238000013473 artificial intelligence Methods 0.000 description 30
- 238000012216 screening Methods 0.000 description 28
- 230000008569 process Effects 0.000 description 26
- 238000005516 engineering process Methods 0.000 description 19
- 230000003993 interaction Effects 0.000 description 19
- 230000006399 behavior Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 14
- 238000012544 monitoring process Methods 0.000 description 14
- 238000004891 communication Methods 0.000 description 12
- 230000007547 defect Effects 0.000 description 12
- 238000011160 research Methods 0.000 description 12
- 238000012550 audit Methods 0.000 description 9
- 238000009826 distribution Methods 0.000 description 8
- 238000009825 accumulation Methods 0.000 description 7
- 238000011156 evaluation Methods 0.000 description 7
- 238000012502 risk assessment Methods 0.000 description 7
- 238000010801 machine learning Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 6
- 238000005457 optimization Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 239000000872 buffer Substances 0.000 description 4
- 230000001788 irregular Effects 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 238000012954 risk control Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000013079 data visualisation Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000000586 desensitisation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000003211 malignant effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000000547 structure data Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4016—Transaction verification involving fraud or risk level assessment in transaction processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
- G06N5/041—Abduction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Molecular Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Accounting & Taxation (AREA)
- Medical Informatics (AREA)
- Computer Security & Cryptography (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a method and a device for generating a customer financial transaction risk report, which relate to the technical field of data processing, and the method comprises the following steps: aiming at a target client, determining user input data corresponding to each data feature type and target client acquisition data according to the preset data feature type of the client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; based on a large language model, generating an analysis report of a target client under different data characteristic types according to user input data corresponding to the different data characteristic types and target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode. The invention is used for improving the generation efficiency and flexibility of the financial transaction risk report of the client and improving the user experience.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a device for generating a customer financial transaction risk report.
Background
This section is intended to provide a background or context to the embodiments of the invention that are recited in the claims. The description herein is not admitted to be prior art by inclusion in this section.
At present, the case screening and checking of the customer financial transaction risk is one of the core contents of the banking control financial transaction work, and the main workflow comprises the following steps:
(1) The system generates a suspicious early warning of the system according to the suspicious model;
(2) By manually carrying out early warning on the system and combining analysis of multiple dimensions of customer conditions, whether suspicious cases relate to suspicious needs to be submitted to people is manually confirmed, and in the process, more detailed customer financial transaction risk suspicious screening investigation reports are required to be written for eliminating suspicious cases and reporting suspicious cases.
In the process, more manpower is required to carry out analysis and writing, and meanwhile, the condition that the investigation report writing quality is poor and irregular exists.
For example, the following schemes are adopted to form a customer financial transaction risk report, but new problems are brought about, as follows:
1. when AI (Artificial Intelligence ) is adopted to generate an adjustment report at present, the report is generated only by means of data in a system, and the conditions of stronger templating and homogeneity exist, so that the report does not meet the specification and the requirement of bank financial risk transaction event supervision;
2. The generated report content is only a list of the content, the arrangement of the report key content is not formed, and the effect of assisting the manual screening and judging of the suspicious cases of the financial transaction risks of the clients is limited.
Disclosure of Invention
The embodiment of the invention provides a method for generating a customer financial transaction risk report, which is used for improving the efficiency and flexibility of generating the customer financial transaction risk report and improving user experience, and comprises the following steps:
aiming at a target client, determining user input data corresponding to each data feature type and target client acquisition data according to the preset data feature type of the client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user;
based on a large language model, generating an analysis report of a target client under different data characteristic types according to user input data corresponding to the different data characteristic types and target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode.
The embodiment of the invention also provides a device for generating the customer financial transaction risk report, which is used for improving the efficiency and the flexibility of generating the customer financial transaction risk report and improving the user experience, and comprises the following steps:
the data acquisition module is used for determining user input data corresponding to each data characteristic type and target client acquisition data according to the preset data characteristic type of the client financial transaction risk aiming at the target client; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user;
the report generation module is used for generating an analysis report of the target client under different data characteristic types based on the large language model according to the user input data corresponding to the different data characteristic types and the target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode.
The embodiment of the invention also provides computer equipment, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the client financial transaction risk report generation method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the client financial transaction risk report generation method when being executed by a processor.
The embodiment of the invention also provides a computer program product, which comprises a computer program, wherein the computer program is executed by a processor to realize the client financial transaction risk report generation method.
In the embodiment of the invention, aiming at a target client, user input data and target client acquisition data corresponding to each data feature type are determined according to the preset data feature type of the client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user; based on a large language model, generating an analysis report of a target client under different data characteristic types according to user input data corresponding to the different data characteristic types and target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data and generating an intelligent report expressed in a natural language mode, so that the client financial transaction risk report can be automatically established and generated based on the large language model technology by combining the user input data and the target client collected data, the generation efficiency of the client financial transaction risk report is improved, the defects of templatization and standardization of the traditional AI generation report are overcome, and the flexibility of generating the client financial transaction risk report is improved; by combining the user input data, the manual judgment of the financial transaction risk level in the report is also realized, the accuracy of the screening judgment of the suspicious case of the customer financial transaction risk by the auxiliary staff is improved, and the user experience is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
FIG. 1 is a flowchart of a method for generating a risk report for a financial transaction of a customer according to an embodiment of the present invention;
FIG. 2 is a diagram showing a specific example of a method for generating a risk report for a financial transaction of a customer according to an embodiment of the present invention;
FIG. 3 is a diagram showing a specific example of a method for generating a risk report for a financial transaction of a customer according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating a specific example of acquiring user input data in a method for generating a risk report for a financial transaction according to an embodiment of the present invention;
FIG. 5 is a diagram showing a specific example of a method for generating a risk report for a financial transaction of a customer according to an embodiment of the present invention;
FIG. 6 is a diagram showing an example of a device for generating a risk report for a financial transaction of a customer according to an embodiment of the present invention;
FIG. 7 is a diagram showing an exemplary configuration of a risk report generating device for a financial transaction of a customer according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of a computer device for customer financial transaction risk report generation in accordance with an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present invention and their descriptions herein are for the purpose of explaining the present invention, but are not to be construed as limiting the invention.
The term "and/or" is used herein to describe only one relationship, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist together, and B exists alone. In addition, the term "at least one" herein means any one of a plurality or any combination of at least two of a plurality, for example, including at least one of A, B, C, and may mean including any one or more elements selected from the group consisting of A, B and C.
In the description of the present specification, the terms "comprising," "including," "having," "containing," and the like are open-ended terms, meaning including, but not limited to. Reference to the terms "one embodiment," "a particular embodiment," "some embodiments," "for example," etc., means that a particular feature, structure, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. The sequence of steps involved in the embodiments is used to schematically illustrate the practice of the present application, and is not limited thereto and may be appropriately adjusted as desired.
The data acquisition, storage, use, processing and the like in the technical scheme meet the relevant regulations of national laws and regulations.
