Revenue guarantee monitoring and early warning system based on AI artificial intelligence

Abstract

The invention provides an AI artificial intelligence-based income guarantee monitoring and early warning system, and relates to the field of big data processing, and the system comprises a big data collection module which collects business data of an income related business system, and synchronizes the business data to a related income protection support cloud database; the data monitoring module is used for monitoring the related income protection support cloud database, selecting training data from the business data and training a plurality of decision models; determining abnormal data, and inputting the abnormal data into a data early warning module; and the data early warning module is used for generating corresponding early warning information according to the abnormal data, and uploading the abnormal data and the early warning information to a related income protection support cloud database. By automatically accessing the business data of each business system to the data monitoring module and the decision model, the abnormal business data can be analyzed and early warned in time, and the efficiency and accuracy of monitoring and early warning are improved.

Description

Technical Field

[0001] The present application relates to the field of big data processing technology, and in particular to an income security monitoring and early warning system based on AI artificial intelligence. Background Art

[0002] In the communications industry, billing revenue is an important part of the industry's development, especially the revenue guarantee of the billing link of mobile communications operators. Revenue guarantee refers to the work of operators to ensure complete and accurate billing and charging of customers and partners by ensuring the processing, operation and process of service provision, so as to ensure operating income. It includes requirements for the completeness, accuracy and timeliness of billing data.

[0003] However, this implementation method is not suitable in today's rapidly developing mobile business. Taking traffic revenue assurance as an example, due to the huge increase in business volume, the huge number of daily call orders, and the increasing amount of user data and customer data, the conventional audit system comparison method involves too much manual participation and is easily affected by personal subjective consciousness. There are problems such as large amounts of data that are difficult to manage, time-consuming, difficult to locate problems, and insufficient flexibility in setting monitoring and warning thresholds. Summary of the invention

[0004] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, the first purpose of this application is to propose an income security monitoring and early warning system based on AI artificial intelligence.

[0006] The second objective of the present application is to provide an electronic device.

[0007] The third object of the present application is to provide a computer-readable storage medium.

[0008] A fourth object of the present application is to provide a computer program product.

[0009] To achieve the above objectives, the first embodiment of the present application proposes an income security monitoring and early warning method based on AI artificial intelligence, including:

[0010] Related income protection supports cloud database, data monitoring module, big data collection module, data decision module, and data early warning module;

[0011] The big data collection module is connected to the revenue-related business system, and is used to collect business data of the revenue-related business system, pre-process the business data, and synchronize the business data to the relevant revenue protection support cloud database;

[0012] The data monitoring module is connected to the relevant revenue protection support cloud database, and the data monitoring module is used to monitor the relevant revenue protection support cloud database, select training data from the business data and train multiple decision models;

[0013] The data decision module is connected to the data monitoring module, and the data decision module is used to process the input business data using the trained decision model to determine abnormal data, and input the abnormal data into the data early warning module;

[0014] The data warning module is connected to the data decision module and the relevant income protection support cloud database. The data warning module is used to generate corresponding warning information based on the abnormal data, and upload the abnormal data and the warning information to the relevant income protection support cloud database.

[0015] Optionally, the big data acquisition module includes:

[0016] A text extraction module, used to identify the file type of the business data and extract keywords from the file title and metadata;

[0017] A parsing module, used to parse the business data and convert the business data into structured data or unstructured data;

[0018] A data cleaning module, used to retrieve missing data and erroneous data in the business data, and correct the missing data and erroneous data;

[0019] The data output module is used to input the cleaned business data into the relevant revenue protection support cloud database.

[0020] Optionally, the data monitoring module includes:

[0021] A data change monitoring module, used for monitoring the relevant revenue protection support cloud database, and when new business data is added to the relevant revenue protection support cloud database, synchronizing the newly added business data to the training module in the form of files;

[0022] A training module, used to mark the labels of the training data and input them into the decision model to train the decision model, wherein the decision model includes a supervised learning model and an unsupervised learning model, and the labels are types corresponding to the training data;

[0023] Reinforcement learning module, used to learn the optimal behavior strategy in interaction with the environment.

[0024] Optionally, the data decision module includes:

[0025] A feature extraction module, used to extract features of the business data according to the trained decision model;

[0026] A feature analysis module, used for standardizing the features and adding weights to the features;

[0027] The early warning analysis module is used to determine abnormal data based on the comparison between the characteristics and preset weights.

