Financial data processing method, device and equipment based on dynamic evolution mechanism

Abstract

Embodiments of the present application provide a financial data processing method, device and equipment based on dynamic evolution mechanism, the method comprises: receiving monitored financial data stream; determining the complexity index of the financial data stream; when the complexity index has a first relationship with a first preset threshold, determining the confidence of the processing result of the financial data stream by different preset models; dynamically adjusting the weight of each preset model by combining the confidence and the preset adjustment parameter, and the adjusted weight is positively correlated with the confidence of the corresponding preset model; reprocessing the financial data stream based on the preset model after weight adjustment.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of financial technology, and in particular to a financial data processing method, apparatus and equipment based on a dynamic evolution mechanism. Background Technology

[0002] The financial industry generates a vast amount of unstructured data, and with the expansion of transaction volume and business types, the types and quantity of unstructured data are also increasing. However, processing unstructured data is extremely challenging. First, there are hundreds of common types, such as research reports, announcements, policies, and public opinion news. Second, the content within each type is complex and varied, with significant differences depending on the subject, time, and type, such as authorization contracts and procurement contracts. Third, the volume is enormous, with daily growth reaching terabytes. Fourth, the value is high, as unstructured information contains much important, cross-referenced, and workflow information. Fifth, it demands high stability and reliability, as the sheer volume of types and content makes manual verification difficult. Summary of the Invention

[0003] To address the aforementioned technical problems, embodiments of the present invention provide a financial data processing method based on a dynamic evolution mechanism, comprising: Receive monitored financial data streams; Determine the complexity index of the financial data flow; When the complexity index has a first relationship with the first preset threshold, the confidence level of the processing results of the financial data stream by different preset models is determined; The weights of each preset model are dynamically adjusted based on the confidence level and preset adjustment parameters. The adjusted weights are positively correlated with the confidence level of the corresponding preset model. The financial data stream is reprocessed based on the pre-defined model after weight adjustment.

[0004] In one embodiment, determining the complexity index of the financial data stream includes: Determine the entropy value, format, and semantic density of financial data; The complexity index of the financial data is determined based on the entropy value, the format involved, and the semantic density.

[0005] In one embodiment, the step of dynamically adjusting the weights of each of the preset models by combining the confidence level and preset adjustment parameters includes: The weights of each preset model are dynamically adjusted based on the confidence level and preset temperature parameters.

[0006] In one embodiment, dynamically adjusting the weights of each preset model by combining the confidence level and preset temperature parameters includes: The confidence level of the financial data stream is converted into a data set of a specified dimension; Determine the quotient of each element in the data set with respect to the temperature parameter; Perform exponential operations on the quotients in the data set; The results of the exponential calculation are normalized to obtain new weights for each of the preset models.

[0007] In one embodiment, the method further includes: The confidence level of the result generated when the preset model reprocesses the financial data stream after weight adjustment is obtained, and this confidence level is defined as the new confidence level. At the same time, the confidence level of the result generated when the preset model processes the financial data stream without weight adjustment is defined as the original confidence level. The credibility of each preset model is verified based on the difference between the new confidence level and the original confidence level.

[0008] In one embodiment, the method further includes: When the confidence level exceeds the second preset threshold, the complexity index is updated based on the confidence level or the difference between the new confidence level and the original confidence level, or the weights of the preset model are updated again.

[0009] In one embodiment, the method further includes: The data distribution divergence of the financial data streams obtained at different times is detected; An alarm is triggered when the detection result exceeds the third preset threshold.

[0010] In one embodiment, the method further includes: When the complexity index has a second relationship with the first preset threshold, a preset emergency handling strategy is triggered; When the complexity index has a third relationship with the first preset threshold, the preset model is controlled to process the financial data stream based on the preset conventional processing strategy; The first relationship includes a complexity index exceeding the first preset threshold; the second relationship includes a complexity index within a range less than the first preset threshold and having a first difference from the first preset threshold; and the third relationship includes a complexity index within a range less than the first preset threshold and having a second difference from the first preset threshold, wherein the first difference is smaller than the second difference.

