Operation and maintenance management method and system for business and financial collaborative self-healing system, and medium

Abstract

The embodiment of the invention provides a business and financial collaborative self-healing system operation and maintenance management method and system and a medium, and the method comprises the steps: obtaining a large amount of log data of business and financial systems through a log collection technology, extracting the features of the log data, and obtaining the data features; the data features are preprocessed, invalid data features are removed, and valid data features are obtained; analyzing the log data based on the effective data features, and inputting analysis information into a data evaluation model to obtain a data evaluation result; analyzing an abnormal state of the log data based on the data evaluation result, and judging whether the abnormal state meets requirements or not; the method comprises the following steps of: generating abnormal log data if the requirements are not met, generating operation and maintenance data if the requirements are met, performing feature extraction on the log data, screening out effective data features, and analyzing an abnormal state of the analyzed data through a data evaluation model, so that the abnormal log data in the operation and maintenance process are accurately analyzed; and flexibility and security of data operation and maintenance management are realized.

Description

Technical Field

[0001] The present application relates to the field of operation and maintenance management technology, and specifically, to a method, system and medium for operation and maintenance management of a business-finance collaborative self-healing system. Background Art

[0002] The technical specification for online monitoring of abnormal logs in business and finance collaboration is guided by the overall plan of the "Research and Application Project on Automatic Repair Technology of Business and Finance Collaboration Problems Based on Artificial Intelligence". According to the construction principles of practicality, advancement, scalability, integration, security, economy and standardization, it defines the system architecture, system functions, key technologies, logical models, external interfaces, operation and maintenance management, and technical requirements of the online monitoring system for abnormal logs in business and finance collaboration, and provides technical support for the research and application of online monitoring technology for abnormal logs in business and finance collaboration. The existing operation and maintenance management methods of business and finance collaboration self-healing systems cannot parse and evaluate the data based on the effective data features of log data, making it difficult to accurately screen out abnormal log data in the operation and maintenance process, reducing the accuracy of operation and maintenance data monitoring, and causing large operation and maintenance management errors. Effective technical solutions are urgently needed for the above problems. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a business-finance collaborative self-healing system operation and maintenance management method, system and medium, by extracting features from log data, screening out valid data features, parsing the valid data features, and analyzing the abnormal state of the parsed data through a data evaluation model, thereby accurately analyzing the abnormal log data in the operation and maintenance process, and achieving flexibility and security in data operation and maintenance management.

[0004] The embodiment of the present application also provides a business-finance collaborative self-healing system operation and maintenance management method, including:

[0005] Obtain a large amount of log data from business and financial systems through log collection technology, extract the features of the log data, and obtain data features;

[0006] Preprocess the data features, remove invalid data features, and obtain valid data features;

[0007] Parse the log data based on the effective data features to obtain parsed information, input the parsed information into the data evaluation model to obtain the data evaluation results;

[0008] Analyze the abnormal state of the log data based on the data evaluation result, and determine whether the abnormal state meets the requirements;

[0009] If the requirements are not met, abnormal log data is generated and marked;

[0010] If the requirements are met, operation and maintenance data will be generated and transmitted to the terminal in real time.

[0011] Optionally, in the business-finance collaborative self-healing system operation and maintenance management method described in the embodiment of the present application, a large amount of log data of the business and financial systems is obtained through log collection technology, and features of the log data are extracted to obtain data features, which specifically include:

[0012] Based on the collection types of several sensors, corresponding types of collection information are generated, and different types of collection data are generated;

[0013] Obtain different types of collected data, extract features for each type of collected data, and generate log data features corresponding to different types of sensors;

[0014] Analyze the data similarity of log data features corresponding to different types of sensors;

[0015] Adjust the data type of log data features based on data similarity to generate the final log data features;

[0016] The final log data features of different types are fused to generate data features.

[0017] Optionally, in the business-finance collaborative self-healing system operation and maintenance management method described in the embodiment of the present application, the data features are preprocessed to remove invalid data features to obtain valid data features, specifically including:

[0018] Acquire data features, analyze abnormal states of data features, and analyze duplicate information, missing information, and abnormal information of data features based on abnormal states;

[0019] Based on the repetition information of the data feature, it is determined whether the repetition degree of the data feature meets the set interval. If so, a first valid data feature is obtained. If not, the repetition features of the data feature are screened out, and the repetition features are deduplicated based on the repetition features.

