AI-assisted large-scale data anomaly detection and early warning system

Abstract

The invention discloses an AI-assisted large-scale data anomaly detection and early warning system, which comprises a detection unit and an early warning unit, and is characterized in that the detection unit can obtain data information, extract feature information in the obtained information, judge whether the feature information is abnormal after obtaining the feature information, start the early warning system when the feature information is abnormal, and send the early warning unit to the early warning unit when the feature information is abnormal. The detection unit comprises a data acquisition module, a data filtering module, a feature capturing module, a data classification module, a data processing module, an AI module, an input module, a comparison module and a judgment module. When the system is implemented, the detection unit can check financial data in the enterprise operation process so as to calculate the profit and loss condition or the tax payment condition or various expenditure conditions of an enterprise, when the calculation result is different from the actual data, abnormal information can be judged through the judgment module, and the enterprise can be timely managed. And then the early warning unit sends out early warning information so as to remind a user.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and specifically to an AI-assisted large-scale data anomaly detection and early warning system. Background Art

[0002] A large amount of data is generated in the course of business operations. Among them, financial data is the most important and requires a lot of manpower and material resources every year. Financial data not only affects the company's own profits and losses, but also whether the company pays taxes in accordance with laws and regulations.

[0003] In the prior art, when statistical calculations are performed on financial data, they are usually performed manually. However, in the course of business operations, the larger the scale and the more employees a company has, the larger the financial data generated, and the more troublesome it is to perform manual calculations.

[0004] Nowadays, the methods of using computer programs to calculate financial data are becoming more and more mature. However, in the existing accounting methods, for each type of financial data, a corresponding set of calculation methods needs to be set up. When the amount of data is large (at the end of the year or the end of the quarter), due to the limited operating capacity of the system, the results will be exported slowly. In order to meet the needs of special periods, it is necessary to spend a lot of costs, which obviously cannot meet the needs of more companies. For this reason, this application proposes an AI-assisted large-scale data anomaly detection and early warning system. Summary of the invention

[0005] To this end, the present application provides an AI-assisted large-scale data anomaly detection and early warning system to solve the problem in the prior art that when the amount of data is large, the result export will be slow due to the limited system operating capacity. In order to meet the needs of special periods, it is necessary to spend a lot of cost, which obviously cannot meet the needs of more enterprises.

[0006] In order to achieve the above objectives, this application provides the following technical solutions:

[0007] In the first aspect, an AI-assisted large-scale data anomaly detection and early warning system includes a detection unit and an early warning unit. The detection unit can acquire data information and extract feature information from the acquired information. After acquiring the feature information, it is determined whether the feature information is abnormal. When the feature information is abnormal, the early warning system is activated and a warning signal is issued.

[0008] The detection unit includes a data acquisition module, a data filtering module, a feature capture module, a data classification module, a data processing module, an AI module, an input module, a comparison module and a judgment module, wherein the output end of the data acquisition module is connected to the input end of the feature capture module, the output end of the feature capture module is connected to the input end of the data classification module, the output end of the data classification module is connected to the input end of the data processing module, the output end of the AI ​​module is connected to the input end of the data processing module, the output end of the data processing module is connected to the input end of the comparison module, the output end of the input module is connected to the input end of the comparison module, the output end of the input module is connected to the input end of the comparison module, and the output end of the comparison module is connected to the input end of the judgment module;

[0009] The data acquisition module can acquire financial data in the business operation process of the enterprise; the feature capture module is used to capture the feature values ​​of the financial data to obtain the feature information of each financial data; the data filtering module removes the unnecessary financial data according to the feature information captured by the feature capture module; the data classification module classifies the financial data according to the data feature information captured by the feature capture module; the data processing module can process different types of financial data according to preset rules and output the calculation results of each type of financial data; the AI ​​module is used to allocate the computing power of the data processing module according to the amount of different types of financial data; the input module is used to input comparison information; the judgment module is used to compare the calculation results with the comparison information to determine whether the data is abnormal;

[0010] The early warning unit includes a connection module, an early warning information editing module, and an early warning information generation module. The input end of the connection module is connected to the output end of the judgment module, the output end of the connection module is connected to the input end of the early warning information editing module, and the output end of the early warning information editing module is connected to the input end of the early warning information generation module.