[0028] To achieve the above-mentioned purpose, a second aspect of the present application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0029] The memory stores computer-executable instructions;

[0030] The processor executes the computer-executable instructions stored in the memory to implement the system as described in any one of the first aspects.

[0031] To achieve the above-mentioned purpose, the fourth aspect embodiment of the present application proposes a computer-readable storage medium, in which computer-readable storage medium is stored computer execution instructions, and when the computer execution instructions are executed by a processor, they are used to implement a system as described in any one of the first aspects.

[0032] To achieve the above-mentioned purpose, the fifth aspect of the present application provides a computer program product, which, when executed by a processor, implements the system described in any one of the first aspects.

[0033] The AI-based income security monitoring and early warning system, device, electronic device and storage medium provided in this application realize timely analysis and early warning of abnormal business data by automatically accessing the business data of each business system to the data monitoring module and decision-making model, thereby improving the efficiency and accuracy of monitoring and early warning.

[0034] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0036] Figure 1 A schematic diagram of the structure of an AI-based income security monitoring and early warning system provided in an embodiment of the present application;

[0037] Figure 2 A machine learning schematic diagram provided for an embodiment of the present application;

[0038] Figure 3 A feature classification schematic diagram provided for an embodiment of the present application. DETAILED DESCRIPTION

[0039] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0040] Artificial Intelligence (AI) is a discipline that studies and develops theories, methods, technologies and applications for simulating, extending and expanding human intelligence. It aims to enable computer systems to perform complex tasks similar to human intelligence, such as perception, learning, reasoning, decision-making and natural language communication. The development of artificial intelligence involves multiple disciplines, including computer science, psychology, philosophy, neuroscience, mathematics and economics. Artificial intelligence is a discipline that studies the use of computers to simulate certain human thinking processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). It mainly includes the principles of computer intelligence, the manufacture of computers similar to human brain intelligence, and the realization of higher-level applications of computers. Artificial intelligence will involve disciplines such as computer science, psychology, philosophy and linguistics. It can be said that it is almost all disciplines of natural sciences and social sciences, and its scope has far exceeded the scope of computer science. The relationship between artificial intelligence and thinking science is the relationship between practice and theory. Artificial intelligence is at the technical application level of thinking science and is one of its application branches. From the perspective of thinking, artificial intelligence is not limited to logical thinking. It is necessary to consider figurative thinking and inspirational thinking to promote the breakthrough development of artificial intelligence. Mathematics is often considered to be the basic science of many disciplines. Mathematics has also entered the fields of language and thinking. The artificial intelligence discipline must also borrow mathematical tools. Mathematics not only plays a role in standard logic, fuzzy mathematics, etc., but also enters the artificial intelligence discipline. They will promote each other and develop faster.

[0041] A method for achieving billing revenue guarantee through auditing is proposed in the related technology. Through BOSS system monitoring and analysis, audit information is extracted from each link of the billing processing flow from collection to accounting. The method is characterized by: including three aspects of audit: audit of network / billing original data integrity, end-to-end balance audit of billing process and special audit of multi-network element billing data consistency. The audit of network / billing original data integrity is to check and monitor the continuity of original call record serial number and audit the original call record file. The end-to-end balance audit of billing process is to check the balance between steps in the whole billing process. The special audit of multi-network element billing data consistency is to verify the consistency of billing call records generated between different network elements of the switching network. The present invention realizes: the function of checking and monitoring the continuity of original call record serial number; the function of auditing the preprocessing module; the function of auditing the balance of all billing services, which simplifies the means of discovering call record loss.

[0042] The above scheme can check and monitor the continuity of the original call record serial number; audit the pre-processing module; and audit the balance of the entire billing business. Through BOSS system monitoring and analysis, audit information is extracted from each link of the billing processing flow from collection to accounting. It is characterized by: including three aspects of audit: network / billing original data integrity audit, billing process end-to-end balance audit and multi-network element billing data consistency special audit. The network / billing original data integrity audit is to check and monitor the continuity of the original call record serial number and audit the original call record file. The end-to-end balance audit of the billing process is to audit the balance between steps in the entire billing process. The multi-network element billing data consistency special audit is to verify the consistency of the billing call records generated between different network elements in the switching network.