[0020] Based on the missing information of the data feature, determine whether the missing data feature affects the semantic information of the data. If so, generate compensation information, and fill the missing data feature based on the compensation information. If not, obtain a second valid data feature.

[0021] Analyze the abnormal data features based on the abnormal information of the data features, remove the abnormal data features, and obtain the third effective data features;

[0022] The first valid data feature, the second valid data feature and the third valid data feature are integrated to obtain a valid data feature.

[0023] Optionally, in the business-finance collaborative self-healing system operation and maintenance management method described in the embodiment of the present application, log data is parsed based on effective data features to obtain parsed information, and the parsed information is input into a data evaluation model to obtain a data evaluation result, specifically including:

[0024] Obtain log data, and use natural language processing technology to perform word segmentation, part-of-speech tagging, and named entity recognition on the text information in the log data to obtain processing results;

[0025] Analyze the semantic state of the data text based on the processing result, convert the unstructured text into structured data based on the semantic state, and generate effective data features, wherein the effective data features include timestamp, keyword or word frequency features;

[0026] Input valid data features into the parsing model, output parsing data, analyze the word meaning based on the parsing data, and generate parsing information;

[0027] The parsed information is evaluated based on the data evaluation model to obtain the data evaluation result.

[0028] Optionally, in the business-finance collaborative self-healing system operation and maintenance management method described in the embodiment of the present application, the data evaluation model construction and training method includes:

[0029] Obtain historical log data, generate a training set, input the training set into the initial model for iterative training, and generate training results;

[0030] Determine whether the training results have converged. If so, input the parsing information into the data evaluation model to evaluate the accuracy of the word meaning of the data and obtain the evaluation results;

[0031] Determine whether the evaluation result meets the accuracy requirement, and if so, generate the data evaluation result;

[0032] If the accuracy requirements are not met, the parameters of the data evaluation model are adjusted.

[0033] If not, then generate adjustment information, adjust the training data in the training set based on the adjustment information, generate a new training set, and perform secondary training on the initial model based on the new training set until the model converges.

[0034] Optionally, in the business-finance collaborative self-healing system operation and maintenance management method described in the embodiment of the present application, analyzing the abnormal state of the log data based on the data evaluation result and judging whether the abnormal state meets the requirements specifically includes:

[0035] Obtaining data evaluation results, comparing the data evaluation results with set evaluation standard information, and obtaining a data state deviation rate;

[0036] Determine whether the data state deviation rate is greater than or equal to a set state deviation rate threshold;

[0037] If it is greater than or equal to, abnormal log data is generated;

[0038] If it is less than, it is determined that the evaluation result of the log data meets the requirements.

[0039] In a second aspect, an embodiment of the present application provides a business-finance collaborative self-healing system operation and maintenance management system, the system comprising: a memory and a processor, the memory comprising a program of a business-finance collaborative self-healing system operation and maintenance management method, and when the program of the business-finance collaborative self-healing system operation and maintenance management method is executed by the processor, the following steps are implemented:

[0040] Obtain a large amount of log data from business and financial systems through log collection technology, extract the features of the log data, and obtain data features;

[0041] Preprocess the data features, remove invalid data features, and obtain valid data features;

[0042] Parse the log data based on the effective data features to obtain parsed information, input the parsed information into the data evaluation model to obtain the data evaluation results;

[0043] Analyze the abnormal state of the log data based on the data evaluation result, and determine whether the abnormal state meets the requirements;

[0044] If the requirements are not met, abnormal log data is generated and marked;

[0045] If the requirements are met, operation and maintenance data will be generated and transmitted to the terminal in real time.

[0046] Optionally, in the business-finance collaborative self-healing system operation and maintenance management system described in the embodiment of the present application, a large amount of log data of the business and financial systems is obtained through log collection technology, and features of the log data are extracted to obtain data features, specifically including:

[0047] Based on the collection types of several sensors, corresponding types of collection information are generated, and different types of collection data are generated;

[0048] Obtain different types of collected data, extract features for each type of collected data, and generate log data features corresponding to different types of sensors;

[0049] Analyze the data similarity of log data features corresponding to different types of sensors;

[0050] Adjust the data type of log data features based on data similarity to generate the final log data features;

[0051] The final log data features of different types are fused to generate data features.