[0011] The connection module is used to connect the judgment module with the warning information editing module. The warning information editing module edits the warning information according to the judgment result of the judgment module. The warning information generation module is used to publish the warning information edited by the warning information editing module.

[0012] Preferably, the data processing module has a plurality of corresponding data processing nodes, each corresponding data processing node is used to process a type of financial data, and each corresponding data processing node stores an algorithm for processing the corresponding financial data.

[0013] Preferably, a general node is also provided in the data processing module, and the output end of the AI ​​module is connected to the input end of the general node.

[0014] Preferably, the AI ​​module can also count the amount of each type of financial data. When the amount of one type of financial data is much larger than the amount of another type of financial data, the AI ​​module calls the general node so that the general node calculates the financial data with a larger amount of data.

[0015] Preferably, the amount of one type of financial data a that needs to be processed is recorded as A, and the amount of another type of financial data b that needs to be processed is recorded as B. When A / B is greater than 2, the AI ​​module calls the general node to calculate the financial data a.

[0016] Preferably, the AI ​​module can also estimate the time it takes to complete the calculation of each type of financial data. When the time it takes to complete the calculation of one type of financial data is much longer than the time it takes to complete the calculation of another type of financial data, the AI ​​module calls the general node so that the general node calculates the financial data that takes longer to complete.

[0017] Preferably, the processing completion time of one type of financial data c is recorded as C, and the processing completion time of another type of financial data d is recorded as D. When C / D is greater than 2, the AI ​​module calls the general node to calculate the financial data C.

[0018] Preferably, a database is also included, and the algorithms in each corresponding data processing node are stored in the database. The AI ​​module can copy the algorithms in the database to the general node so that the general node can calculate the financial data.

[0019] Preferably, a storage module is also included, and the storage module can save the judgment result of the judgment module and the warning information generated by the warning information generation module.

[0020] Preferably, the early warning unit further includes a wireless communication module, and the wireless communication module can send the early warning information generated by the early warning information generating module to the communication terminal.

[0021] Compared with the prior art, this application has at least the following beneficial effects:

[0022] When this system is implemented, the detection unit can calculate the financial data of the enterprise during its business operations to work out the enterprise's profit and loss situation, tax situation or various expenses. When the calculated result is different from the actual data, the judgment module can identify abnormal information, and then the early warning unit will issue an early warning message to remind the user.

[0023] When this system is implemented, an AI module is set up. The AI ​​module can adjust the processing speed of the data processing module, that is, with less increase in cost, the operating speed of the data processing module is greatly improved, so that various financial data can be calculated quickly and efficiently, meeting the needs of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more intuitively illustrate the prior art and the present application, exemplary drawings are given below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing the present application; for example, those skilled in the art are capable of easily making conventional adjustments or further optimizations to the addition / reduction / attribution division, specific shapes, positional relationships, connection methods, dimensional ratios, etc. of certain units (components) based on the technical concepts and exemplary drawings disclosed in the present application.

[0025] Figure 1 A module diagram of an AI-assisted large-scale data anomaly detection and early warning system provided in Example 1 of the present application. DETAILED DESCRIPTION

[0026] The present application is further described below in detail through specific embodiments in conjunction with the accompanying drawings.

[0027] An AI-assisted large-scale data anomaly detection and early warning system includes a detection unit and an early warning unit. The detection unit can acquire data information and extract feature information from the acquired information. After acquiring the feature information, it is determined whether the feature information is abnormal. When the feature information is abnormal, the early warning system is activated and a warning signal is issued.

[0028] The detection unit includes a data acquisition module, a data filtering module, a feature capture module, a data classification module, a data processing module, an AI module, an input module, a comparison module and a judgment module, wherein the output end of the data acquisition module is connected to the input end of the feature capture module, the output end of the feature capture module is connected to the input end of the data classification module, the output end of the data classification module is connected to the input end of the data processing module, the output end of the AI ​​module is connected to the input end of the data processing module, the output end of the data processing module is connected to the input end of the comparison module, the output end of the input module is connected to the input end of the comparison module, the output end of the input module is connected to the input end of the comparison module, and the output end of the comparison module is connected to the input end of the judgment module;