Embodiments of the present invention relate to the following terms, as explained below:
1. large language model (LLM, large Language Model): the artificial intelligent AI model with a strong application prospect is a deep learning model trained by using a large amount of text data;
2. man-machine interaction: the interaction between a person and a computer comprises data interaction, image interaction, voice interaction, behavior interaction and the like;
3. AI generation: automatically generating results such as texts, pictures, sound and video according to input instructions through a trained Artificial Intelligence (AI) model;
4. bank funds transfer risk suspicious screening survey report: in the process that a bank funds transfer risk worker manually analyzes suspicious early-warning cases of a bank funds transfer risk system at multiple angles, screening and investigation are carried out on the cases for eliminating the suspicious cases and reporting the suspicious cases based on the principle of 'risk is the same' or 'knowing your clients', and investigation records and reports are reserved.
At present, the case screening and checking of customer financial transaction risk is one of core contents of bank control malignant financial transaction work, and the main workflow comprises the following steps:
(1) The system generates a suspicious early warning of the system according to the suspicious model;
(2) By manually carrying out early warning on the system and combining analysis of multiple dimensions of customer conditions, whether suspicious cases relate to suspicious needs to be submitted to people is manually confirmed, and in the process, more detailed customer financial transaction risk suspicious screening investigation reports are required to be written for eliminating suspicious cases and reporting suspicious cases.
In the process, more manpower is required to carry out analysis and writing, and meanwhile, the condition that the investigation report writing quality is poor and irregular exists.
For example, the following schemes are adopted to form a customer financial transaction risk report, but new problems are brought about, as follows:
1. when AI (Artificial Intelligence ) is adopted to generate an adjustment report at present, the report is generated only by means of data in a system, and the conditions of stronger templating and homogeneity exist, so that the report does not meet the specification and the requirement of bank financial risk transaction event supervision;
2. the generated report content is only a list of the content, the arrangement of the report key content is not formed, and the effect of assisting the manual screening and judging of the suspicious cases of the financial transaction risks of the clients is limited.
In order to solve the above-mentioned problems, an embodiment of the present invention provides a method for generating a risk report of a customer financial transaction, which is used to increase the efficiency and flexibility of generating a risk report of a customer financial transaction and improve the user experience, and referring to fig. 1, the method may include:
step 101: aiming at a target client, determining user input data corresponding to each data feature type and target client acquisition data according to the preset data feature type of the client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user;
step 102: based on a large language model, generating an analysis report of a target client under different data characteristic types according to user input data corresponding to the different data characteristic types and target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode.
In the step (101) of the process (step(s),
User input data and target customer collected data may be categorized and matched according to the type of financial transaction risk data characteristics of the target customer. For example, if the target customer is primarily a stock investor, the user input data and target customer acquisition data may be categorized according to stock trading characteristics, such as trading frequency, trading amount, time to stay, etc. Meanwhile, finer classification can be performed according to factors such as investment style, risk bearing capacity and the like of clients.
In step 102, user input data of the target client under different data feature types and target client collected data can be inferred and analyzed using intelligent analysis report generation techniques based on a large language model. For example, a large language model may be utilized to generate an analysis report regarding stock investment strategies, including suggestions for market trends, stock selections, trading strategies, etc., based on the investment goals and risk bearing capabilities of the user. Meanwhile, the analysis capability and the output effect of the large language model can be optimized and improved continuously according to the feedback and the demands of users.
Through the two steps, the generation and output of the financial transaction risk analysis report of the target client under different data characteristic types can be realized, so that the client is helped to better know own investment risk and formulate corresponding investment strategies. Meanwhile, the accuracy and the practicability of the analysis report can be improved through continuous data accumulation and model optimization, and more intelligent and personalized service is provided for clients.
In the embodiment of the invention, aiming at a target client, user input data and target client acquisition data corresponding to each data feature type are determined according to the preset data feature type of the client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user; based on a large language model, generating an analysis report of a target client under different data characteristic types according to user input data corresponding to the different data characteristic types and target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode. The embodiment of the invention can automatically establish and generate the customer financial transaction risk report based on a large language model technology by combining user input data and target customer acquisition data, thereby improving the generation efficiency of the customer financial transaction risk report, avoiding the disadvantages of templatization and standardization of the traditional AI generation report and improving the flexibility of generating the customer financial transaction risk report; by combining the user input data, the manual judgment of the financial transaction risk level in the report is also realized, the accuracy of the screening judgment of the suspicious case of the customer financial transaction risk by the auxiliary staff is improved, and the user experience is improved.
On the basis of the above embodiments, the present invention may further consider the following aspects:
1. for the processing of user input data and target customer collected data, natural language processing techniques may be employed to convert text information into computer-processable data. For example, text may be preprocessed using word segmentation, part-of-speech tagging, named entity recognition, etc., to extract keywords and entity information related to financial transaction risk.
2. When the analysis report is generated, the large language model can be combined with a deep learning technology to perform deep learning and inference on user input data and target client collected data, so that a more accurate and comprehensive risk report is generated. For example, historical financial transaction data may be modeled using a deep learning model such as a Recurrent Neural Network (RNN) or long short term memory network (LSTM) to predict future financial transaction risk.
3. The invention can also combine manual auditing and judging to improve the accuracy and reliability of the report while automatically generating the customer financial transaction risk report. For example, a manual auditing link can be set, and professionals audit and verify the generated report to ensure the quality and credibility of the report.
4. The invention can also combine the visualization technology to present the financial transaction risk report to the user in a more visual way. For example, the data in the risk report can be displayed in a more visual form by using graphical means such as a bar graph, a line graph, a pie chart and the like, so that the reading and understanding ability of a user is improved.
In summary, the embodiment of the invention combines the technologies of natural language processing, deep learning, visualization and the like, achieves the purpose of automatically and intelligently generating the customer financial transaction risk report, improves the efficiency and accuracy of generating the report, and improves the user experience.
When the method is implemented, firstly, aiming at a target client, user input data and target client acquisition data corresponding to each data feature type are determined according to the preset data feature type of the financial transaction risk of the client; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user.
In specific implementation, for a target client, first, according to a preset data feature type of a client financial transaction risk, user input data corresponding to each data feature type and target client acquisition data can be determined. These user input data are used primarily to characterize the financial transaction risk level that the user input describes the existence of the target customer for the corresponding data characteristic type. And the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in the historical financial transaction data of the user.
After determining these data feature types, the data may be processed and analyzed using machine learning algorithms. These algorithms may include regression analysis, decision trees, neural networks, etc. for modeling user input data and target customer collected data and predicting the level of risk that the target customer may have in future financial transactions.
At the same time, a large amount of historical financial transaction data may also be processed and analyzed using data mining techniques to discover patterns and laws hidden in the data. These patterns and laws may be used to predict and pre-warn future financial transactions to avoid possible financial risks.
Finally, the model may be continuously optimized and updated in order to improve the accuracy and reliability of the predictions. This can be achieved by using online learning techniques, enabling models to be self-optimized and learned continuously from up-to-date data to accommodate changes in market and continuous changes in customer demand.
In the embodiment, aiming at the problem that the generated report content in the prior art is only a list of content, and the arrangement of report key content is not formed, and the effect of assisting in manually carrying out bank funds transfer risk suspicious case discrimination judgment is limited, the embodiment of the invention provides the method, firstly aiming at a target client, according to the preset data characteristic types of the client financial transaction risk, determining the user input data corresponding to each data characteristic type and the target client acquisition data, innovatively combining a large language model and a man-machine interaction mode, and improving the defects of templatization and standardization of the AI generated report.