[0011] Another embodiment of the present invention also provides a financial data processing device based on a dynamic evolution mechanism, comprising: The receiving module is used to receive the monitored financial data stream; The first determining module is used to determine the complexity index of the financial data flow; The second determining module is used to determine the confidence level of the processing results of the financial data stream by different preset models when the complexity index has a first relationship with the first preset threshold. The adjustment module is used to dynamically adjust the weights of each preset model by combining the confidence level and preset adjustment parameters. The adjusted weights are positively correlated with the confidence level of the corresponding preset model. The processing module is used to reprocess the financial data stream based on the preset model after weight adjustment.

[0012] Another embodiment of the present invention also provides an electronic device, comprising: One or more processors; Memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the financial data processing method based on the dynamic evolution mechanism as described in any one of the above descriptions.

[0013] Based on the above, the beneficial effects of this invention include: monitoring financial data flows to determine their complexity index; then using the relationship between this complexity index and a preset threshold to infer whether model weights need adjustment; and promptly adjusting model weights when adjustment is required to ensure the processing accuracy and results of the financial data flows. This significantly improves data processing accuracy and enhances the collaborative efficiency of multiple models in scenarios with sudden changes in data complexity.

[0014] The advantages and other features of this application will be set forth in the following description. The objectives and other advantages of this application can be realized and obtained through the structures specifically pointed out in the written description and drawings.

[0015] The technical solution of this application will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the financial data processing method based on a dynamic evolution mechanism in an embodiment of the present invention.

[0018] Figure 2This is a flowchart illustrating the financial data processing method based on a dynamic evolution mechanism in an application embodiment of the present invention.

[0019] Figure 3 This is a flowchart illustrating a financial data processing method based on a dynamic evolution mechanism according to another embodiment of the present invention.

[0020] Figure 4 This is a structural block diagram of a financial data processing device based on a dynamic evolution mechanism in an embodiment of the present invention. Detailed Implementation

[0021] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but these are not intended to limit the scope of the invention.

[0022] It should be understood that various modifications can be made to the embodiments disclosed herein. Therefore, the following description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope of this disclosure will be apparent to those skilled in the art.

[0023] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present disclosure and, together with the general description of the disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the disclosure.

[0024] These and other features of the invention will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0025] It should also be understood that although the invention has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of the invention, which have the features described in the claims and are therefore all within the scope of protection defined herein.

[0026] The above and other aspects, features and advantages of this disclosure will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0027] Specific embodiments of the present disclosure are described thereafter with reference to the accompanying drawings; however, it should be understood that the disclosed embodiments are merely examples of the present disclosure and can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the present disclosure. Therefore, the specific structural and functional details disclosed herein are not intended to be limiting, but merely to serve as the basis and representative basis for the claims to teach those skilled in the art to use the present disclosure in a variety of substantially any suitable detailed structures.

[0028] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this disclosure.

[0029] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0030] To address the shortcomings of existing solutions, such as the inability to respond in real-time to changes in data complexity, leading to a significant decrease in the processing accuracy of financial data streams and poor performance in multi-model collaborative processing; the tendency for fixed model weights to result in cross-modal feature fusion efficiency below 65%; and the sharp drop in system processing efficiency when the complexity of financial data streams is high, such as with a large variety of data types and formats, these are some of the financial data processing problems that need to be addressed. Figure 1 As shown, this embodiment of the invention provides a financial data processing method based on a dynamic evolution mechanism, including: S1: Receives monitored financial data streams; S2: Determine the complexity index of the financial data flow; S3: When the complexity index has a first relationship with the first preset threshold, determine the confidence level of the processing results of the financial data stream by different preset models; S4: The weights of each preset model are dynamically adjusted based on the confidence level and preset adjustment parameters. The adjusted weights are positively correlated with the confidence level of the corresponding preset model. S5: Reprocess the financial data stream based on the preset model after weight adjustment.