[0029] The data acquisition module can acquire financial data during the business operation of the enterprise; the feature capture module is used to capture the feature values ​​of the financial data to obtain the feature information of each financial data; the data filtering module removes the unnecessary financial data according to the feature information captured by the feature capture module; the data classification module classifies the financial data according to the data feature information captured by the feature capture module; the data processing module can process different types of financial data according to preset rules and output the calculation results of each type of financial data; the AI ​​module is used to allocate the computing power of the data processing module according to the amount of different types of financial data; the input module is used to input comparison information, and the input comparison information is the real financial data of the enterprise. When the real financial data is different from the calculated result, it represents the occurrence of an abnormal situation; the judgment module is used to compare the calculation result with the comparison information to determine whether the data is abnormal;

[0030] The early warning unit includes a connection module, an early warning information editing module, and an early warning information generation module. The input end of the connection module is connected to the output end of the judgment module, the output end of the connection module is connected to the input end of the early warning information editing module, and the output end of the early warning information editing module is connected to the input end of the early warning information generation module.

[0031] The connection module is used to connect the judgment module with the warning information editing module. The warning information editing module edits the warning information according to the judgment result of the judgment module. The warning information generation module is used to publish the warning information edited by the warning information editing module.

[0032] When this system is implemented, the detection unit can calculate the financial data of the enterprise during its business operations to work out the enterprise's profit and loss situation, tax situation or various expenses. When the calculated result is different from the actual data, the judgment module can identify abnormal information, and then the early warning unit will issue an early warning message to remind the user.

[0033] When this system is implemented, an AI module is set up. The AI ​​module can adjust the processing speed of the data processing module, that is, with less increase in cost, the operating speed of the data processing module is greatly improved, so that various financial data can be calculated quickly and efficiently, meeting the needs of the enterprise.

[0034] The data processing module has multiple corresponding data processing nodes, each corresponding data processing node is used to process a type of financial data, and each corresponding data processing node stores an algorithm for processing the corresponding financial data. One type of financial data or one type of data corresponds to one processing node, which can speed up the operation efficiency of the system.

[0035] A general node is also provided in the data processing module, and the output end of the AI ​​module is connected to the input end of the general node.

[0036] The AI ​​module can also count the amount of each type of financial data. When the amount of one type of financial data is much larger than the amount of another type of financial data, the AI ​​module calls the general node so that the general node calculates the financial data with a larger amount of data.

[0037] The amount of one type of financial data a that needs to be processed is recorded as A, and the amount of the other type of financial data b that needs to be processed is recorded as B. When A / B is greater than 2, the AI ​​module calls the general node to calculate the financial data a.

[0038] It also includes a database, and the algorithms in each corresponding data processing node are stored in the database. The AI ​​module can copy the algorithms in the database to the general node so that the general node can calculate the financial data.

[0039] It also includes a storage module, which can save the judgment results of the judgment module and the warning information generated by the warning information generation module.

[0040] The early warning unit also includes a wireless communication module, which can send the early warning information generated by the early warning information generating module to the communication terminal.

[0041] When implementing this embodiment, when the amount of one type of financial data is much larger than the amount of other financial data, it will take more time to process this type of financial data, and the user will get the result more slowly. In order to solve this problem and speed up the export of results, an AI module is used. The AI ​​module can call the algorithm required by the financial data, thereby speeding up the export of results.

[0042] In this embodiment, A / B may be greater than x, and the specific value of x may be preset according to the user. In this embodiment, x=2.

[0043] Embodiment 1

[0044] An AI-assisted large-scale data anomaly detection and early warning system includes a detection unit and an early warning unit. The detection unit can acquire data information and extract feature information from the acquired information. After acquiring the feature information, it is determined whether the feature information is abnormal. When the feature information is abnormal, the early warning system is activated and a warning signal is issued.