In one embodiment, the data feature types include: the method comprises the steps of one or any combination of a client name, a risk rating condition type, a client basic information data type, a client due investigation related data type, a hit feature hotword data type, a client last year transaction information data feature, a client reported report data feature, a transaction opponent risk data, a transaction opponent collection accumulated amount data feature, a transaction opponent payment accumulated amount data feature and a' sub-transaction channel data feature.
In one embodiment, the data feature types may further include: the client investment preference data type, the client asset configuration data type, the client risk bearing capacity data type, the client investment income data type, the client transaction behavior data type, the client credit rating data type, the client industry data type, the client regional distribution data type, the client age structure data type, the client sex proportion data type and the like.
By acquiring the plurality of data characteristics, the clients can be analyzed and evaluated more comprehensively and accurately, so that the client requirements, risk preferences and behavior patterns can be better known, and more accurate and personalized services and products can be provided for banks. Meanwhile, by carrying out multidimensional analysis and evaluation on clients, a more accurate risk evaluation and risk management tool can be provided for banks, and the banks can be helped to manage and control risks better.
For example, the operator clicks the sub-term analysis page entering the "customer name and risk rating condition", "customer basic information", "due investigation", "hit feature hotword", "last year transaction information", "customer reported report", "opponent suspicious report", "opponent collection accumulation amount five before", "opponent payment accumulation amount five before", "sub-transaction channel transaction condition" to conduct the study, judgment and analysis of each module, thereby forming the above user input data.
For another example, the operator may click on the sub-term analysis page entering "customer name", "risk tag", "due investigation", "account information", "product information", "recent suspicious report", "recent big report" to perform the study, judgment and analysis of each module, thereby forming the above-mentioned user input data.
For example, an operator may enter a sub-term analysis page of "customer base information", "due diligence", "account information" and "product information" to perform research, judgment and analysis of each module by inputting a customer name or risk tag, thereby forming user input data. Meanwhile, an operator can enter a corresponding analysis page to conduct research, judgment and analysis by clicking links of analysis pages such as a client name and risk rating condition, hit characteristic hot words, transaction information of the last year, a client reported report, a transaction opponent suspicious report, five transaction opponent collection accumulation amount, five transaction opponent payment accumulation amount, transaction channel transaction condition and the like, so that user input data are formed.
When the operator performs the study, judgment and analysis of each module, the operator needs to comprehensively use various information and analysis methods by combining the specific situation and business background of the customer to perform deep thinking and judgment. Meanwhile, operators are required to continuously adjust and optimize the analysis methods and indexes of each module according to the feedback and actual demands of clients so as to improve the accuracy and reliability of the input data of users.
In the specific implementation, after user input data corresponding to each data feature type and target client acquisition data are determined according to the preset data feature types of the client financial transaction risk aiming at the target client, an analysis report of the target client under different data feature types is generated based on a large language model according to the user input data corresponding to different data feature types and the target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode.
In the embodiment, based on a large language model technology, the system data collected and arranged by the system and the operation behavior of operators are obtained through operation behavior analysis of man-machine interaction, and a set of tools capable of automatically and effectively generating bank funds transfer risk suspicious investigation reports are generated by utilizing AI generation, so that the writing efficiency of the bank funds transfer risk suspicious investigation reports is improved, and meanwhile, the defects of templatizing and standardization of the traditional AI generation reports are overcome.
In the specific implementation, after user input data corresponding to each data feature type and target client acquisition data are determined according to the preset data feature types of the client financial transaction risk aiming at the target client, an analysis report of the target client under different data feature types is generated based on a large language model according to the user input data corresponding to different data feature types and the target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode.
For example, in the generation process of the bank funds transfer risk suspicious screening survey report, a corresponding analysis report can be generated by using a large language model according to the transaction data, the funds flow direction, the account information and other data feature types of the bank clients. The process can be realized by man-machine interaction, an operator can input related data information through a system interface, and the system can automatically generate a corresponding investigation report.
In addition, by collecting and sorting system data, the result analysis can be performed in combination with the operation behaviors of operators. Thus, a set of tools capable of automatically and effectively generating bank funds transfer risk suspicious screening survey reports can be generated by using AI. Compared with the traditional manual report writing, the automatic report generating mode can improve efficiency and avoid the defects of templating and standardization. By the method, the bank can monitor the fund transfer condition of the customer better, discover suspicious transactions in time and take corresponding measures to control risks. Meanwhile, the automatic report generation mode can help banks to better know client demands and market changes, and provides powerful support for future business development.
In one embodiment, further comprising:
the large language model is built as follows:
constructing a training sample library according to the historical report data; the history report data comprises user input data corresponding to different clients and history record data of target client acquisition data;
and performing instruction fine adjustment and weight fine adjustment on the large language model by using the training sample library to obtain a trained large language model.
In the above method, to further enhance the performance of the large language model, the following steps may be performed:
firstly, preprocessing data in a training sample library, including data cleaning, data normalization, data conversion and the like, so as to eliminate noise and abnormal values in the data and ensure the quality and reliability of the data.
Next, the large language model is trained using the preprocessed training sample library. During the training process, a variety of different optimization algorithms, such as random gradient descent (SGD), adam, etc., may be employed to find optimal model parameters.
After training is completed, the large language model can be evaluated and debugged using the test set. The performance of the large language model can be further improved by adjusting the model parameters and the optimization algorithm.
And finally, deploying and applying the trained large language model. In practical application, the performance of the large language model can be continuously optimized through interaction with a user so as to better meet the requirements of the user.
It should be noted that in building a large language model, the structure and parameters of the model should be determined according to specific application scenarios and requirements. Meanwhile, in order to protect the privacy and safety of users, the data needs to be subjected to desensitization treatment, encryption protection and other measures.
In the above embodiment, the process of building a large language model further includes the steps of:
1. and (3) data collection: user input data corresponding to different clients and historical record data of target client collected data are collected from the historical report data. Such data may include personal information of the customer, purchase records, search history, feedback comments, and the like.
2. Data preprocessing: the collected data is cleaned, sorted and standardized to ensure accuracy and consistency of the data.
3. Building a training sample library: and constructing a training sample library according to the preprocessed data. Each sample includes user input data and historical data for the target customer's collected data.
4. Model training: and training the large language model by using a training sample library, and enabling the model to better understand and generate natural language through repeated iteration and optimization.
5. Model evaluation and adjustment: in the training process, the performance and effect of the model need to be evaluated regularly, and the model is adjusted and optimized according to the evaluation result. This may include adjusting parameters of the model, increasing or decreasing the number of layers, changing the activation function, etc.
6. Instruction fine tuning and weight fine tuning: instruction fine tuning and weight fine tuning are also required during training of large language models. Instruction fine tuning refers to making small adjustments to a model to better understand and generate natural language. The fine-tuning of the weight refers to adjusting the weight of each word or grammar in the model according to the occurrence frequency and importance of each word or grammar.
7. Model deployment: the trained large language model may be deployed into a practical system to help customers understand and generate natural language better.
In the implementation, first, aiming at a target client, according to the preset data feature types of the financial transaction risk of the client, user input data corresponding to each data feature type and target client acquisition data are determined. These data characteristic types may include aspects of the customer's transaction behavior, account information, credit status, and the like.