[0052] Optionally, in the business-finance collaborative self-healing system operation and maintenance management system described in the embodiment of the present application, data features are preprocessed to remove invalid data features to obtain valid data features, specifically including:

[0053] Acquire data features, analyze abnormal states of data features, and analyze duplicate information, missing information, and abnormal information of data features based on abnormal states;

[0054] Based on the repetition information of the data feature, it is determined whether the repetition degree of the data feature meets the set interval. If so, a first valid data feature is obtained. If not, the repetition features of the data feature are screened out, and the repetition features are deduplicated based on the repetition features.

[0055] Based on the missing information of the data feature, determine whether the missing data feature affects the semantic information of the data. If so, generate compensation information, and fill the missing data feature based on the compensation information. If not, obtain a second valid data feature.

[0056] Analyze the abnormal data features based on the abnormal information of the data features, remove the abnormal data features, and obtain the third effective data features;

[0057] The first valid data feature, the second valid data feature and the third valid data feature are integrated to obtain a valid data feature.

[0058] In the third aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a business-finance collaborative self-healing system operation and maintenance management method program, and when the business-finance collaborative self-healing system operation and maintenance management method program is executed by a processor, the steps of the business-finance collaborative self-healing system operation and maintenance management method as described in any one of the above items are implemented.

[0059] As can be seen from the above, the embodiment of the present application provides a business-finance collaborative self-healing system operation and maintenance management method, system and medium, which obtain a large amount of log data of business and financial systems through log collection technology, extract the characteristics of the log data, and obtain data characteristics; pre-process the data characteristics, eliminate invalid data characteristics, and obtain valid data characteristics; parse the log data based on the valid data characteristics to obtain parsing information, input the parsing information into the data evaluation model, and obtain data evaluation results; analyze the abnormal state of the log data based on the data evaluation results, and judge whether the abnormal state meets the requirements; if the requirements are not met, generate abnormal log data, and mark the abnormal log data; if the requirements are met, generate operation and maintenance data, and transmit the operation and maintenance data to the terminal in real time; extract features from the log data, screen out valid data characteristics, parse the valid data characteristics, and analyze the abnormal state of the parsed data through the data evaluation model, so as to accurately analyze the abnormal log data in the operation and maintenance process, and realize the flexibility and security of data operation and maintenance management. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0061] Figure 1 A flowchart of the business-finance collaborative self-healing system operation and maintenance management method provided in an embodiment of the present application;

[0062] Figure 2 A flow chart of a method for acquiring data features of a business-finance collaborative self-healing system operation and maintenance management method provided in an embodiment of the present application;

[0063] Figure 3 A flow chart of a method for acquiring effective data features for the business-finance collaborative self-healing system operation and maintenance management method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0065] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0066] Please refer to Figure 1 , Figure 1 This is a flow chart of a business-finance collaborative self-healing system operation and maintenance management method in some embodiments of the present application. The business-finance collaborative self-healing system operation and maintenance management method is used in a terminal device, and the business-finance collaborative self-healing system operation and maintenance management method includes the following steps:

[0067] S101, obtaining a large amount of log data of business and financial systems through log collection technology, extracting features of the log data, and obtaining data features;

[0068] S102, preprocessing the data features, eliminating invalid data features, and obtaining valid data features;

[0069] S103, parsing the log data based on the valid data features to obtain parsing information, and inputting the parsing information into a data evaluation model to obtain a data evaluation result;

[0070] S104, analyzing the abnormal state of the log data based on the data evaluation result, and determining whether the abnormal state meets the requirements;

[0071] S105, if the requirements are not met, then abnormal log data is generated and the abnormal log data is marked; if the requirements are met, then operation and maintenance data is generated and the operation and maintenance data is transmitted to the terminal in real time.