[0045] The detection unit includes a data acquisition module, a data filtering module, a feature capture module, a data classification module, a data processing module, an AI module, an input module, a comparison module and a judgment module, wherein the output end of the data acquisition module is connected to the input end of the feature capture module, the output end of the feature capture module is connected to the input end of the data classification module, the output end of the data classification module is connected to the input end of the data processing module, the output end of the AI ​​module is connected to the input end of the data processing module, the output end of the data processing module is connected to the input end of the comparison module, the output end of the input module is connected to the input end of the comparison module, the output end of the input module is connected to the input end of the comparison module, and the output end of the comparison module is connected to the input end of the judgment module;

[0046] The data acquisition module can acquire financial data during the business operation of the enterprise; the feature capture module is used to capture the feature values ​​of the financial data to obtain the feature information of each financial data; the data filtering module removes the unnecessary financial data according to the feature information captured by the feature capture module; the data classification module classifies the financial data according to the data feature information captured by the feature capture module; the data processing module can process different types of financial data according to preset rules and output the calculation results of each type of financial data; the AI ​​module is used to allocate the computing power of the data processing module according to the amount of different types of financial data; the input module is used to input comparison information, and the input comparison information is the real financial data of the enterprise. When the real financial data is different from the calculated result, it represents the occurrence of an abnormal situation; the judgment module is used to compare the calculation result with the comparison information to determine whether the data is abnormal;

[0047] The early warning unit includes a connection module, an early warning information editing module, and an early warning information generation module. The input end of the connection module is connected to the output end of the judgment module, the output end of the connection module is connected to the input end of the early warning information editing module, and the output end of the early warning information editing module is connected to the input end of the early warning information generation module.

[0048] The connection module is used to connect the judgment module with the warning information editing module. The warning information editing module edits the warning information according to the judgment result of the judgment module. The warning information generation module is used to publish the warning information edited by the warning information editing module.

[0049] When this system is implemented, the detection unit can calculate the financial data of the enterprise during its business operations to work out the enterprise's profit and loss situation, tax situation or various expenses. When the calculated result is different from the actual data, the judgment module can identify abnormal information, and then the early warning unit will issue an early warning message to remind the user.

[0050] When this system is implemented, an AI module is set up. The AI ​​module can adjust the processing speed of the data processing module, that is, with less increase in cost, the operating speed of the data processing module is greatly improved, so that various financial data can be calculated quickly and efficiently, meeting the needs of the enterprise.

[0051] The data processing module has multiple corresponding data processing nodes, each corresponding data processing node is used to process a type of financial data, and each corresponding data processing node stores an algorithm for processing the corresponding financial data. One type of financial data or one type of data corresponds to one processing node, which can speed up the operation efficiency of the system.

[0052] A general node is also provided in the data processing module, and the output end of the AI ​​module is connected to the input end of the general node.

[0053] The AI ​​module can also estimate the time it takes to complete the calculation of each type of financial data. When the time it takes to complete the calculation of one type of financial data is much longer than the time it takes to complete the calculation of another type of financial data, the AI ​​module calls the general node to enable the general node to calculate the financial data that takes longer to complete.

[0054] The processing completion time of one type of financial data c is recorded as C, and the processing completion time of another type of financial data d is recorded as D. When C / D is greater than 2, the AI ​​module calls the general node to calculate the financial data C.

[0055] It also includes a database, and the algorithms in each corresponding data processing node are stored in the database. The AI ​​module can copy the algorithms in the database to the general node so that the general node can calculate the financial data.

[0056] It also includes a storage module, which can save the judgment results of the judgment module and the warning information generated by the warning information generation module.

[0057] The early warning unit also includes a wireless communication module, which can send the early warning information generated by the early warning information generating module to the communication terminal.

[0058] When this embodiment is implemented, when the time for processing one type of financial data is much longer than the time for processing other financial data, the system will export the results of all data more slowly, that is, some data have already produced results, while some data will take a longer time to produce results. In order to solve this problem, an AI module is used. The AI ​​module can call the algorithm required by the financial data, so that the results of each data can be exported at the same or similar time.

[0059] In this embodiment, C / D may be greater than y, and the specific value of y may be preset according to the user. In this embodiment, y=2.

[0060] The technical features of the above embodiments may be arbitrarily combined (as long as there is no contradiction in the combination of these technical features). To make the description concise, not all possible combinations of the technical features in the above embodiments are described; these embodiments that are not explicitly written should also be considered to be within the scope of this specification.