And then, based on the large language model, generating an analysis report of the target client under different data feature types according to the user input data corresponding to the different data feature types and the target client acquisition data. These reports are expressed in natural language, while intelligent reasoning and judgment are performed to provide a more accurate and comprehensive risk assessment.
In order to better realize the process, the system data collected and arranged by the system are combined with the operation behaviors of operators based on a large language model technology through operation behavior analysis of man-machine interaction, and a set of tools capable of automatically and effectively generating bank funds transfer risk suspicious screening investigation reports are established by utilizing AI generation. Therefore, the writing efficiency of bank funds transfer risk suspicious screening investigation reports can be improved, and the defects of templatization and standardization of the traditional AI generation report can be avoided.
When training a large language model, a training sample library needs to be built according to a certain mode. This training sample library should include user input data corresponding to the different clients and historical record data of the target client's collected data. Then, the training sample library is utilized to conduct instruction fine adjustment and weight fine adjustment on the large language model, and the trained large language model is obtained. The large language model can be better adapted to actual application scenes, and accuracy and efficiency of risk assessment are improved.
In one embodiment, further comprising:
and sequencing and visually displaying the analysis reports corresponding to each data feature type according to the sequence from high to low of the financial transaction risk level of each data feature type in the acquired user input data and/or according to the sequence from front to back of the data moment of each data feature type in the user input data, so as to obtain a report display interface.
In the report presentation interface, each analysis report is ordered and visually presented according to the order of the financial transaction risk level of each data feature type in the user input data from high to low and/or according to the order of the data time of each data feature type in the user input data from first to last. In this way, the user can more clearly understand the content and focus of each analysis report to better make investment decisions.
Meanwhile, the report display interface also supports various interactive operations, and a user can check analysis reports of different data characteristic types in a clicking, sliding and other modes and compare and analyze the analysis reports. In addition, the report presentation interface also supports various data visualization modes, such as bar charts, line charts, pie charts, and the like, so as to better present data and provide more visual investment decision references.
In a word, the analysis reports corresponding to each data feature type are sequenced and visually displayed according to the financial transaction risk level and the data time sequence, and various interactive operation and data visualization modes are supported, so that the report display interface can help a user to more comprehensively know investment risks and opportunities, and accuracy and efficiency of investment decision are improved.
In one embodiment, the report presentation interface carries jump links of analysis reports corresponding to different data feature types; the jump link is used for jumping and expanding the analysis report corresponding to the data characteristic type. A step of
The report presentation interface also carries links to comparison reports for different data feature types for jumping analysis reports comparing the different data feature types.
In this way, a user may conveniently obtain detailed analysis reports corresponding to different data feature types and be able to quickly compare the analysis reports of the different data feature types to better understand differences and similarities between the data feature types. Meanwhile, the jump link is used, so that the user can conveniently develop the analysis report corresponding to the data characteristic type, and the use experience of the user is further improved.
In addition, the report display interface also carries links of data distribution reports aiming at different data characteristic types, and is used for jumping to check the data distribution conditions of the corresponding data characteristic types. In this way, the user can intuitively understand the data distribution condition of the data feature types, so as to better understand the distribution characteristics and rules of the data feature types.
In summary, the report presentation interface enables a user to conveniently acquire and analyze differences and similarities between data feature types, while also intuitively knowing the data distribution of the data feature types, by providing skip links for analysis reports and links for comparison reports corresponding to different data feature types, and links for data distribution reports. The design thought can greatly improve the use efficiency and convenience of users.
In one embodiment, the large language model may also automatically determine and generate a corresponding analysis report according to different data feature types. For example, when there are multiple different financial transaction risk levels in the input user data, the large language model may automatically identify and generate analysis reports corresponding to the different risk levels, thereby helping the user to more fully understand the financial transaction risk situation of the target customer.
In addition, the large language model can also sequence and visually display the analysis reports corresponding to each data feature type according to the sequence from the beginning to the end of the data moment of each data feature type in the user input data. For example, when a plurality of different financial transaction risk levels exist, the large language model can rank the analysis report corresponding to the earliest data feature type according to the sequence of the data time, so that a user can more conveniently check and analyze the financial transaction risk condition of the target client.
Meanwhile, the report display interface carries jump links of analysis reports corresponding to different data characteristic types. For example, when a user wants to view an analysis report of a specific data feature type, the user can directly skip and spread the analysis report of the data feature type by clicking on a corresponding skip link, so that the user can more quickly and accurately view and analyze financial transaction risk situations of a target client.
In the embodiment, a suspected monitoring view of the bank funds transfer risk can be extracted, and an operator can be intuitively assisted to operate according to the risk situation; the customer risk view can be extracted, an operator can be intuitively assisted to operate according to the risk situation, and the method is more intuitive and instructive.
In addition, a risk monitoring module can be constructed for real-time monitoring and early warning of bank funds transfer risks. The module can generate a risk monitoring report by utilizing AI based on a large language model technology, and timely discover and early warn potential funds transfer risks.
In an embodiment, the risk monitoring module may comprise the steps of:
1. and (3) data acquisition: relevant data of bank funds transfer is collected through means of a system interface or a crawler program and the like, including but not limited to transaction amount, transaction time, transaction counter-party information, transaction flow and the like.
2. Data preprocessing: the collected data is subjected to preprocessing operations such as cleaning, de-duplication, standardization and the like so as to ensure the quality and usability of the data.
3. Risk assessment: based on the large language model technology, the preprocessed data is input into a risk assessment model, and an assessment report of the bank funds transfer risk is generated. The report may include information on risk level, risk type, risk early warning, etc.
4. Risk early warning: and sending out early warning signals in time for the funds transfer with potential risks according to the risk assessment report, and reminding operators to pay attention to and further process. The early warning signal can include short message notification, mail notification, system popup window and other modes.
5. Report generation: and generating a risk monitoring report according to the risk assessment result and the early warning condition. The report may include information such as detailed analysis of the risk of funds transfer, pre-warning conditions, countermeasures, etc., for reference and use by the operator.
By establishing the risk monitoring module, the real-time monitoring and early warning of the bank funds transfer risk can be realized, and the bank funds safety and risk management level can be improved. Meanwhile, by combining a large language model technology and man-machine interaction operation behavior analysis, the accuracy and efficiency of risk monitoring can be improved, and powerful support is provided for robust operation of banks.
Two specific examples are given below to illustrate specific applications of the method of the invention.
First embodiment:
the bank funds transfer risk suspicious case screening and checking is one of the core contents of the bank funds transfer risk work, and the main workflow comprises the following steps: the method comprises the steps that (1) a system generates a suspicious early warning of the system according to a suspicious model; (2) By manually analyzing the system early warning condition and combining with the analysis of a plurality of dimensions of the customer condition, whether suspicious cases relate to suspicious needs to be submitted to people is manually confirmed, and in the process, a more detailed bank funds transfer risk suspicious screening survey report needs to be written for eliminating suspicious cases and reporting suspicious cases. In the process, more manpower is required to carry out analysis and writing, and meanwhile, the condition that the investigation report writing quality is poor and irregular exists.