[0031] In this embodiment, the system receives monitored financial data streams in real time. These financial data streams are multi-source data streams, such as data received from an Apache Kafka cluster (which may, but is not limited to, 16 nodes). The data formats involved in these financial data streams are variable, for example, they may involve several, dozens, or hundreds of financial data formats. After obtaining the financial data stream, the system automatically determines a complexity index that characterizes the data complexity. The higher the complexity index, the higher the data complexity. When determining the complexity index, the system preferably measures and calculates it from multiple different dimensions, including but not limited to the types of data formats involved. After determining the data complexity, the system compares the complexity index with a preset first threshold to determine the data relationship. When the data relationship is characterized as a primary relationship, the system determines the confidence level of the results generated by each preset model when collaboratively processing the financial data stream. The preset models may include, but are not limited to, optical character recognition models, large language models, and visual language models. Next, based on the confidence level and preset adjustment parameters, the weights of each preset model are dynamically adjusted. The adjusted weights are positively correlated with the confidence level of the corresponding preset model; that is, the higher the weight, the higher the model's confidence level, and vice versa. After adjusting the weights of each preset model, the system will reprocess the obtained financial data stream based on the preset models with adjusted weights, thereby improving the accuracy of the data processing results.

[0032] Based on the above, the beneficial effects of this embodiment include: monitoring financial data flows to determine their complexity index; then using the relationship between this complexity index and a preset threshold to infer whether model weights need adjustment; and promptly adjusting model weights when adjustment is required to ensure the processing accuracy and results of the financial data flows. This significantly improves data processing accuracy and enhances the collaborative efficiency of multiple models in scenarios with sudden changes in data complexity.

[0033] In one embodiment, determining the complexity index of the financial data stream includes: S201: Determine the entropy value, format, and semantic density of the financial data stream; S202: Determine the complexity index of the financial data stream based on the entropy value, the format involved, and the semantic density.

[0034] like Figure 2As shown in this embodiment, when determining the complexity index, the system first determines the entropy value of the financial data stream, the types of data formats involved, and the semantic density, such as the density of specified semantics. Then, based on the determined entropy value, the involved formats, and the semantic density, the system calculates and determines the complexity index of the financial data stream. For example, the complexity index of the financial data stream can be calculated and determined based on a preset formula, combining the entropy value, the involved formats, and the semantic density, using either summation or weighted summation methods; the specific method is not unique.

[0035] Furthermore, the step of dynamically adjusting the weights of each preset model by combining the confidence level and preset adjustment parameters includes: S401: Dynamically adjust the weights of each preset model based on the confidence level and preset temperature parameters.

[0036] In this embodiment, the preset temperature parameter T is used to adjust the scaling ratio of the input value, affecting the final weight distribution. A smaller T results in a larger weight for high confidence and a smaller weight for low confidence; a larger T results in a more even weight distribution.

[0037] Specifically, the dynamic adjustment of the weights of each preset model based on the confidence level and preset temperature parameters includes: S402: Convert the confidence level into a data set of a specified dimension; S403: Determine the quotient of each element in the data set with respect to the temperature parameter; S404: Perform an exponential operation on the quotients in the data set; S405: Normalize the results of the exponential calculation to obtain new weights for each of the preset models.

[0038] For example, the confidence scores of the corresponding financial data stream can be converted into a NumPy array, then each data element can be divided by the temperature parameter T, and finally an exponential operation can be performed on the entire array. This step is equivalent to performing a non-linear transformation on the input data, amplifying or reducing the differences between different confidence scores. After the exponential operation is completed, the system will normalize the result so that the sum of the weights is 1. The above steps are equivalent to applying the softmax function to the exponential calculation result, returning a probability distribution where each value represents a weight corresponding to a preset model. Alternatively, the weights can be adjusted through... Figure 2 The diagram illustrates how to initiate online incremental learning and generate adversarial examples.

[0039] In another embodiment, the method further includes: S6: Obtain the confidence level of the result generated when the preset model reprocesses the financial data stream after weight adjustment, and define the confidence level as the new confidence level. At the same time, define the confidence level of the result generated when the preset model processes the financial data stream without weight adjustment as the original confidence level. S7: Verify the credibility of each preset model based on the difference between the new confidence level and the original confidence level.