[0043] However, this implementation method is not suitable in today's rapidly developing mobile business. Taking traffic revenue assurance as an example, due to the huge increase in business volume, the huge number of daily call orders, and the increasing amount of user data and customer data, the conventional audit system comparison method involves too much manual participation and is easily affected by personal subjective consciousness. There are problems such as large amounts of data that are difficult to manage, time-consuming, difficult to locate problems, and insufficient flexibility in setting monitoring and warning thresholds.

[0044] This proposal is a revenue security monitoring and early warning technology based on AI. It includes the following steps: first, set up an AI-based revenue security monitoring and early warning intelligent system, then build a revenue security cloud database and cloud file service system respectively, then access the data files of each business system through big data technology, and complete the reading, cleaning, analysis, and storage work. Finally, based on a large amount of data, intelligent algorithms, and machine learning, train artificial intelligence to analyze and monitor data, autonomously adjust the early warning threshold, and push alarms to users when problematic data is found.

[0045] The main technical problem to be solved by this proposal is: using big data + intelligent algorithms to give the revenue security system intelligent capabilities, so that monitoring and early warning do not require human intervention throughout the process. The system can conduct autonomous learning based on historical alarm data, and analysis rules, data reports, alarm thresholds, and alarm templates can be generated autonomously. It can analyze revenue differences from all aspects of "horizontally" - "vertically" - "point", effectively and quickly locate the causes of audit differences, improve the audit efficiency of large call records, effectively reduce revenue losses, and safeguard the billing of revenue security services, providing real-time, accurate, reliable, and refined nuclear security measures.

[0046] The embodiment of the present application provides an income security monitoring and early warning system based on AI artificial intelligence. Figure 1 This is a schematic diagram of the structure of an AI-based income security monitoring and early warning system provided in an embodiment of the present application. Figure 1 As shown, the system includes:

[0047] Related income protection supports cloud database 10, data monitoring module 30, big data collection module 20, data decision module 40, data early warning module 50;

[0048] The big data collection module 20 is connected to the revenue-related business system, and the big data collection module 20 is used to collect business data of the revenue-related business system, pre-process the business data, and synchronize the business data to the relevant revenue protection support cloud database 10;

[0049] The data monitoring module 30 is connected to the relevant revenue protection support cloud database 10, and the data monitoring module 30 is used to monitor the relevant revenue protection support cloud database 10, select training data from the business data and train multiple decision models;

[0050] The data decision module 40 is connected to the data monitoring module 30, and the data decision module 40 is used to process the input business data using the trained decision model to determine abnormal data, and input the abnormal data into the data early warning module 50;

[0051] The data warning module 50 is connected to the data decision module 40 and the relevant income protection support cloud database 10. The data warning module 50 is used to generate corresponding warning information according to the abnormal data, and upload the abnormal data and the warning information to the relevant income protection support cloud database 10.

[0052] Optionally, the big data acquisition module 20 includes:

[0053] A text extraction module 21, used to identify the file type of the business data and extract keywords from the file title and metadata;

[0054] The parsing module 22 is used to parse the business data and convert the business data into structured data or unstructured data;

[0055] A data cleaning module 23, used to retrieve missing data and erroneous data in the business data, and correct the missing data and erroneous data;

[0056] The data output module 24 is used to input the cleaned business data into the relevant revenue protection support cloud database 10.

[0057] In this embodiment, the text extraction module 21 uses text mining and machine learning algorithms to automatically identify the file type, and uses natural language processing (NLP) technology to extract keywords from the file title and metadata to assist in classification and achieve intelligent classification of files. Keywords related to billing data services include but are not limited to the following aspects:

[0058] Billing system: It is composed of modules such as call record collection, preprocessing, standard batch pricing, preferential batch pricing, high-price control, monitoring, parameter management, and interfaces. For mobile billing systems, it also includes a roaming processing module.

[0059] Billing mode: including prepaid and postpaid, as well as fixed ratio, fixed amount, interval range and other billing modes.

[0060] Call detail record (CDR): Call detail record, which records detailed information during the communication process, such as user ID, calling number, called number, start time, end time, call duration, Internet traffic, etc.

[0061] Online billing: A real-time interactive billing method applicable to prepaid services. It monitors user accounts in real time through the Online Billing System (OCS).

[0062] Offline billing: a non-real-time billing method that periodically uploads data to the billing system for settlement, suitable for postpaid services.

[0063] Converged billing: Combines the functions of online billing and offline billing, supports a unified service-oriented interface, and adapts to the billing needs of the 5G era.