The first embodiment is based on a large language model technology, and through operation behavior analysis of man-machine interaction, the system data collected and arranged by the system are combined with operation behavior of operators, a set of tools capable of automatically and effectively generating bank funds transfer risk suspicious investigation reports are generated by utilizing AI generation, so that the writing efficiency of the bank funds transfer risk suspicious investigation reports is improved, and meanwhile, the defects of templatizing and standardization of traditional AI generation reports are overcome. The basic working steps are as follows:
in a first specific embodiment, as shown in fig. 2, the following steps may be included:
1. the operator enters a bank funds transfer risk suspicious monitoring view to check the overall risk condition of suspicious transactions, and a specific page is shown in fig. 3;
2. the operator clicks the sub-item analysis page which enters the client name and risk rating condition, the client basic information, the due investigation, the hit characteristic hot word, the last year transaction information, the client reported report, the suspicious report of the transaction opponent, the first five transaction opponent collection accumulated amount, the first five transaction opponent payment accumulated amount and the transaction condition of the sub-transaction channel to conduct the research and judgment of each module, and the specific page is shown in figure 4;
3. The system records the operation sequence of operators and the input research and judgment conclusion and analysis content of each module, extracts and forms manual analysis data to form a table, as shown in table 1, wherein the table 1 is firstly ordered according to the suspicious degree, and if the suspicious degree is consistent, the table 1 is ordered from small to large according to the operation time;
TABLE 1
4. The system data collected by the bank funds transfer risk system data layer generates a system client rating report of each module of client name and risk rating condition, client basic information, due investigation, hit characteristic hotword, last year transaction information, client reported report, transaction opponent suspicious report, five transaction opponent collection accumulation amount, five transaction opponent payment accumulation amount and transaction channel transaction condition through a large language model
5. And (3) sequencing the final report according to the sequence of the step 3, and simultaneously inserting contents input by an operator into the 'module research analysis condition' reserved for each module to form the final report.
6. In the process of composing the final report, the system automatically generates a research analysis conclusion of each module according to research analysis content input by an operator, and generates a final bank funds transfer risk suspicious screening investigation report in a summarizing way.
7. By adopting the mode, the writing efficiency of bank funds transfer risk suspicious screening investigation reports can be improved, and the defects of templatizing and standardization of the traditional AI generation report can be avoided.
8. In the process of writing the final report, the system can also automatically check and audit the analysis conclusion of the research judgment of each module so as to ensure the quality and accuracy of the report.
9. By adopting the mode, the time and energy for manually writing the report can be reduced, the working efficiency is improved, and the quality and accuracy of the report can be improved.
10. In practicing the present embodiment, automated tools or manual methods may be employed.
11. In the process of adopting the automation tool, data acquisition and arrangement can be carried out through an API interface or a webpage application mode.
12. In the process of adopting a manual mode, data can be acquired and arranged in a mode of manually operating a bank funds transfer risk monitoring view and the like.
13. In practicing this particular embodiment, care is taken to protect the privacy and data security of the customer.
14. In processes employing automated tools, data needs to be encrypted and securely stored.
15. In the process of adopting a manual mode, training and management on operators are required to be enhanced, and accuracy and safety of data are ensured.
Second embodiment:
the customer bank funds transfer risk rating is one of the core contents of the bank funds transfer risk job, and the main workflow thereof comprises the following steps: (1) The system generates a system initial evaluation result according to the client rating model; (2) The system rating results are manually confirmed by analyzing the system rating results and combining multiple dimensions of the customer situation, and in the process, more detailed rating survey reports are required to be written for rating tasks related to customers with higher risks. In the process, more manpower is required to carry out analysis and writing, and meanwhile, the condition that the investigation report writing quality is poor and irregular exists.
The second embodiment is based on a large language model technology, and the system data collected and arranged by the system and the operation behavior of operators are obtained through operation behavior analysis of man-machine interaction, and a set of tools capable of automatically and effectively generating bank funds transfer risk suspicious investigation reports are generated by utilizing AI generation, so that the writing efficiency of the bank funds transfer risk suspicious investigation reports is improved, and meanwhile, the defects of templatizing and standardization of the traditional AI generation reports are overcome.
In a second specific embodiment, as shown in fig. 5, the following steps may be included:
1. the operator enters a customer bank funds transfer risk rating view to check the overall risk condition of the customer, and a specific page is shown in fig. 6;
2. the operator clicks the sub-term analysis page entering the client name, the risk tag, the due-job investigation, the account information, the product information, the recent suspicious report and the recent large report to conduct the research, the judgment and the analysis of each module, and the specific page is shown in fig. 4;
3. the system records the operation sequence of operators and the input research and judgment conclusion and analysis content of each module, extracts and forms manual analysis data to form a table, as shown in table 2, wherein the table 2 is firstly ordered according to the suspicious degree, and if the suspicious degree is consistent, the table is ordered from small to large according to the operation time;
TABLE 2
4. The system data collected by the bank funds transfer risk system data layer generates a system client rating report of each module of 'client name', 'risk tag', 'due job investigation', 'account information', 'product information', 'recent suspicious report', 'recent large report' through a large language model;
5. And (3) sequencing the final report according to the sequence of the step 3, and simultaneously inserting contents input by an operator into the 'module research analysis condition' reserved for each module to form the final report.
In a second embodiment, in the analysis process of the bank funds transfer risk rating system, not only is the system primary evaluation result generated by the customer bank funds transfer risk rating model considered, but also a plurality of dimensional analyses of the customer situation by the operator are fully combined. The analysis method combining artificial intelligence and human expert can evaluate the risk level of the customer more accurately.
During operation, an operator may view the overall risk of the customer by entering a customer banking funds transfer risk rating view. This view may display detailed information of the customer, including customer name, risk tags, due diligence, account information, product information, and recent suspicious reports, among others. Through the information, operators can more comprehensively know the condition of clients and more accurately evaluate the risk level of the clients.
The operator can click the sub-item analysis page entering each module to conduct research, judgment and analysis. For example, in a "customer name" module, an operator may view details of the customer, including the customer's name, address, contact, etc.; in the 'risk tag' module, an operator can primarily judge the risk level of a client according to the risk tag given by the system; in the "due diligence" module, operators can check the due diligence report of the clients and know the information of the clients such as the business condition, the financial condition and the like.
During the analysis, the system records the order of operation of the operators and the content of the inputs and refines and forms manual analysis data. These data will be used to generate the final customer banking funds transfer risk rating report.
The final report will be ordered according to the degree of suspicion and in order of time of operation from big to small. The analysis of each module will be displayed according to the content entered by the operator in the corresponding module. The design can make the report more personalized and targeted, and simultaneously avoid the defects of templatization and standardization of the traditional AI generation report.
By the method in the second embodiment, the bank can evaluate the risk level of the customer more effectively and provide the customer with a better quality financial service. Meanwhile, the method can also improve the writing efficiency of the bank funds transfer risk suspicious screening investigation report and ensure the quality and consistency of the report.
Third embodiment:
the third embodiment is based on a further optimization of the second embodiment, enabling automated risk rating report generation and auditing by employing large language model techniques.
In a third embodiment, after the operator enters the customer banking funds transfer risk rating view, the operator may click on the "generate report" button, and the system automatically generates the customer banking funds transfer risk rating report. The report includes not only basic information of the customer, risk tags, due diligence, account information, product information, recent suspicious reports, etc., but also increases the assessment of the overall risk situation of the customer and the analysis conclusion and advice for each module.