[0040] In this embodiment, the confidence level of the processing result generated when the pre-adjusted model reprocesses the financial data stream is defined as the new confidence level. The confidence level of the result generated when the pre-adjusted model processes the financial data stream without weight adjustment is defined as the original confidence level. After obtaining the new and original confidence levels, the system calculates the difference between the two confidence levels and uses this difference to verify the credibility of each pre-adjusted model. For example, the difference between the two confidence levels can be calculated and compared with a preset standard, such as a rate of change. If the difference is ≤5%, the credibility meets the requirements, and the weight update matches the current scenario. If the difference is >5%, the credibility does not meet the requirements.

[0041] Specifically, the method further includes: S8: When the confidence level exceeds the second preset threshold, update the complexity index based on the confidence level or the difference between the new confidence level and the original confidence level, or update the weights of the preset model again.

[0042] In this embodiment, the aforementioned preset standard configuration is a second preset threshold. When the confidence level exceeds the second preset threshold, the system will update the complexity index based on the determined confidence level or based on the difference between the new confidence level and the original confidence level.

[0043] like Figure 3 As shown, the method further includes: S9: Detect the data distribution divergence of the financial data streams obtained at different times; S10: An alarm is triggered when the detection result exceeds the third preset threshold.

[0044] In this embodiment, the distribution of the input data is detected to see if it changes significantly over time. For example, KL divergence (Kullback-Leibler divergence, relative entropy) can be used to quantify the difference between the current data distribution and the historical reference distribution. When the KL divergence exceeds a third preset threshold, such as, but not limited to, 0.15, the system issues an alarm.

[0045] The changes in data distribution, such as altered user behavior, sensor aging, and sudden environmental changes, can lead to a decline in the performance of the preset model. KL divergence is a metric with clear mathematical meaning, and the third preset threshold can be optimized based on historical data and business tolerance to achieve a balance between sensitivity and false alarm rate. By detecting changes in data distribution as described above, such changes can be identified immediately, preventing the system from operating in a faulty state. Upon triggering an alarm, the system can automatically take action, such as retraining the model, switching to a backup model, adjusting fusion weights, or notifying maintenance personnel for intervention.

[0046] In another embodiment, the method further includes: S11: When the complexity index has a second relationship with the first preset threshold, a preset emergency handling strategy is triggered; S12: When the complexity index has a third relationship with the first preset threshold, the preset model is controlled to process the financial data stream based on the preset conventional processing strategy; The first relationship includes a complexity index exceeding the first preset threshold; the second relationship includes a complexity index within a range less than the first preset threshold and having a first difference from the first preset threshold; and the third relationship includes a complexity index within a range less than the first preset threshold and having a second difference from the first preset threshold, wherein the first difference is smaller than the second difference.

[0047] In this embodiment, when a second relationship exists between the complexity index and the first preset threshold, a preset emergency handling strategy will be triggered, such as notifying maintenance personnel to handle the situation; the specific strategy is not unique. Furthermore, when a third relationship exists between the complexity index and the first preset threshold, the preset model will be controlled to process the financial data stream based on a preset conventional processing strategy. For example, the preset model may not be weighted; the financial data stream can be processed directly based on the preset model under its current weight state. Alternatively, other preset processing strategies may be used; the specific strategy is not unique.

[0048] Specifically, in this embodiment, the first relationship includes a complexity index exceeding the first preset threshold, such as a complexity index greater than or equal to 90 if the first preset threshold is 90; the second relationship includes a complexity index within a range less than the first preset threshold and having a first difference from the first preset threshold, such as a complexity index between 90 and 70; the third relationship includes a complexity index within a range less than the first preset threshold and having a second difference from the first preset threshold, where the first difference is less than the second difference, such as a complexity index less than 70, etc., and the specific relationship is not unique.

[0049] The solution in this embodiment establishes a three-level response mechanism by determining the relationship between the complexity index and a first preset threshold. Based on this relationship, different levels of response are triggered, including processing the financial data stream according to a preset conventional processing method, dynamically adjusting model weights, and an emergency processing method. This can cope with different data state scenarios and different complexity abrupt changes, ensuring stable data processing accuracy while ensuring the efficiency and stability of the collaborative cooperation among various models. Especially for data with a large number of different data formats, the method based on the above embodiment still achieves excellent performance. In practical applications, the temperature parameter can be, but is not limited to, 0.5, and the weight update can be defaulted to once every 200ms.