[0064] Billing engine: A system that executes billing rules and generates billing results.

[0065] The parsing module 22 uses the distributed computing framework Spark to parse and process the file data in parallel, and uses Apache Avro to parse semi-structured or unstructured data. Apache Spark is a powerful distributed computing framework that can efficiently process various data types including semi-structured and unstructured data. Spark can import data in a variety of ways, such as using the sc.textFile method to read text files, or using the Spark SQL data source API to read files in formats such as JSON and CSV. These data sources may contain semi-structured or unstructured data. For unstructured data, such as text or images, Spark can be combined with machine learning libraries (such as MLlib) or deep learning frameworks (such as TensorFlow) to extract useful features. For example, the TF-IDF algorithm can be used to extract features of text data, or a pre-trained deep learning model can be used to extract image features. Spark SQL allows users to convert semi-structured data (such as JSON) into a structured table format (such as DataFrame), so that SQL statements can be used for query and analysis.

[0066] The data cleaning module 23 uses MapReduce big data processing tools to clean and convert data, and adopts Apache Kafka stream processing technology to process real-time data streams to ensure timely data processing.

[0067] MapReduce is a programming model for distributed processing of large-scale data sets. It consists of two main functions: Map and Reduce, and a framework to manage the scheduling and execution of tasks. Among big data processing tools, MapReduce is often used for data cleaning and transformation. In the Map phase, the input data is split into multiple independent fragments, each of which is processed concurrently by the Map function. The Map function reads the input data and performs data cleaning and transformation operations. This may include removing invalid or incomplete records, correcting erroneous or inconsistent data, converting data formats, extracting useful fields, etc.

[0068] The data output module 24 uses Apache HBase technology to store the cleaned and converted data in a distributed cloud database. Apache ZooKeeper is used to monitor the file data processing status at all times to prevent missing files or duplicate data entry. The ELK Stack log management tool is used to collect and analyze logs during the data processing process, and automatically alarm when a failure occurs. The Apache Oozie job scheduler is used to optimize system resource allocation and task execution order, use machine learning algorithms to predict the time and resources required for data processing, dynamically adjust task scheduling, and make full use of server resources to achieve efficient data collection, cleaning, and warehousing.

[0069] Optionally, the data monitoring module 30 includes:

[0070] A data change monitoring module 31 is used to monitor the relevant revenue protection support cloud database 10, and when new business data is added to the relevant revenue protection support cloud database 10, the newly added business data is synchronized to the training module in the form of a file;

[0071] A training module 32, used to mark the labels of the training data and input them into the decision model to train the decision model, wherein the decision model includes a supervised learning model and an unsupervised learning model, and the labels are types corresponding to the training data;

[0072] The reinforcement learning module 33 is used to learn the optimal behavior strategy in the interaction with the environment.

[0073] Figure 2 A machine learning schematic diagram provided in an embodiment of the present application, such as Figure 2 As shown, in this embodiment, the training module 32 uses supervised learning algorithms: support vector machine (SVM), decision tree, neural network, etc., to learn and predict from labeled data. In linear regression, the model can be expressed as y=wx+b, where y is the target variable, x is the input feature, w is the weight, and b is the bias. The model learns the best weight and bias parameters by minimizing the loss function. Unsupervised learning algorithms are used in the data decision system: support clustering algorithms (K-means, hierarchical clustering), association rule learning, which are used to discover patterns and structures from unlabeled data.

[0074] In the K-means clustering algorithm, the goal is to minimize the sum of the distances between the data points and the centers of the clusters to which they belong. The clustering structure of the data is learned by iteratively updating the cluster centers.

[0075] The reinforcement learning module 33 uses reinforcement learning algorithms: supports Q-learning, deep reinforcement learning, etc., to learn the optimal behavior strategy in the interaction with the environment. In the Q-learning algorithm, the optimal action strategy is learned by updating the Q value function. The update formula of the Q value function is:

[0076] Q(s,a)=(1-α)*Q(s,a)+α*(r+γ*max(Q(s',a'))), where s is the state, a is the action, r is the reward, α is the learning rate, and γ is the discount factor.

[0077] Optionally, the data decision module 40 includes:

[0078] A feature extraction module 41 is used to extract features of the business data according to the trained decision model;

[0079] A feature analysis module 42, used to standardize the features and add weights to the features;

[0080] The early warning analysis module 43 is used to determine abnormal data based on the comparison between the characteristics and preset weights.