[0072] It should be noted that by extracting data features and processing them, invalid data features are eliminated and the reflection accuracy of log data is improved. In addition, log data is evaluated through a data evaluation model to analyze the abnormal state of the log data, thereby marking the abnormal data and improving the analysis accuracy of abnormal data in the operation and maintenance management process.

[0073] Please refer to Figure 2 , Figure 2 This is a flow chart of a data feature acquisition method of a business-finance collaborative self-healing system operation and maintenance management method in some embodiments of the present application. According to an embodiment of the present invention, a large amount of log data of business and financial systems is acquired through log collection technology, and the features of the log data are extracted to obtain data features, which specifically include:

[0074] S201, based on the collection types of several sensors, generating corresponding types of collection information and generating different types of collection data;

[0075] S202, acquiring different types of collected data, performing feature extraction on the different types of collected data respectively, and generating log data features corresponding to different types of sensors;

[0076] S203, analyzing the data similarity of log data features corresponding to different types of sensors;

[0077] S204, adjusting the data type of the log data feature based on the data similarity to generate a final log data feature;

[0078] S205: Fusion different types of final log data features to generate data features.

[0079] It should be noted that different types of data are collected through different types of sensors to obtain log data of different latitudes. The data types are integrated and classified by analyzing the data similarity to improve the classification accuracy. The final log data is fused to obtain data features to ensure the validity of the data features.

[0080] Please refer to Figure 3 , Figure 3 This is a flow chart of a method for obtaining effective data features of a business-finance collaborative self-healing system operation and maintenance management method in some embodiments of the present application. According to an embodiment of the present invention, data features are preprocessed to remove invalid data features to obtain effective data features, specifically including:

[0081] S301, acquiring data features, analyzing abnormal states of the data features, and analyzing repeated information, missing information, and abnormal information of the data features according to the abnormal states;

[0082] S302, judging whether the repetition degree of the data feature meets the set interval based on the repetition information of the data feature, if so, obtaining the first valid data feature, if not, filtering out the repetitive features of the data feature, and performing deduplication processing on the repetitive data features based on the repetitive features;

[0083] S303, judging whether the missing data feature affects the semantic information of the data based on the missing information of the data feature, if so, generating compensation information, and filling the missing data feature based on the compensation information, if not, obtaining a second valid data feature;

[0084] S304, analyzing the abnormal data features based on the abnormal information of the data features, removing the abnormal data features, and obtaining a third valid data feature;

[0085] S305, integrating the first valid data feature, the second valid data feature and the third valid data feature to obtain a valid data feature.

[0086] It should be noted that by analyzing the missing information, abnormal information and repeated information of data features, the abnormal data is processed in different ways according to different data states, so as to obtain different valid data features, and the different valid data features are fused to obtain accurate valid data features.

[0087] According to an embodiment of the present invention, log data is parsed based on valid data features to obtain parsed information, and the parsed information is input into a data evaluation model to obtain a data evaluation result, which specifically includes:

[0088] Obtain log data, and use natural language processing technology to perform word segmentation, part-of-speech tagging, and named entity recognition on the text information in the log data to obtain processing results;

[0089] Analyze the semantic state of the data text based on the processing results, convert the unstructured text into structured data based on the semantic state, and generate effective data features, which include timestamps, keywords or word frequency features;

[0090] Input valid data features into the parsing model, output parsing data, analyze the word meaning based on the parsing data, and generate parsing information;

[0091] The parsed information is evaluated based on the data evaluation model to obtain the data evaluation result.

[0092] It should be noted that the data is analyzed by analyzing the semantic state of the data text to obtain valid data features, and the valid data features are input into the parsing model to parse the log data to improve the accuracy of data evaluation.

[0093] According to an embodiment of the present invention, a method for constructing and training a data evaluation model includes:

[0094] Obtain historical log data, generate a training set, input the training set into the initial model for iterative training, and generate training results;

[0095] Determine whether the training results have converged. If so, input the parsing information into the data evaluation model to evaluate the accuracy of the word meaning of the data and obtain the evaluation results;

[0096] Determine whether the evaluation result meets the accuracy requirement, and if so, generate the data evaluation result;

[0097] If the accuracy requirements are not met, the parameters of the data evaluation model are adjusted.