[0050] like Figure 4 As shown, another embodiment of the present invention also provides a financial data processing device based on a dynamic evolution mechanism, comprising: The receiving module is used to receive the monitored financial data stream; The first determining module is used to determine the complexity index of the financial data flow; The second determining module is used to determine the confidence level of the processing results of the financial data stream by different preset models when the complexity index has a first relationship with the first preset threshold. The adjustment module is used to dynamically adjust the weights of each preset model by combining the confidence level and preset adjustment parameters. The adjusted weights are positively correlated with the confidence level of the corresponding preset model. The processing module is used to reprocess the financial data stream based on the preset model after weight adjustment.

[0051] In one embodiment, determining the complexity index of the financial data stream includes: Determine the entropy value, format, and semantic density of financial data; The complexity index of the financial data is determined based on the entropy value, the format involved, and the semantic density.

[0052] In one embodiment, the step of dynamically adjusting the weights of each of the preset models by combining the confidence level and preset adjustment parameters includes: The weights of each preset model are dynamically adjusted based on the confidence level and preset temperature parameters.

[0053] In one embodiment, dynamically adjusting the weights of each preset model by combining the confidence level and preset temperature parameters includes: The confidence level of the financial data stream is converted into a data set of a specified dimension; Determine the quotient of each element in the data set with respect to the temperature parameter; Perform exponential operations on the quotients in the data set; The results of the exponential calculation are normalized to obtain new weights for each of the preset models.

[0054] In one embodiment, the device further includes: The module is used to obtain the confidence level of the result generated by the preset model after weight adjustment when it reprocesses the financial data stream, and to define the confidence level as the new confidence level. At the same time, the confidence level of the result generated by the preset model when it processes the financial data stream without weight adjustment is defined as the original confidence level. The verification module is used to verify the credibility of each of the preset models based on the difference between the new confidence level and the original confidence level.

[0055] In one embodiment, the device further includes: The update module is used to update the complexity index based on the confidence level or the difference between the new confidence level and the original confidence level when the confidence level exceeds the second preset threshold, or to update the weights of the preset model again.

[0056] In one embodiment, the device further includes: The detection module is used to detect the data distribution dispersion of the financial data streams obtained at different times; The triggering module is used to trigger an alarm when the detection result exceeds the third preset threshold.

[0057] In one embodiment, the device further includes: The first judgment module is used to trigger a preset emergency handling strategy when the complexity index has a second relationship with the first preset threshold. The second judgment module is used to control the preset model to process the financial data stream based on a preset conventional processing strategy when the complexity index has a third relationship with the first preset threshold. The first relationship includes a complexity index exceeding the first preset threshold; the second relationship includes a complexity index within a range less than the first preset threshold and having a first difference from the first preset threshold; and the third relationship includes a complexity index within a range less than the first preset threshold and having a second difference from the first preset threshold, wherein the first difference is smaller than the second difference.

[0058] Another embodiment of the present invention also provides an electronic device, comprising: One or more processors; Memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the financial data processing method based on the dynamic evolution mechanism as described in any one of the above descriptions.

[0059] Furthermore, one embodiment of the present invention also provides a storage medium storing a computer program, which, when executed by a processor, implements the financial data processing method based on a dynamic evolution mechanism as described above. It should be understood that the various solutions in this embodiment have the corresponding technical effects in the above-described method embodiments, and will not be repeated here.

[0060] Furthermore, embodiments of the present invention also provide a computer program product, which is tangibly stored on a computer-readable medium and includes computer-readable instructions that, when executed, cause at least one processor to perform a financial data processing method based on a dynamic evolution mechanism, such as the embodiments described above.

[0061] It should be noted that the computer storage medium of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access storage medium (RAM), a read-only storage medium (ROM), an erasable programmable read-only storage medium (EPROM or flash memory), an optical fiber, a portable compact disk read-only storage medium (CD-ROM), an optical storage medium, a magnetic storage medium, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program configured for use by or in connection with an instruction execution system, system, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, antenna, optical fiber, RF, etc., or any suitable combination thereof.