[0081] Figure 3 A feature classification diagram provided in an embodiment of the present application, such as Figure 3 As shown, in this embodiment, the features of the data are extracted in the feature extraction module 41, and multiple groups are formed according to the features to form a feature analysis model. The formula for data feature extraction is set as:

[0082] (Fea 01 ,Fea 02 ,Fea 03 ,Fea 03 ...Fea 0n );

[0083] The feature analysis module 42 then extracts the feature points (Fea 01 ,Fea 02 ,Fea 03 ,Fea 03 ...Fea 0n ) is standardized and at the feature point (Fea 01 ,Fea 02 ,Fea 03 ,Fea 03 ...Fea 0n ) and add the weights to get

[0084] Then the early warning analysis module 43 updates and determines a characteristic point threshold x; finally, if Then the income data needs warning, otherwise No warning is required.

[0085] The embodiment of the present application introduces big data capabilities in the field of revenue security data processing. It automatically connects the business data related to the revenue of each business system to the revenue security monitoring and early warning system, and stores it in big data Hadoop and HBASE after pre-processing, providing data support for artificial intelligence training.

[0086] The embodiment of the present application introduces a machine learning algorithm in the field of income security data analysis, and performs machine learning. After a certain period of learning, the decision model can perform intelligent analysis and monitoring of the data. If abnormal levels of data are found in the data, the data will be pushed to the data early warning module.

[0087] In order to implement the above embodiments, the present application also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the system provided by the above embodiments.

[0088] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the system provided by the above embodiments.

[0089] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the system provided by the above embodiments when executed by a processor.

[0090] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0091] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.

[0092] The present application is expected to provide an implementation scheme for users to selectively block the use or access of personal information data. That is, the present disclosure is expected to provide hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, the risk can be minimized by limiting data collection and deleting the data. In addition, when applicable, such personal information is de-identified to protect the privacy of the user.

[0093] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0094] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0095] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0096] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0097] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0098] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0099] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0100] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. An income security monitoring and early warning system based on AI artificial intelligence, characterized in that: include: Related income protection supports cloud database, data monitoring module, big data collection module, data decision module, and data early warning module; The big data collection module is connected to the revenue-related business system, and is used to collect business data of the revenue-related business system, pre-process the business data, and synchronize the business data to the relevant revenue protection support cloud database; The data monitoring module is connected to the relevant revenue protection support cloud database, and the data monitoring module is used to monitor the relevant revenue protection support cloud database, select training data from the business data and train multiple decision models; The data decision module is connected to the data monitoring module, and the data decision module is used to process the input business data using the trained decision model to determine abnormal data, and input the abnormal data into the data early warning module; The data warning module is connected to the data decision module and the relevant income protection support cloud database. The data warning module is used to generate corresponding warning information based on the abnormal data, and upload the abnormal data and the warning information to the relevant income protection support cloud database.

2. The system according to claim 1, characterized in that The big data acquisition module includes: A text extraction module, used to identify the file type of the business data and extract keywords from the file title and metadata; A parsing module, used to parse the business data and convert the business data into structured data or unstructured data; A data cleaning module, used to retrieve missing data and erroneous data in the business data, and correct the missing data and erroneous data; The data output module is used to input the cleaned business data into the relevant revenue protection support cloud database.

3. The system according to claim 1, characterized in that The data monitoring module comprises: A data change monitoring module, used for monitoring the relevant revenue protection support cloud database, and when new business data is added to the relevant revenue protection support cloud database, synchronizing the newly added business data to the training module in the form of files; A training module, used to mark the labels of the training data and input them into the decision model to train the decision model, wherein the decision model includes a supervised learning model and an unsupervised learning model, and the labels are types corresponding to the training data; Reinforcement learning module, used to learn the optimal behavior strategy in interaction with the environment.

4. The system according to claim 3, characterized in that The data decision module includes: A feature extraction module, used to extract features of the business data according to the trained decision model; A feature analysis module, used for standardizing the features and adding weights to the features; The early warning analysis module is used to determine abnormal data based on the comparison between the characteristics and preset weights.

5. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the system according to any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the system according to any one of claims 1 to 4 when executed by a processor.

7. A computer program product, characterized in that The invention comprises a computer program, which implements the system according to any one of claims 1 to 4 when being executed by a processor.

Images