[0098] It should be noted that by continuously training the model through historical log data, an accurate data evaluation model can be obtained, the output accuracy of the data evaluation model can be improved, and the output results of the data evaluation model can be ensured to be close to the actual results.

[0099] If not, then generate adjustment information, adjust the training data in the training set based on the adjustment information, generate a new training set, and perform secondary training on the initial model based on the new training set until the model converges.

[0100] According to an embodiment of the present invention, analyzing the abnormal state of the log data based on the data evaluation result and determining whether the abnormal state meets the requirements specifically includes:

[0101] Obtaining data evaluation results, comparing the data evaluation results with set evaluation standard information, and obtaining a data state deviation rate;

[0102] Determine whether the data state deviation rate is greater than or equal to a set state deviation rate threshold;

[0103] If it is greater than or equal to, abnormal log data is generated;

[0104] If it is less than, it is determined that the evaluation result of the log data meets the requirements.

[0105] It should be noted that by analyzing the deviation of the data evaluation results, the log data is analyzed for abnormality, abnormal log data is obtained, and the judgment of the abnormal state of the log data is improved.

[0106] According to an embodiment of the present invention, it also includes: acquiring log data, analyzing the log data based on the analysis rule, generating an abnormal log, and obtaining abnormal information;

[0107] The abnormal information is distinguished and classified by a predefined or trained classification model, and key features are extracted from the abnormal log data, wherein the key features include error codes, error messages, involved service modules or operations that trigger the abnormality;

[0108] Standardize or normalize the feature data, train the classification model based on the abnormal log sample data, and optimize the model parameters;

[0109] Based on the classification model, the model can accurately classify the newly generated abnormal logs and generate new abnormal log data;

[0110] Based on the data type of the abnormal log data analysis, the category of the abnormal log data is obtained.

[0111] It should be noted that by analyzing the abnormal information in the log data, the key features in the log data are extracted, and the feature data is standardized, so that the abnormal log samples can be accurately classified.

[0112] In a second aspect, an embodiment of the present application provides an operation and maintenance management system for a business-finance collaborative self-healing system, the system comprising: a memory and a processor, the memory comprising a program of an operation and maintenance management method for a business-finance collaborative self-healing system, and when the program of the operation and maintenance management method for a business-finance collaborative self-healing system is executed by the processor, the following steps are implemented:

[0113] Obtain a large amount of log data from business and financial systems through log collection technology, extract the features of the log data, and obtain data features;

[0114] Preprocess the data features, remove invalid data features, and obtain valid data features;

[0115] Parse the log data based on the effective data features to obtain parsed information, input the parsed information into the data evaluation model to obtain the data evaluation results;

[0116] Analyze the abnormal status of log data based on the data evaluation results and determine whether the abnormal status meets the requirements;

[0117] If the requirements are not met, abnormal log data is generated and marked;

[0118] If the requirements are met, operation and maintenance data will be generated and transmitted to the terminal in real time.

[0119] It should be noted that by extracting data features and processing them, invalid data features are eliminated and the reflection accuracy of log data is improved. In addition, log data is evaluated through a data evaluation model to analyze the abnormal state of the log data, thereby marking the abnormal data and improving the analysis accuracy of abnormal data in the operation and maintenance management process.

[0120] According to an embodiment of the present invention, a large amount of log data of business and financial systems is obtained through log collection technology, and features of the log data are extracted to obtain data features, which specifically include:

[0121] Based on the collection types of several sensors, corresponding types of collection information are generated, and different types of collection data are generated;

[0122] Obtain different types of collected data, extract features for each type of collected data, and generate log data features corresponding to different types of sensors;

[0123] Analyze the data similarity of log data features corresponding to different types of sensors;

[0124] Adjust the data type of log data features based on data similarity to generate the final log data features;

[0125] The final log data features of different types are fused to generate data features.

[0126] It should be noted that different types of data are collected through different types of sensors to obtain log data of different latitudes. The data types are integrated and classified by analyzing the data similarity to improve the classification accuracy. The final log data is fused to obtain data features to ensure the validity of the data features.

