Cloud financial system and method based on artificial intelligence

Abstract

The invention discloses a cloud financial system and method based on artificial intelligence, and relates to the technical field of financial data management, and the system comprises a financial management platform which is in communication connection with a data collection and preprocessing module, an abnormal data analysis module, a monitoring and early warning module, a performance optimization module and a report generation and distribution module, the modules are in electric signal connection; the data acquisition preprocessing module is used for acquiring financial data from each financial data source. According to the method, the change trend of financial data is continuously monitored, historical data and industry standards are combined, the abnormal resource demand index is calculated, early warning mechanisms of different levels are triggered according to a set risk threshold value, a resource allocation strategy is dynamically adjusted according to the real-time load condition, it is ensured that efficient operation is still kept under the high-concurrency condition, and the resource allocation efficiency is improved. The response speed of the system is improved, potential financial risks can be effectively prevented and dealt with, and financial health of enterprises is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of financial data management, and in particular to a cloud financial system and method based on artificial intelligence. Background Art

[0002] As the scale of enterprises expands and the scope of business expands, the workload of financial management increases dramatically. Traditional financial management methods require a lot of labor costs and time costs. Intelligent financial systems combined with artificial intelligence technology can automatically process large amounts of financial data. Through automation and intelligent technology, it reduces manual intervention and significantly improves the efficiency of financial management. It can also update and analyze financial data in real time, provide accurate financial decision-making recommendations, and help companies better grasp market opportunities and optimize resource allocation.

[0003] As the amount of financial data increases, cloud financial systems are prone to performance bottlenecks, resulting in slower processing speeds and affecting the real-time and response speed of financial data analysis. Therefore, how to analyze the resource requirements of financial data to balance the operational requirements of financial data and improve the processing capacity of the system is a problem we need to solve. To this end, a cloud financial system and method based on artificial intelligence is proposed. Summary of the invention

[0004] The purpose of the present invention is to provide a cloud financial system and method based on artificial intelligence to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: In a first aspect, a cloud financial system based on artificial intelligence includes a financial management platform, wherein the financial management platform is communicatively connected to a data acquisition and preprocessing module, an abnormal data analysis module, a monitoring and early warning module, a performance optimization module, and a report generation and distribution module, wherein electrical signals are connected between the modules; The data collection and preprocessing module is used to collect financial data from various financial data sources and preprocess the financial data to obtain a financial data sequence table, ensure the quality and consistency of the input data, and lay the foundation for subsequent data processing and analysis; The abnormal data analysis module analyzes the financial data in the financial data sequence table, identifies the financial data with abnormal resource requirements, and provides accurate financial analysis results; The monitoring and early warning module continuously monitors the financial status based on the abnormal resource demand analysis results, identifies dynamic changes in financial data, analyzes potential risks and triggers early warning mechanisms to promptly detect financial risks and anomalies; The performance optimization module dynamically adjusts the system's resource allocation strategy based on the early warning results and real-time load conditions to ensure that the system remains efficient under high concurrency conditions and meets data processing requirements; The report generation and distribution module automatically generates performance optimization reports according to user needs and distributes them via email or text messages to clarify the computing resource requirements of financial data.

[0006] A further improvement of the technical solution of the present invention is that the data acquisition preprocessing module includes: Identify available financial data sources and establish secure connections to various data sources, including ERP systems, bank interfaces, spreadsheets and databases; Extract relevant financial data from each identified financial data source and integrate data extracted from different data sources into a unified data set; Preprocess the financial data integrated in the data set, including data cleaning, data format conversion and data standardization. Data cleaning is used to remove redundancy, errors and outliers in the data. Redundant data includes repeated records or irrelevant information, which need to be deleted. Error data includes incorrect values, formats or logic, which need to be corrected. Outliers are caused by input errors or system anomalies, which need to be identified and processed. Data in different formats are converted into a unified format through data format conversion, and data are standardized to ensure comparability between different data. The preprocessed financial data are arranged in chronological order to form a continuous time series, and then a structured financial data sequence table is generated.