The third embodiment also adds an audit link in order to ensure the quality and accuracy of the report. The operator can submit the generated report to the superior auditor for auditing. The auditor can audit the report by entering a customer bank funds transfer risk rating audit view, and can add audit comments and modification comments. If the audit is not passed, the operator needs to revise and resubmit the report according to the audit opinion.
The third embodiment has the advantages of high automation degree, high-quality customer bank funds transfer risk rating report generation speed, and reduced time and cost for manual writing and auditing. In addition, by adopting a large language model technology, the generated report is more natural and smooth, and the readability and the intelligibility are improved.
In a word, the third embodiment automatically generates and audits the customer bank funds transfer risk rating report, improves the working efficiency and quality, reduces the cost and error rate, and provides more effective support for the bank funds transfer risk management work.
Fourth embodiment:
the fourth embodiment further introduces a machine learning technique to continuously optimize the customer banking funds transfer risk rating model.
In a fourth embodiment, the system collects a large amount of customer banking funds-transfer risk rating data, including historical analysis data of operators, customer base information, account information, product information, recent suspicious reports, and the like. The system would then use this data to train a machine learning model that would enable it to automatically learn and identify patterns and features of customer banking funds transfer risk.
When an operator performs the risk rating of the customer bank funds transfer, the system automatically analyzes the data of the customer according to the learned model and generates a corresponding risk rating report. Meanwhile, the system can continuously update and optimize the model according to new data so as to improve the accuracy and efficiency of risk rating.
An advantage of the fourth embodiment is that the system can automatically learn and optimize through machine learning techniques, constantly improving the accuracy and efficiency of risk ratings. Meanwhile, machine learning can also process a large amount of data, valuable information and modes are extracted from the data, and comprehensive support is provided for money transfer risk management of banks.
In addition, the fourth embodiment may also implement an automated early warning function. The system can early warn the fund transfer risk of the customer bank according to the learned model, discover potential risks and problems in time, and provide timely countermeasures for the bank.
In a word, the fourth embodiment realizes the automatic learning and optimization of the customer bank funds transfer risk rating model by introducing a machine learning technology, improves the working efficiency and quality, reduces the cost and error rate, and provides more intelligent and efficient support for the funds transfer risk management work of the bank.
The specific embodiment provided by the invention has the following advantages: the risk monitoring mechanism is optimized, and the accuracy and timeliness of risk identification are improved; the efficiency of operators is improved; the method combines creatively a large language model and man-machine interaction mode, and serves clients more intelligently; the risk control capability of the bank is improved.
1. The risk monitoring mechanism is optimized, and the accuracy and timeliness of risk identification are improved. By introducing a large language model and a man-machine interaction mode, the AI can better understand natural language and extract valuable information from a large amount of data, thereby improving accuracy and timeliness of risk identification. This not only helps the bank to better manage risk, but also better protects the customer's interests.
2. The efficiency of the operator is improved. By refining the bank funds transfer risk suspicious monitoring view and the customer risk view, operators can more intuitively understand the risk situation, thereby making decisions more quickly and accurately. This not only improves operator efficiency, but also reduces errors due to human factors.
3. The method combines a large language model with a man-machine interaction mode innovatively and serves clients more intelligently. By introducing a large language model and man-machine interaction mode, the AI can better understand the needs and problems of the customer and can automatically generate a more personalized and customized service scheme. This can not only improve customer satisfaction but also improve bank competitiveness.
4. The risk control capability of the bank is improved. By introducing a large language model and a man-machine interaction mode, the AI can better monitor and identify risks and can automatically generate a more personalized and customized risk control scheme. The risk control capability of the bank can be improved, and the business expansion capability of the bank can also be improved.
Of course, it is to be understood that other variations of the above detailed procedures are also possible, and all related variations should fall within the protection scope of the present invention.
In the embodiment of the invention, aiming at a target client, user input data and target client acquisition data corresponding to each data feature type are determined according to the preset data feature type of the client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user; based on a large language model, generating an analysis report of a target client under different data characteristic types according to user input data corresponding to the different data characteristic types and target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data and generating an intelligent report expressed in a natural language mode, so that the client financial transaction risk report can be automatically established and generated based on the large language model technology by combining the user input data and the target client collected data, the generation efficiency of the client financial transaction risk report is improved, the defects of templatization and standardization of the traditional AI generation report are overcome, and the flexibility of generating the client financial transaction risk report is improved; by combining the user input data, the manual judgment of the financial transaction risk level in the report is also realized, the accuracy of the screening judgment of the suspicious case of the customer financial transaction risk by the auxiliary staff is improved, and the user experience is improved.
In the embodiment of the invention, aiming at a target client, user input data and target client acquisition data corresponding to each data feature type are determined according to the preset data feature type of the client financial transaction risk. The user input data is used for representing financial transaction risk levels which are input by the user and describe the existence of target clients in corresponding data feature types, and the target client acquisition data is used for representing feature parameters corresponding to different data feature types in historical financial transaction data of the user.
Based on the large language model, according to user input data corresponding to different data feature types and target client collected data, an analysis report of the target client under the different data feature types is generated. The large language model is used for reasoning the user input data and the target customer collected data and generating an intelligent report expressed in a natural language mode, so that the customer financial transaction risk report can be automatically established and generated based on the large language model technology and combining the user input data and the target customer collected data.
The method and the system have the advantages that the template and standard disadvantages of the traditional AI generation report are avoided while the generation efficiency of the client financial transaction risk report is improved, and the flexibility of generating the client financial transaction risk report is improved. By combining the user input data, the manual judgment of the financial transaction risk level in the report is also realized, the accuracy of the screening judgment of the suspicious case of the customer financial transaction risk by the auxiliary staff is improved, and the user experience is improved.
In addition, the embodiment of the invention also provides a method for generating the customer financial transaction risk report, which comprises the following steps: acquiring user input data and target client acquisition data; inputting user input data and target client collected data into a large language model for reasoning; generating an intelligent report expressed in a natural language mode according to the reasoning result; the intelligent report is output as a customer financial transaction risk report.
In the embodiment of the invention, the client financial transaction risk report is automatically established and generated through a large language model technology, so that the defects of templatization and standardization of the traditional AI generation report are avoided, and the flexibility of generating the client financial transaction risk report is improved. Meanwhile, by combining the user input data, the manual judgment of the financial transaction risk level in the report is realized, the accuracy of the screening judgment of the suspicious case of the customer financial transaction risk by the auxiliary staff is improved, and the user experience is improved.
The embodiment of the invention also provides a client financial transaction risk report generating device, which is expressed in the following embodiment. Because the principle of the device for solving the problem is similar to that of the client financial transaction risk report generation method, the implementation of the device can refer to the implementation of the client financial transaction risk report generation method, and the repetition is omitted.
The embodiment of the invention also provides a device for generating the customer financial transaction risk report, which is used for improving the efficiency and the flexibility of generating the customer financial transaction risk report and improving the user experience, as shown in fig. 7, and comprises the following steps:
the data acquisition module 701 is configured to determine, for a target client, user input data corresponding to each data feature type and target client acquisition data according to a preset data feature type of a client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user;
the report generating module 702 is configured to generate an analysis report for the target client under different data feature types according to the user input data corresponding to the different data feature types and the target client collected data based on the large language model; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode.