[0127] According to an embodiment of the present invention, data features are preprocessed to remove invalid data features and obtain valid data features, which specifically includes:

[0128] Acquire data features, analyze abnormal states of data features, and analyze duplicate information, missing information, and abnormal information of data features based on abnormal states;

[0129] Based on the repetition information of the data feature, it is determined whether the repetition degree of the data feature meets the set interval. If so, a first valid data feature is obtained. If not, the repetition features of the data feature are screened out, and the repetition features are deduplicated based on the repetition features.

[0130] Based on the missing information of the data feature, determine whether the missing data feature affects the semantic information of the data. If so, generate compensation information, and fill the missing data feature based on the compensation information. If not, obtain a second valid data feature.

[0131] Analyze the abnormal data features based on the abnormal information of the data features, remove the abnormal data features, and obtain the third effective data features;

[0132] The first valid data feature, the second valid data feature and the third valid data feature are integrated to obtain a valid data feature.

[0133] It should be noted that by analyzing the missing information, abnormal information and repeated information of data features, the abnormal data is processed in different ways according to different data states, so as to obtain different valid data features, and the different valid data features are fused to obtain accurate valid data features.

[0134] According to an embodiment of the present invention, log data is parsed based on valid data features to obtain parsed information, and the parsed information is input into a data evaluation model to obtain a data evaluation result, which specifically includes:

[0135] Obtain log data, and use natural language processing technology to perform word segmentation, part-of-speech tagging, and named entity recognition on the text information in the log data to obtain processing results;

[0136] Analyze the semantic state of the data text based on the processing results, convert the unstructured text into structured data based on the semantic state, and generate effective data features, which include timestamps, keywords or word frequency features;

[0137] Input valid data features into the parsing model, output parsing data, analyze the word meaning based on the parsing data, and generate parsing information;

[0138] The parsed information is evaluated based on the data evaluation model to obtain the data evaluation result.

[0139] It should be noted that the data is analyzed by analyzing the semantic state of the data text to obtain valid data features, and the valid data features are input into the parsing model to parse the log data to improve the accuracy of data evaluation.

[0140] According to an embodiment of the present invention, a method for constructing and training a data evaluation model includes:

[0141] Obtain historical log data, generate a training set, input the training set into the initial model for iterative training, and generate training results;

[0142] Determine whether the training results have converged. If so, input the parsing information into the data evaluation model to evaluate the accuracy of the word meaning of the data and obtain the evaluation results;

[0143] Determine whether the evaluation result meets the accuracy requirement, and if so, generate the data evaluation result;

[0144] If the accuracy requirements are not met, the parameters of the data evaluation model are adjusted.

[0145] It should be noted that by continuously training the model through historical log data, an accurate data evaluation model can be obtained, the output accuracy of the data evaluation model can be improved, and the output results of the data evaluation model can be ensured to be close to the actual results.

[0146] If not, then generate adjustment information, adjust the training data in the training set based on the adjustment information, generate a new training set, and perform secondary training on the initial model based on the new training set until the model converges.

[0147] According to an embodiment of the present invention, analyzing the abnormal state of the log data based on the data evaluation result and determining whether the abnormal state meets the requirements specifically includes:

[0148] Obtaining data evaluation results, comparing the data evaluation results with set evaluation standard information, and obtaining a data state deviation rate;

[0149] Determine whether the data state deviation rate is greater than or equal to a set state deviation rate threshold;

[0150] If it is greater than or equal to, abnormal log data is generated;

[0151] If it is less than, it is determined that the evaluation result of the log data meets the requirements.

[0152] It should be noted that by analyzing the deviation of the data evaluation results, the log data is analyzed for abnormality, abnormal log data is obtained, and the judgment of the abnormal state of the log data is improved.

[0153] According to an embodiment of the present invention, it also includes: acquiring log data, analyzing the log data based on the analysis rule, generating an abnormal log, and obtaining abnormal information;

[0154] The abnormal information is distinguished and classified by a predefined or trained classification model, and key features are extracted from the abnormal log data, wherein the key features include error codes, error messages, involved service modules or operations that trigger the abnormality;

[0155] Standardize or normalize the feature data, train the classification model based on the abnormal log sample data, and optimize the model parameters;

[0156] Based on the classification model, the model can accurately classify the newly generated abnormal logs and generate new abnormal log data;

[0157] Based on the data type of the abnormal log data analysis, the category of the abnormal log data is obtained.

[0158] It should be noted that by analyzing the abnormal information in the log data, the key features in the log data are extracted, and the feature data is standardized, so that the abnormal log samples can be accurately classified.

[0159] A third aspect of the present invention provides a computer-readable storage medium, which includes a business-finance collaborative self-healing system operation and maintenance management method program. When the business-finance collaborative self-healing system operation and maintenance management method program is executed by a processor, the steps of the business-finance collaborative self-healing system operation and maintenance management method as described in any one of the above items are implemented.

[0160] The present invention discloses a business-finance collaborative self-healing system operation and maintenance management method, system and medium. The method obtains a large amount of log data of business and financial systems through log collection technology, extracts features of the log data, and obtains data features; pre-processes the data features, eliminates invalid data features, and obtains valid data features; parses the log data based on the valid data features to obtain parsing information, inputs the parsing information into a data evaluation model, and obtains data evaluation results; analyzes the abnormal state of the log data based on the data evaluation results, and determines whether the abnormal state meets the requirements; if the requirements are not met, generates abnormal log data, and marks the abnormal log data; if the requirements are met, generates operation and maintenance data, and transmits the operation and maintenance data to a terminal in real time; extracts features from the log data, screens out valid data features, parses the valid data features, and analyzes the abnormal state of the parsed data through a data evaluation model, thereby accurately analyzing the abnormal log data in the operation and maintenance process, and realizing the flexibility and security of data operation and maintenance management.

[0161] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0162] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0163] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0164] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0165] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

Claims

1. A business-finance collaborative self-healing system operation and maintenance management method, characterized in that: include: Obtain a large amount of log data from business and financial systems through log collection technology, extract the features of the log data, and obtain data features; Preprocess the data features, remove invalid data features, and obtain valid data features; Parse the log data based on the effective data features to obtain parsed information, input the parsed information into the data evaluation model to obtain the data evaluation results; Analyze the abnormal state of the log data based on the data evaluation result, and determine whether the abnormal state meets the requirements; If the requirements are not met, abnormal log data is generated and marked; If the requirements are met, operation and maintenance data will be generated and transmitted to the terminal in real time.

2. The method for operation and maintenance management of the business-finance collaborative self-healing system according to claim 1 is characterized in that: Through log collection technology, a large amount of log data from business and financial systems is obtained, and the features of the log data are extracted to obtain data features, including: Based on the collection types of several sensors, corresponding types of collection information are generated, and different types of collection data are generated; Obtain different types of collected data, extract features for each type of collected data, and generate log data features corresponding to different types of sensors; Analyze the data similarity of log data features corresponding to different types of sensors; Adjust the data type of log data features based on data similarity to generate the final log data features; The final log data features of different types are fused to generate data features.

3. The method for operation and maintenance management of the business-finance collaborative self-healing system according to claim 2 is characterized in that: Preprocess the data features, remove invalid data features, and obtain valid data features, including: Acquire data features, analyze abnormal states of data features, and analyze duplicate information, missing information, and abnormal information of data features based on abnormal states; Based on the repetition information of the data feature, it is determined whether the repetition degree of the data feature meets the set interval. If so, a first valid data feature is obtained. If not, the repetition features of the data feature are screened out, and the repetition features are deduplicated based on the repetition features. Based on the missing information of the data feature, determine whether the missing data feature affects the semantic information of the data. If so, generate compensation information, and fill the missing data feature based on the compensation information. If not, obtain a second valid data feature. Analyze the abnormal data features based on the abnormal information of the data features, remove the abnormal data features, and obtain the third effective data features; The first valid data feature, the second valid data feature and the third valid data feature are integrated to obtain a valid data feature.

4. The method for operation and maintenance management of the business-finance collaborative self-healing system according to claim 3 is characterized in that: Parse the log data based on the valid data features to obtain parsed information, input the parsed information into the data evaluation model, and obtain the data evaluation results, which specifically include: Obtain log data, and use natural language processing technology to perform word segmentation, part-of-speech tagging, and named entity recognition on the text information in the log data to obtain processing results; Analyze the semantic state of the data text based on the processing result, convert the unstructured text into structured data based on the semantic state, and generate effective data features, wherein the effective data features include timestamp, keyword or word frequency features; Input valid data features into the parsing model, output parsing data, analyze the word meaning based on the parsing data, and generate parsing information; The parsed information is evaluated based on the data evaluation model to obtain the data evaluation result.