Claims

1. An AI-assisted large-scale data anomaly detection and early warning system, characterized in that: It includes a detection unit and an early warning unit. The detection unit can acquire data information and extract characteristic information from the acquired information. After acquiring the characteristic information, it determines whether the characteristic information is abnormal. When the characteristic information is abnormal, the early warning system is activated and a warning signal is issued. The detection unit includes a data acquisition module, a data filtering module, a feature capture module, a data classification module, a data processing module, an AI module, an input module, a comparison module and a judgment module, wherein the output end of the data acquisition module is connected to the input end of the feature capture module, the output end of the feature capture module is connected to the input end of the data classification module, the output end of the data classification module is connected to the input end of the data processing module, the output end of the AI ​​module is connected to the input end of the data processing module, the output end of the data processing module is connected to the input end of the comparison module, the output end of the input module is connected to the input end of the comparison module, the output end of the input module is connected to the input end of the comparison module, and the output end of the comparison module is connected to the input end of the judgment module; The data acquisition module can acquire financial data in the business operation process of the enterprise; the feature capture module is used to capture the feature values ​​of the financial data to obtain the feature information of each financial data; the data filtering module removes the unnecessary financial data according to the feature information captured by the feature capture module; the data classification module classifies the financial data according to the data feature information captured by the feature capture module; the data processing module can process different types of financial data according to preset rules and output the calculation results of each type of financial data; the AI ​​module is used to allocate the computing power of the data processing module according to the amount of different types of financial data; the input module is used to input comparison information; the judgment module is used to compare the calculation results with the comparison information to determine whether the data is abnormal; The early warning unit includes a connection module, an early warning information editing module, and an early warning information generation module. The input end of the connection module is connected to the output end of the judgment module, the output end of the connection module is connected to the input end of the early warning information editing module, and the output end of the early warning information editing module is connected to the input end of the early warning information generation module. The connection module is used to connect the judgment module with the warning information editing module. The warning information editing module edits the warning information according to the judgment result of the judgment module. The warning information generation module is used to publish the warning information edited by the warning information editing module.

2. The AI-assisted large-scale data anomaly detection and early warning system according to claim 1, characterized in that: The data processing module has a plurality of corresponding data processing nodes, each corresponding data processing node is used to process a type of financial data, and each corresponding data processing node stores an algorithm for processing the corresponding financial data.

3. The AI-assisted large-scale data anomaly detection and early warning system according to claim 2, characterized in that: A general node is also provided in the data processing module, and the output end of the AI ​​module is connected to the input end of the general node.

4. The AI-assisted large-scale data anomaly detection and early warning system according to claim 3, characterized in that: The AI ​​module can also count the amount of each type of financial data. When the amount of one type of financial data is much larger than the amount of another type of financial data, the AI ​​module calls the general node so that the general node calculates the financial data with a larger amount of data.

5. The AI-assisted large-scale data anomaly detection and early warning system according to claim 4, characterized in that: The amount of one type of financial data a that needs to be processed is recorded as A, and the amount of the other type of financial data b that needs to be processed is recorded as B. When A / B is greater than 2, the AI ​​module calls the general node to calculate the financial data a.

6. The AI-assisted large-scale data anomaly detection and early warning system according to claim 3, characterized in that: The AI ​​module can also estimate the time it takes to complete the calculation of each type of financial data. When the time it takes to complete the calculation of one type of financial data is much longer than the time it takes to complete the calculation of another type of financial data, the AI ​​module calls the general node to enable the general node to calculate the financial data that takes longer to complete.

7. The AI-assisted large-scale data anomaly detection and early warning system according to claim 6, characterized in that: The processing completion time of one type of financial data c is recorded as C, and the processing completion time of another type of financial data d is recorded as D. When C / D is greater than 2, the AI ​​module calls the general node to calculate the financial data C.

8. The AI-assisted large-scale data anomaly detection and early warning system according to claim 3, characterized in that: It also includes a database, and the algorithms in each corresponding data processing node are stored in the database. The AI ​​module can copy the algorithms in the database to the general node so that the general node can calculate the financial data.

9. The AI-assisted large-scale data anomaly detection and early warning system according to claim 1, characterized in that: It also includes a storage module, which can save the judgment results of the judgment module and the warning information generated by the warning information generation module.

10. The AI-assisted large-scale data anomaly detection and early warning system according to claim 1, characterized in that: The early warning unit also includes a wireless communication module, which can send the early warning information generated by the early warning information generating module to the communication terminal.

Images