[0062] Furthermore, those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0063] This 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 block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can 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, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0064] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction set implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0065] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of protection of this application is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of one or more embodiments of this application as described above, which are not provided in detail for the sake of brevity.

Claims

1. A financial data processing method based on a dynamic evolution mechanism, characterized in that, include: Receive monitored financial data streams; Determine the complexity index of the financial data flow; When the complexity index has a first relationship with the first preset threshold, the confidence level of the processing results of the financial data stream by different preset models is determined; The weights of each preset model are dynamically adjusted based on the confidence level and preset adjustment parameters. The adjusted weights are positively correlated with the confidence level of the corresponding preset model. The financial data stream is reprocessed based on the pre-defined model after weight adjustment.

2. The financial data processing method based on a dynamic evolution mechanism according to claim 1, characterized in that, Determining the complexity index of the financial data stream includes: Determine the entropy value, format, and semantic density of the financial data stream; The complexity index of the financial data stream is determined based on the entropy value, the format involved, and the semantic density.

3. The financial data processing method based on a dynamic evolution mechanism according to claim 1, characterized in that, The step of dynamically adjusting the weights of each preset model by combining the confidence level and preset adjustment parameters includes: The weights of each preset model are dynamically adjusted based on the confidence level and preset temperature parameters.

4. The financial data processing method based on a dynamic evolution mechanism according to claim 1, characterized in that, The dynamic adjustment of the weights of each preset model by combining the confidence level and preset temperature parameters includes: The confidence level of the financial data stream is converted into a data set of a specified dimension; Determine the quotient of each element in the data set with respect to the temperature parameter; Perform exponential operations on the quotients in the data set; The results of the exponential calculation are normalized to obtain new weights for each of the preset models.

5. The financial data processing method based on a dynamic evolution mechanism according to claim 1, characterized in that, The method further includes: The confidence level of the result generated when the preset model reprocesses the financial data stream after weight adjustment is obtained, and this confidence level is defined as the new confidence level. At the same time, the confidence level of the result generated when the preset model processes the financial data stream without weight adjustment is defined as the original confidence level. The credibility of each preset model is verified based on the difference between the new confidence level and the original confidence level.

6. The financial data processing method based on a dynamic evolution mechanism according to claim 5, characterized in that, The method further includes: When the confidence level exceeds the second preset threshold, the complexity index is updated based on the confidence level or the difference between the new confidence level and the original confidence level, or the weights of the preset model are updated again.

7. The financial data processing method based on a dynamic evolution mechanism according to claim 1, characterized in that, The method further includes: The data distribution divergence of the financial data streams obtained at different times is detected; An alarm is triggered when the detection result exceeds the third preset threshold.

8. The financial data processing method based on a dynamic evolution mechanism according to claim 1, characterized in that, The method further includes: When the complexity index has a second relationship with the first preset threshold, a preset emergency handling strategy is triggered; When the complexity index has a third relationship with the first preset threshold, the preset model is controlled to process the financial data stream based on the preset conventional processing strategy; The first relationship includes a complexity index exceeding the first preset threshold; the second relationship includes a complexity index within a range less than the first preset threshold and having a first difference from the first preset threshold; and the third relationship includes a complexity index within a range less than the first preset threshold and having a second difference from the first preset threshold, wherein the first difference is smaller than the second difference.

9. A financial data processing device based on a dynamic evolution mechanism, characterized in that, include: The receiving module is used to receive the monitored financial data stream; The first determining module is used to determine the complexity index of the financial data flow; The second determining module is used to determine the confidence level of the processing results of the financial data stream by different preset models when the complexity index has a first relationship with the first preset threshold. The adjustment module is used to dynamically adjust the weights of each preset model by combining the confidence level and preset adjustment parameters. The adjusted weights are positively correlated with the confidence level of the corresponding preset model. The processing module is used to reprocess the financial data stream based on the preset model after weight adjustment.

10. An electronic device, comprising: One or more processors; Memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the financial data processing method based on a dynamic evolution mechanism as described in any one of claims 1-8.

Images