5. The method for operation and maintenance management of the business-finance collaborative self-healing system according to claim 4 is characterized in that: The construction and training method of the data evaluation model includes: Obtain historical log data, generate a training set, input the training set into the initial model for iterative training, and generate training results; Determine whether the training results have converged. If so, input the parsing information into the data evaluation model to evaluate the accuracy of the word meaning of the data and obtain the evaluation results; Determine whether the evaluation result meets the accuracy requirement, and if so, generate the data evaluation result; If the accuracy requirements are not met, the parameters of the data evaluation model are adjusted. If not, then generate adjustment information, adjust the training data in the training set based on the adjustment information, generate a new training set, and perform secondary training on the initial model based on the new training set until the model converges.

6. The method for operation and maintenance management of the business-finance collaborative self-healing system according to claim 5 is characterized in that: Analyze the abnormal state of the log data based on the data evaluation result and determine whether the abnormal state meets the requirements, specifically including: Obtaining data evaluation results, comparing the data evaluation results with set evaluation standard information, and obtaining a data state deviation rate; Determine whether the data state deviation rate is greater than or equal to a set state deviation rate threshold; If it is greater than or equal to, abnormal log data is generated; If it is less than, it is determined that the evaluation result of the log data meets the requirements.

7. A business-finance collaborative self-healing system operation and maintenance management system, characterized in that: The system includes: a memory and a processor, wherein the memory includes a program of a business-finance collaborative self-healing system operation and maintenance management method, and when the program of the business-finance collaborative self-healing system operation and maintenance management method is executed by the processor, the following steps are implemented: Obtain a large amount of log data from business and financial systems through log collection technology, extract the features of the log data, and obtain data features; Preprocess the data features, remove invalid data features, and obtain valid data features; Parse the log data based on the effective data features to obtain parsed information, input the parsed information into the data evaluation model to obtain the data evaluation results; Analyze the abnormal state of the log data based on the data evaluation result, and determine whether the abnormal state meets the requirements; If the requirements are not met, abnormal log data is generated and marked; If the requirements are met, operation and maintenance data will be generated and transmitted to the terminal in real time.

8. The business-finance collaborative self-healing system operation and maintenance management system according to claim 7 is characterized in that: Through log collection technology, a large amount of log data from business and financial systems is obtained, and the features of the log data are extracted to obtain data features, including: Based on the collection types of several sensors, corresponding types of collection information are generated, and different types of collection data are generated; Obtain different types of collected data, extract features for each type of collected data, and generate log data features corresponding to different types of sensors; Analyze the data similarity of log data features corresponding to different types of sensors; Adjust the data type of log data features based on data similarity to generate the final log data features; The final log data features of different types are fused to generate data features.

9. The business-finance collaborative self-healing system operation and maintenance management system according to claim 8 is characterized in that: Preprocess the data features, remove invalid data features, and obtain valid data features, including: Acquire data features, analyze abnormal states of data features, and analyze duplicate information, missing information, and abnormal information of data features based on abnormal states; Based on the repetition information of the data feature, it is determined whether the repetition degree of the data feature meets the set interval. If so, a first valid data feature is obtained. If not, the repetition features of the data feature are screened out, and the repetition features are deduplicated based on the repetition features. Based on the missing information of the data feature, determine whether the missing data feature affects the semantic information of the data. If so, generate compensation information, and fill the missing data feature based on the compensation information. If not, obtain a second valid data feature. Analyze the abnormal data features based on the abnormal information of the data features, remove the abnormal data features, and obtain the third effective data features; The first valid data feature, the second valid data feature and the third valid data feature are integrated to obtain a valid data feature.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a business-finance collaborative self-healing system operation and maintenance management method program. When the business-finance collaborative self-healing system operation and maintenance management method program is executed by a processor, the steps of the business-finance collaborative self-healing system operation and maintenance management method as described in any one of claims 1 to 6 are implemented.

Images