In one embodiment, further comprising:
The large language model building module is used for:
the large language model is built as follows:
constructing a training sample library according to the historical report data; the history report data comprises user input data corresponding to different clients and history record data of target client acquisition data;
and performing instruction fine adjustment and weight fine adjustment on the large language model by using the training sample library to obtain a trained large language model.
In one embodiment, the data feature types include: the method comprises the steps of one or any combination of a client name, a risk rating condition type, a client basic information data type, a client due investigation related data type, a hit feature hotword data type, a client last year transaction information data feature, a client reported report data feature, a transaction opponent risk data, a transaction opponent collection accumulated amount data feature, a transaction opponent payment accumulated amount data feature and a' sub-transaction channel data feature.
In one embodiment, further comprising:
a report presentation interface generation interface for:
and sequencing and visually displaying the analysis reports corresponding to each data feature type according to the sequence from high to low of the financial transaction risk level of each data feature type in the acquired user input data and/or according to the sequence from front to back of the data moment of each data feature type in the user input data, so as to obtain a report display interface.
In one embodiment, the report presentation interface carries jump links of analysis reports corresponding to different data feature types; the jump link is used for jumping and expanding the analysis report corresponding to the data characteristic type.
In one embodiment, the data acquisition module 701 further comprises: and the data screening module is used for screening the historical financial transaction data of the target clients according to preset data screening conditions to obtain target client acquisition data meeting the conditions.
The data screening conditions may include, but are not limited to: among the historical financial transaction data of the target customer, transaction data with transaction amount greater than the set threshold value, or transaction data with transaction time points within the set time range, and the like.
By setting the data screening conditions, the target customer acquisition data meeting the requirements can be acquired more accurately, and further the analysis accuracy of the financial transaction risk of the target customer is improved.
In one embodiment, the user input data may include, but is not limited to: descriptive information of the target client by the user, subjective evaluation information of financial transaction risk of the target client by the user, and the like; the target customer collected data may include, but is not limited to: historical financial transaction data such as transaction amount, transaction time, transaction counter-parties, transaction channels, and the like of the target customer.
In one embodiment, the large language model may also be used to deep learn user input data and target customer collected data, predict and infer new unlabeled data using unsupervised learning, and generate intelligent reports expressed in natural language.
In one embodiment, the large language model may be further used to perform cluster analysis on the user input data and the target client collected data, discover and analyze risk level distribution conditions under different data feature types, and further generate an analysis report for the target client under different data feature types.
In one embodiment, the report generation module 702 may also be configured to: and generating a corresponding visual chart, such as a pie chart, a histogram, a line chart and the like, according to the analysis report under different data characteristic types, so that a user can more intuitively know the financial transaction risk condition of the target client under different data characteristic types.
The embodiment of the invention provides a computer device for realizing all or part of contents in the client financial transaction risk report generation method, which specifically comprises the following contents:
A processor (processor), a memory (memory), a communication interface (Communications Interface), and a bus; the processor, the memory and the communication interface complete communication with each other through the bus; the communication interface is used for realizing information transmission between related devices; the computer device may be a desktop computer, a tablet computer, a mobile terminal, or the like, and the embodiment is not limited thereto. In this embodiment, the computer device may be implemented with reference to an embodiment for implementing the method for generating a risk report of a customer financial transaction and an embodiment for implementing the apparatus for generating a risk report of a customer financial transaction, and the contents thereof are incorporated herein, and the repetition is omitted.
Fig. 8 is a schematic block diagram of a system configuration of a computer device 1000 according to an embodiment of the present application. As shown in fig. 8, the computer device 1000 may include a central processor 1001 and a memory 1002; the memory 1002 is coupled to the central processor 1001. Notably, this fig. 8 is exemplary; other types of structures may also be used in addition to or in place of the structures to implement telecommunications functions or other functions.
In one embodiment, the customer financial transaction risk report generation functionality may be integrated into the central processor 1001. The central processor 1001 may be configured to control, among other things, the following:
Aiming at a target client, determining user input data corresponding to each data feature type and target client acquisition data according to the preset data feature type of the client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user;
based on a large language model, generating an analysis report of a target client under different data characteristic types according to user input data corresponding to the different data characteristic types and target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode.
In another embodiment, the customer financial transaction risk report generating device may be configured separately from the central processor 1001, for example, the customer financial transaction risk report generating device may be configured as a chip connected to the central processor 1001, and the customer financial transaction risk report generating function is implemented under the control of the central processor.
As shown in fig. 8, the computer device 1000 may further include: a communication module 1003, an input unit 1004, an audio processor 1005, a display 1006, a power supply 1007. It is noted that the computer device 1000 need not include all of the components shown in FIG. 8; in addition, the computer device 1000 may further include components not shown in fig. 8, to which reference is made to the related art.
As shown in fig. 8, the central processor 1001, sometimes also referred to as a controller or operational control, may include a microprocessor or other processor device and/or logic device, and the central processor 1001 receives input and controls the operation of the various components of the computer device 1000.
The memory 1002 may be, for example, one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. The information about failure may be stored, and a program for executing the information may be stored. And the central processor 1001 can execute the program stored in the memory 1002 to realize information storage or processing, and the like.
The input unit 1004 provides input to the central processor 1001. The input unit 1004 is, for example, a key or a touch input device. The power supply 1007 is used to provide power to the computer device 1000. The display 1006 is used for displaying display objects such as images and characters. The display may be, for example, but not limited to, an LCD display.
The memory 1002 may be a solid state memory such as Read Only Memory (ROM), random Access Memory (RAM), SIM card, and the like. But also a memory which holds information even when powered down, can be selectively erased and provided with further data, an example of which is sometimes referred to as EPROM or the like. Memory 1002 may also be some other type of device. Memory 1002 includes a buffer memory 1021 (sometimes referred to as a buffer). The memory 1002 may include an application/function storage 1022, the application/function storage 1022 for storing application programs and function programs or for executing a flow of operations of the computer apparatus 1000 by the central processor 1001.
The memory 1002 may also include a data store 1023, the data store 1023 for storing data such as contacts, digital data, pictures, sounds, and/or any other data used by a computer device. The driver store 1024 of the memory 1002 can include various drivers for the computer device for communication functions and/or for performing other functions of the computer device (e.g., messaging applications, address book applications, etc.).
The communication module 1003 is a transmitter/receiver 1003 that transmits and receives signals via an antenna 1008. A communication module (transmitter/receiver) 1003 is coupled to the central processor 1001 to provide an input signal and receive an output signal, which may be the same as in the case of a conventional mobile communication terminal.
Based on different communication technologies, a plurality of communication modules 1003, such as a cellular network module, a bluetooth module, and/or a wireless lan module, etc., may be provided in the same computer device. The communication module (transmitter/receiver) 1003 is also coupled to a speaker 1009 and a microphone 1010 via an audio processor 1005 to provide audio output via the speaker 1009 and to receive audio input from the microphone 1010 to implement usual telecommunications functionality. The audio processor 1005 may include any suitable buffers, decoders, amplifiers and so forth. In addition, an audio processor 1005 is also coupled to the central processor 1001 so that sound can be recorded locally through the microphone 1010 and so that sound stored locally can be played through the speaker 1009.
The embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the client financial transaction risk report generation method when being executed by a processor.