[0007] A further improvement of the technical solution of the present invention is that the abnormal data analysis module includes: Extract the preprocessed and standardized financial data sequence table from the financial data sequence table, and perform preliminary checks on the loaded data, including data type, data range, and data consistency; According to the characteristics of financial data, analyze the amount, transaction frequency and time interval in the financial data, clarify the resource requirements of financial data, and preset the standard range of financial data resource requirements under different requirements based on the analysis of historical data; Perform anomaly detection on financial data, calculate the difference between the actual resource demand of financial data and the standard resource demand, and then calculate the abnormal resource demand index to identify the abnormal resource demand trend of financial data; Interpret and classify the detected financial data to determine its specific type and cause, and combine business knowledge and historical data to infer the cause of the anomaly.

[0008] A further improvement of the technical solution of the present invention is that the calculation process of the abnormal resource demand index is: Extract the actual amount, transaction frequency and time interval from the financial data sequence table, and extract the standard value of amount, standard value of transaction frequency and standard value of time interval set according to historical data; For each financial data point, calculate the difference ratio between the actual value of the amount, transaction frequency and time interval and its standard value, and obtain the amount difference ratio, transaction frequency difference ratio and time interval difference ratio; Calculate the product ratio of the actual value of the amount, transaction frequency and time interval to the standard value, and take the square of the logarithm to measure the overall deviation between the actual value and the standard value, and then calculate the exponential function part to smooth the difference ratio, so that the overall result is more stable and within a reasonable range; The three difference ratios of amount, transaction frequency and time interval are averaged to obtain the total difference ratio, and the total difference ratio is multiplied by the exponential function part to obtain the final abnormal resource demand index.

[0009] A further improvement of the technical solution of the present invention is that the expression of the abnormal resource demand index is as follows: ; In the formula, is the abnormal resource demand index, is the actual amount, is the actual transaction frequency, is the actual time interval, is the standard amount value, which is the center value of the standard amount range. is the standard value of transaction frequency, which is the center value of the standard transaction frequency range. is the standard value of the time interval, which is the center value of the standard time interval range. The value range is between 0 and 1.

[0010] A further improvement of the technical solution of the present invention is that the monitoring and early warning module includes: Receive the calculated abnormal resource demand index and its related financial data from the abnormal data analysis module, and combine it with historical data and real-time data to form a complete monitoring data set; According to business needs and historical data, different monitoring rules are set to determine different risk levels, namely low risk level, medium risk level and high risk level, and the corresponding warning threshold is matched for each risk level in combination with the abnormal resource demand index; Continuously monitor new financial data, update the abnormal resource demand index in real time, and compare it with the set warning threshold to determine the risk level of the current financial data, and then comprehensively assess the risks of the current financial data; The current risk level of financial data is determined based on the abnormal resource demand index, and the corresponding early warning mechanism is triggered. Early warning information is generated synchronously and then sent to relevant personnel. For low risk levels, email notifications are sent to relevant departments or personnel to remind them to pay attention to potential problems. For medium risk levels, internal alarms are triggered, SMS notifications are sent and logs are recorded, requiring further investigation and analysis. For high risk levels, management is notified immediately and emergency plans are initiated to prevent potential losses.

[0011] A further improvement of the technical solution of the present invention is that: a plurality of the risk levels are provided with a plurality of the warning thresholds in one-to-one correspondence, and the specific correspondence between the risk level and the warning threshold is as follows: The warning threshold of the low risk level is: ; Indicates that the potential problem is minor, but still requires attention; The warning thresholds for the medium risk level are: ; Indicates that there is a certain risk and further investigation is needed; The high risk level warning threshold is: ; Indicates that a significant risk exists and immediate action is required; in, is the abnormal resource demand index, is the upper threshold of the low risk level and the lower threshold of the medium risk level, is the upper threshold of the medium risk level and the lower threshold of the high risk level, , .

[0012] A further improvement of the technical solution of the present invention is that the performance optimization module includes: Receive abnormal resource demand index and related warning results from the monitoring and warning module, and obtain real-time load data of the current system from the system monitoring tool (Prometheus), including CPU usage, memory usage, I / O load, and network bandwidth; According to the early warning results and real-time load monitoring data, the comprehensive load index is calculated, and the bottleneck identification threshold T is set to identify the system bottleneck and determine whether corresponding optimization measures need to be taken, so as to dynamically adjust the resource allocation strategy. The resource allocation strategy includes horizontal expansion, vertical expansion, load balancing, cache mechanism and task scheduling. Horizontal expansion is to add more server instances or container instances to share the load, vertical expansion is to improve the hardware resources of existing servers (such as increasing the number of CPU cores and memory capacity), load balancing is to evenly distribute requests to multiple servers through the load balancer (Nginx), cache mechanism is to use cache technology (Redis) to reduce the number of database queries and improve the response speed, and task scheduling is to reasonably arrange the execution time of background tasks to avoid a large number of computing-intensive operations during peak hours; Dynamically adjust the system's resource allocation according to the established resource allocation strategy. Specific operations include: increasing / decreasing server instances, adjusting resource quotas and load balancing configurations. Through the cloud service provider's API interface, dynamically increase or decrease server instances, adjust resource quotas by modifying the resource configuration of virtual machines or containers, and redistribute traffic by updating the load balancer configuration. After resource adjustment, continuously monitor the system's performance indicators, evaluate the optimization effect, and then adjust the optimization strategy based on the monitoring results. If the optimization effect is not ideal, further adjust the resource allocation strategy and repeat the above steps.

[0013] A further improvement of the technical solution of the present invention is that the expression of the comprehensive load index is as follows: ; In the formula, It is a comprehensive load index, which is used to measure the overall load of the system. is the current CPU usage, is the current memory usage, is the current I / O load, is the current network bandwidth usage, is the standard value of CPU usage, is the standard value of memory usage, is the standard value of I / O load, It is the standard value of network bandwidth utilization.

[0014] In a second aspect, an artificial intelligence-based cloud finance method is implemented based on the artificial intelligence-based cloud finance system, comprising the following steps: Step S1: extract, clean and standardize financial data from multiple financial data sources to form a unified data set, providing a high-quality data foundation for subsequent analysis; Step S2: Conduct in-depth analysis of financial data, calculate abnormal resource demand index, and identify potential financial risks; Step S3: Based on the abnormal resource demand index and real-time load conditions, continuous monitoring is performed to identify dynamic changes in financial data and trigger corresponding early warning mechanisms; Step S4: Combine the warning results and the real-time load conditions to dynamically adjust the system's resource allocation strategy and optimize system performance through a feedback mechanism.

[0015] Due to the adoption of the above technical solution, the present invention has the following technical advances compared with the prior art: 1. The present invention provides a cloud financial system and method based on artificial intelligence. By continuously monitoring the changing trend of financial data, combining historical data and industry standards, the abnormal resource demand index is calculated, and different levels of early warning mechanisms are triggered according to the set risk thresholds. The resource allocation strategy is dynamically adjusted according to the real-time load conditions to ensure that efficient operation is maintained under high concurrency conditions. This not only improves the response speed of the system, but also effectively prevents and responds to potential financial risks and ensures the financial health of the enterprise.

[0016] 2. The present invention provides an artificial intelligence-based cloud financial system and method, which can discover potential rules and trends through in-depth analysis of financial data, thereby helping enterprises identify potential risk points and formulate corresponding response measures. It can also intelligently allocate resources and optimize performance according to real-time load conditions and business needs, dynamically adjust the number of server instances, virtual machine resource configuration, and load balancer configuration, to ensure that system costs are minimized while meeting load requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a schematic diagram of the functional modules of the system of the present invention; Figure 2 Schematic diagram of the workflow of the abnormal data analysis module of the present invention; Figure 3 This is a schematic diagram of the workflow of the monitoring and early warning module of the present invention; Figure 4 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] Embodiment 1, as Figure 1 , Figure 2 As shown, the present invention provides a cloud financial system based on artificial intelligence, including a financial management platform, the financial management platform is communicatively connected with a data acquisition preprocessing module, an abnormal data analysis module, a monitoring and early warning module, a performance optimization module and a report generation and distribution module, wherein the modules are connected by electrical signals; The data collection and preprocessing module is used to collect financial data from various financial data sources and preprocess the financial data to obtain a financial data sequence table, ensure the quality and consistency of the input data, lay the foundation for subsequent data processing and analysis, identify available financial data sources, and establish secure connections with various data sources. The financial data sources include ERP systems, bank interfaces, spreadsheets, and databases. Relevant financial data is extracted from each identified financial data source, and the data extracted from different data sources are integrated to form a unified data set. The financial data integrated in the data set is preprocessed, including data cleaning, data format conversion, and data standardization. Data cleaning is used to remove redundancy, errors, and outliers in the data. Redundant data includes repeated records or irrelevant information, which need to be deleted. Error data includes incorrect values, formats, or logic, which need to be corrected. Outliers are caused by input errors or system anomalies and need to be identified and processed. Data in different formats are converted into a unified format through data format conversion. The data is standardized to ensure comparability between different data. The preprocessed financial data is arranged in chronological order to form a continuous time series, thereby generating a structured financial data sequence table. The abnormal data analysis module analyzes the financial data in the financial data sequence table, identifies the financial data with abnormal resource demand, provides accurate financial analysis results, extracts the preprocessed and standardized financial data sequence table from the financial data sequence table, and performs preliminary checks on the loaded data, including data type, data range, and data consistency. According to the characteristics of the financial data, the amount, transaction frequency, and time interval in the financial data are analyzed to identify the resource demand of the financial data. Based on the analysis of historical data, the standard range of financial data resource demand under different requirements is preset, and the financial data is detected for anomalies. The difference between the actual resource demand of the financial data and the standard resource demand is calculated, and then the abnormal resource demand index is calculated to identify the abnormal resource demand trend of the financial data. The detected financial data is interpreted and classified to determine its specific type and cause, and the cause of the anomaly is inferred by combining business knowledge and historical data. Furthermore, the calculation process of the abnormal resource demand index is: Extract the actual amount, transaction frequency and time interval from the financial data sequence table, and extract the standard value of the amount, transaction frequency and time interval set according to the historical data. For each financial data point, calculate the difference ratio between the actual value of the amount, transaction frequency and time interval and their standard value, obtain the amount difference ratio, transaction frequency difference ratio and time interval difference ratio, calculate the product ratio of the actual value of the amount, transaction frequency and time interval and their standard value, and take the logarithm square to measure the overall deviation between the actual value and the standard value, and then calculate the exponential function part, smooth the difference ratio, so that the overall result is more stable and within a reasonable range, average the three difference ratios of the amount, transaction frequency and time interval to obtain the total difference ratio, and multiply the total difference ratio by the exponential function part to obtain the final abnormal resource demand index; Furthermore, the expression of abnormal resource demand index is as follows: ; In the formula, is the abnormal resource demand index, is the actual amount, is the actual transaction frequency, is the actual time interval, is the standard amount value, which is the center value of the standard amount range. is the standard value of transaction frequency, which is the center value of the standard transaction frequency range. is the standard value of the time interval, which is the center value of the standard time interval range. The value range is between 0 and 1. When the actual amount, transaction frequency and time interval are exactly the same as the standard value, the first part of the formula are all 0, so the result of the entire expression is 0. When the actual value is very different from the standard value, the first part is close to 1, but because the second part The overall result will not exceed 1; The monitoring and early warning module continuously monitors the financial status based on the abnormal resource demand analysis results, identifies the dynamic changes of financial data, analyzes potential risks and triggers early warning mechanisms to timely detect financial risks and anomalies; The performance optimization module dynamically adjusts the system's resource allocation strategy based on the warning results and real-time load conditions to ensure that the system remains efficient under high concurrency conditions and meets data processing requirements; The report generation and distribution module automatically generates performance optimization reports based on user needs and distributes them via email or SMS to clarify the computing resource requirements of financial data.

[0021] Embodiment 2, as Figure 3 As shown, based on Example 1, the present invention provides a technical solution: preferably, the monitoring and early warning module includes: Receive the calculated abnormal resource demand index and its related financial data from the abnormal data analysis module, and combine them with historical data and real-time data to form a complete monitoring data set. According to business needs and historical data, set different monitoring rules, determine different risk levels, namely low risk level, medium risk level and high risk level, and match the corresponding warning threshold for each risk level in combination with the abnormal resource demand index. Continuously monitor new financial data, update the abnormal resource demand index in real time, and compare it with the set warning threshold to determine the risk level of the current financial data, and then comprehensively evaluate the risks of the current financial data. Determine the current risk level of the financial data according to the abnormal resource demand index, and trigger the corresponding warning mechanism, synchronously generate warning information, and then send the warning information to relevant personnel. For low risk levels, send email notifications to relevant departments or personnel to remind them to pay attention to potential problems. For medium risk levels, trigger internal alarms, send SMS notifications and record logs to require further investigation and analysis. For high risk levels, immediately notify management and initiate emergency plans to prevent potential losses. Furthermore, multiple risk levels are set with multiple warning thresholds in one-to-one correspondence. The specific correspondence between the risk levels and the warning thresholds is as follows: The warning thresholds for low risk levels are: ; Indicates that the potential problem is minor, but still requires attention; The warning thresholds for the medium risk level are: ; Indicates that there is a certain risk and further investigation is needed; The warning thresholds for high risk levels are: ; Indicates that a significant risk exists and immediate action is required; in, is the abnormal resource demand index, is the upper threshold of the low risk level and the lower threshold of the medium risk level, is the upper threshold of the medium risk level and the lower threshold of the high risk level, , ; Performance optimization modules include: Receive the abnormal resource demand index and its related warning results from the monitoring and warning module, and obtain the real-time load data of the current system from the system monitoring tool (Prometheus), including CPU usage, memory usage, I / O load and network bandwidth. According to the warning results and real-time load monitoring data, calculate the comprehensive load index and set the bottleneck identification threshold T to identify the system bottleneck and determine whether corresponding optimization measures need to be taken, and then dynamically adjust the resource allocation strategy. The resource allocation strategy includes horizontal expansion, vertical expansion, load balancing, caching mechanism and task scheduling. Horizontal expansion is to add more server instances or container instances to share the load, vertical expansion is to improve the hardware resources of the existing server (such as increasing the number of CPU cores and memory capacity), and load balancing is to evenly distribute requests through the load balancer (Nginx). Evenly distribute to multiple servers. The cache mechanism uses cache technology (Redis) to reduce the number of database queries and improve the response speed. Task scheduling is to reasonably arrange the execution time of background tasks to avoid a large number of computing-intensive operations during peak hours. According to the formulated resource allocation strategy, the system resource allocation is dynamically adjusted. The specific operations include: increasing / decreasing server instances, adjusting resource quotas and load balancing configurations. Through the API interface of the cloud service provider, dynamically increase or decrease server instances, adjust resource quotas by modifying the resource configuration of virtual machines or containers, and redistribute traffic by updating the configuration of the load balancer. After the resources are adjusted, the system performance indicators are continuously monitored to evaluate the optimization effect, and then the optimization strategy is adjusted according to the monitoring results. If the optimization effect is not ideal, the resource allocation strategy is further adjusted and the above steps are repeated; Furthermore, the expression of the comprehensive load index is as follows: ; In the formula, It is a comprehensive load index, which is used to measure the overall load of the system. is the current CPU usage, is the current memory usage, is the current I / O load, is the current network bandwidth usage, is the standard value of CPU usage, is the standard value of memory usage, is the standard value of I / O load, is the standard value of network bandwidth utilization. When the actual data is very close to the standard value, the difference ratio of each item in the formula is close to 0, and the result of the entire expression is close to 0. When it is greater than the bottleneck identification threshold T, it indicates that there is a bottleneck in the system and resources need to be adjusted.

[0022] Embodiment 3, as Figure 4 As shown, based on Embodiment 1-2, the present invention further provides an artificial intelligence-based cloud financial method, which is implemented based on the above-mentioned artificial intelligence-based cloud financial system and includes the following steps: Step S1: extract, clean and standardize financial data from multiple financial data sources to form a unified data set, providing a high-quality data foundation for subsequent analysis; Step S2: Conduct in-depth analysis of financial data, calculate abnormal resource demand index, and identify potential financial risks; Step S3: Based on the abnormal resource demand index and real-time load conditions, continuous monitoring is performed to identify dynamic changes in financial data and trigger corresponding early warning mechanisms; Step S4: Combine the warning results and the real-time load conditions to dynamically adjust the system's resource allocation strategy and optimize system performance through a feedback mechanism.

[0023] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A cloud financial system based on artificial intelligence, including a financial management platform, characterized in that: The financial management platform is connected in communication with a data collection and preprocessing module, an abnormal data analysis module, a monitoring and early warning module, a performance optimization module, and a report generation and distribution module, wherein the modules are connected by electrical signals; The data collection and preprocessing module is used to collect financial data from various financial data sources and preprocess the financial data to obtain a financial data sequence table; The abnormal data analysis module analyzes the financial data in the financial data sequence table to identify the financial data with abnormal resource requirements; The monitoring and early warning module continuously monitors the financial status based on the abnormal resource demand analysis results, identifies dynamic changes in financial data, analyzes potential risks and triggers an early warning mechanism; The performance optimization module dynamically adjusts the system's resource allocation strategy based on the warning results and real-time load conditions; The report generation and distribution module automatically generates performance optimization reports according to user needs and distributes them via email or SMS to clarify the computing resource requirements of financial data.

2. The cloud financial system based on artificial intelligence according to claim 1, characterized in that: The data acquisition preprocessing module includes: Identify available financial data sources and establish secure connections to various data sources, including ERP systems, bank interfaces, spreadsheets and databases; Extract relevant financial data from each identified financial data source and integrate data extracted from different data sources into a unified data set; Pre-process the financial data integrated in the data set, including data cleaning, data format conversion and data standardization; The preprocessed financial data are arranged in chronological order to form a continuous time series, and then a structured financial data sequence table is generated.

3. The cloud financial system based on artificial intelligence according to claim 2, characterized in that: The abnormal data analysis module includes: Extract the preprocessed and standardized financial data sequence table from the financial data sequence table and perform a preliminary check on the loaded data; According to the characteristics of financial data, analyze the amount, transaction frequency and time interval in the financial data, clarify the resource requirements of financial data, and preset the standard range of financial data resource requirements under different requirements based on the analysis of historical data; Perform anomaly detection on financial data, calculate the difference between the actual resource demand of financial data and the standard resource demand, and then calculate the abnormal resource demand index to identify the abnormal resource demand trend of financial data; Interpret and classify the detected financial data to determine its specific type and cause, and combine business knowledge and historical data to infer the cause of the anomaly.

4. The cloud financial system based on artificial intelligence according to claim 3, characterized in that: The calculation process of the abnormal resource demand index is: Extract the actual amount, transaction frequency and time interval from the financial data sequence table, and extract the standard value of amount, standard value of transaction frequency and standard value of time interval set according to historical data; For each financial data point, calculate the difference ratio between the actual value of the amount, transaction frequency and time interval and its standard value, and obtain the amount difference ratio, transaction frequency difference ratio and time interval difference ratio; Calculate the product ratio of the actual value of the amount, transaction frequency and time interval to their standard value, take the square of the logarithm, and then calculate the exponential function part to smooth the difference ratio; The three difference ratios of amount, transaction frequency and time interval are averaged to obtain the total difference ratio, and the total difference ratio is multiplied by the exponential function part to obtain the final abnormal resource demand index.

5. The cloud financial system based on artificial intelligence according to claim 4, characterized in that: The expression of the abnormal resource demand index is as follows: ； In the formula, is the abnormal resource demand index, is the actual amount, is the actual transaction frequency, is the actual time interval, is the standard value of the amount, is the standard value of transaction frequency, is the standard value of the time interval, The value range is between 0 and 1.

6. The cloud financial system based on artificial intelligence according to claim 5, characterized in that: The monitoring and early warning module includes: Receive the calculated abnormal resource demand index and its related financial data from the abnormal data analysis module, and combine it with historical data and real-time data to form a complete monitoring data set; According to business needs and historical data, different monitoring rules are set to determine different risk levels, namely low risk level, medium risk level and high risk level, and the corresponding warning threshold is matched for each risk level in combination with the abnormal resource demand index; Continuously monitor new financial data, update the abnormal resource demand index in real time, and compare it with the set warning threshold to determine the risk level of the current financial data, and then comprehensively assess the risks of the current financial data; The current risk level of financial data is determined based on the abnormal resource demand index, and the corresponding early warning mechanism is triggered, and early warning information is generated simultaneously, and then sent to relevant personnel.

7. The cloud financial system based on artificial intelligence according to claim 6, characterized in that: Multiple risk levels are correspondingly provided with multiple warning thresholds. The specific corresponding relationship between the risk level and the warning threshold is as follows: The warning threshold of the low risk level is: ; The warning thresholds for the medium risk level are: ; The high risk level warning threshold is: ; in, is the abnormal resource demand index, is the upper threshold of the low risk level and the lower threshold of the medium risk level, is the upper threshold of the medium risk level and the lower threshold of the high risk level, , .

8. The cloud financial system based on artificial intelligence according to claim 7, characterized in that: The performance optimization module includes: Receive abnormal resource demand index and related warning results from the monitoring and warning module, and obtain real-time load data of the current system from the system monitoring tool, including CPU usage, memory usage, I / O load and network bandwidth; Based on the early warning results and real-time load monitoring data, the comprehensive load index is calculated and the bottleneck identification threshold T is set to identify system bottlenecks and determine whether corresponding optimization measures need to be taken, thereby dynamically adjusting the resource allocation strategy; Dynamically adjust the system's resource allocation according to the formulated resource allocation strategy. Specific operations include: adding / reducing server instances, adjusting resource quotas and load balancing configurations; After resource adjustment, continuously monitor the system's performance indicators, evaluate the optimization effect, and then adjust the optimization strategy based on the monitoring results. If the optimization effect is not ideal, further adjust the resource allocation strategy and repeat the above steps.

9. The cloud financial system based on artificial intelligence according to claim 8, characterized in that: The expression of the comprehensive load index is as follows: ； In the formula, It is a comprehensive load index, which is used to measure the overall load of the system. is the current CPU usage, is the current memory usage, is the current I / O load, is the current network bandwidth usage, is the standard value of CPU usage, is the standard value of memory usage, is the standard value of I / O load, It is the standard value of network bandwidth utilization.

10. An artificial intelligence-based cloud financial method, implemented based on an artificial intelligence-based cloud financial system according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step S1, extracting, cleaning and standardizing financial data from multiple financial data sources to form a unified data set; Step S2: Conduct in-depth analysis of financial data, calculate abnormal resource demand index, and identify potential financial risks; Step S3: Based on the abnormal resource demand index and real-time load conditions, continuous monitoring is performed to identify dynamic changes in financial data and trigger corresponding early warning mechanisms; Step S4: Combine the warning results and the real-time load conditions to dynamically adjust the system's resource allocation strategy and optimize system performance through a feedback mechanism.

Images