The embodiment of the invention also provides a computer program product, which comprises a computer program, wherein the computer program is executed by a processor to realize the client financial transaction risk report generation method.
In the embodiment of the invention, aiming at a target client, user input data and target client acquisition data corresponding to each data feature type are determined according to the preset data feature type of the client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user; based on a large language model, generating an analysis report of a target client under different data characteristic types according to user input data corresponding to the different data characteristic types and target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data and generating an intelligent report expressed in a natural language mode, so that the client financial transaction risk report can be automatically established and generated based on the large language model technology by combining the user input data and the target client collected data, the generation efficiency of the client financial transaction risk report is improved, the defects of templatization and standardization of the traditional AI generation report are overcome, and the flexibility of generating the client financial transaction risk report is improved; by combining the user input data, the manual judgment of the financial transaction risk level in the report is also realized, the accuracy of the screening judgment of the suspicious case of the customer financial transaction risk by the auxiliary staff is improved, and the user experience is improved.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (13)
1. A method for generating a customer financial transaction risk report, comprising:
aiming at a target client, determining user input data corresponding to each data feature type and target client acquisition data according to the preset data feature type of the client financial transaction risk; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user;
based on a large language model, generating an analysis report of a target client under different data characteristic types according to user input data corresponding to the different data characteristic types and target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode.
2. The method as recited in claim 1, further comprising:
the large language model is built as follows:
constructing a training sample library according to the historical report data; the history report data comprises user input data corresponding to different clients and history record data of target client acquisition data;
And performing instruction fine adjustment and weight fine adjustment on the large language model by using the training sample library to obtain a trained large language model.
3. The method of claim 1, wherein the data characteristic type comprises: the method comprises the steps of one or any combination of a client name, a risk rating condition type, a client basic information data type, a client due investigation related data type, a hit feature hotword data type, a client last year transaction information data feature, a client reported report data feature, a transaction opponent risk data, a transaction opponent collection accumulated amount data feature, a transaction opponent payment accumulated amount data feature and a' sub-transaction channel data feature.
4. The method as recited in claim 1, further comprising:
and sequencing and visually displaying the analysis reports corresponding to each data feature type according to the sequence from high to low of the financial transaction risk level of each data feature type in the acquired user input data and/or according to the sequence from front to back of the data moment of each data feature type in the user input data, so as to obtain a report display interface.
5. The method of claim 4, wherein the report presentation interface carries jump links for analysis reports corresponding to different data characteristic types; the jump link is used for jumping and expanding the analysis report corresponding to the data characteristic type.
6. A customer financial transaction risk report generating device, comprising:
the data acquisition module is used for determining user input data corresponding to each data characteristic type and target client acquisition data according to the preset data characteristic type of the client financial transaction risk aiming at the target client; the user input data is used for representing financial transaction risk levels of the description target clients input by the user and existing in the corresponding data feature types; the target client collected data are used for representing characteristic parameters corresponding to different data characteristic types in historical financial transaction data of a user;
the report generation module is used for generating an analysis report of the target client under different data characteristic types based on the large language model according to the user input data corresponding to the different data characteristic types and the target client acquisition data; the large language model is used for reasoning the user input data and the target client collected data, and generating an intelligent report expressed in a natural language mode.
7. The apparatus as recited in claim 6, further comprising:
the large language model building module is used for:
the large language model is built as follows:
Constructing a training sample library according to the historical report data; the history report data comprises user input data corresponding to different clients and history record data of target client acquisition data;
and performing instruction fine adjustment and weight fine adjustment on the large language model by using the training sample library to obtain a trained large language model.
8. The apparatus of claim 6, wherein the data characteristic type comprises: the method comprises the steps of one or any combination of a client name, a risk rating condition type, a client basic information data type, a client due investigation related data type, a hit feature hotword data type, a client last year transaction information data feature, a client reported report data feature, a transaction opponent risk data, a transaction opponent collection accumulated amount data feature, a transaction opponent payment accumulated amount data feature and a' sub-transaction channel data feature.
9. The apparatus as recited in claim 6, further comprising:
a report presentation interface generation interface for:
and sequencing and visually displaying the analysis reports corresponding to each data feature type according to the sequence from high to low of the financial transaction risk level of each data feature type in the acquired user input data and/or according to the sequence from front to back of the data moment of each data feature type in the user input data, so as to obtain a report display interface.
10. The apparatus of claim 9, wherein the report presentation interface carries jump links for analysis reports corresponding to different data characteristic types; the jump link is used for jumping and expanding the analysis report corresponding to the data characteristic type.
11. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 5 when executing the computer program.
12. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, implements the method of any of claims 1 to 5.
13. A computer program product, characterized in that the computer program product comprises a computer program which, when executed by a processor, implements the method of any of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311636141.8A CN117635153A (en) | 2023-11-30 | 2023-11-30 | Client financial transaction risk report generation method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311636141.8A CN117635153A (en) | 2023-11-30 | 2023-11-30 | Client financial transaction risk report generation method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117635153A true CN117635153A (en) | 2024-03-01 |
Family
ID=90017857
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311636141.8A Pending CN117635153A (en) | 2023-11-30 | 2023-11-30 | Client financial transaction risk report generation method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117635153A (en) |
-
2023
- 2023-11-30 CN CN202311636141.8A patent/CN117635153A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20200210899A1 (en) | Machine learning model training method and device, and electronic device | |
US10176526B2 (en) | Processing system for data elements received via source inputs | |
US20230274351A1 (en) | Processing system to generate risk scores for electronic records | |
JP2005158069A (en) | System, method and computer product for detecting action pattern for financial soundness of business subject | |
CN111882420B (en) | Response rate generation method, marketing method, model training method and device | |
CN111275546A (en) | Financial client fraud risk identification method and device | |
KR102168198B1 (en) | Business default prediction system and operation method thereof | |
CN111738331A (en) | User classification method and device, computer-readable storage medium and electronic device | |
CN112598294A (en) | Method, device, machine readable medium and equipment for establishing scoring card model on line | |
CN115908022A (en) | Abnormal transaction risk early warning method and system based on network modeling | |
CN112801775A (en) | Client credit evaluation method and device | |
CN111179051A (en) | Financial target customer determination method and device and electronic equipment | |
CN109102396A (en) | A kind of user credit ranking method, computer equipment and readable medium | |
CN115409518A (en) | User transaction risk early warning method and device | |
CN111061948B (en) | User tag recommendation method and device, computer equipment and storage medium | |
KR102499181B1 (en) | Loan regular auditing system using artificia intellicence | |
US20100042446A1 (en) | Systems and methods for providing core property review | |
CN117196630A (en) | Transaction risk prediction method, device, terminal equipment and storage medium | |
KR102499182B1 (en) | Loan regular auditing system using artificia intellicence | |
Chen et al. | Predicting a corporate financial crisis using letters to shareholders | |
CN117635153A (en) | Client financial transaction risk report generation method and device | |
CN113220447B (en) | Financial wind control system and method based on edge calculation | |
CN111951099A (en) | Credit card issuing model and application method thereof | |
CN114612239A (en) | Stock public opinion monitoring and wind control system based on algorithm, big data and artificial intelligence | |
CN114462742A (en) | Risk prompting method, device, equipment and computer storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |