CN117216801A - Enterprise financial data safety management system and method based on artificial intelligence - Google Patents
Enterprise financial data safety management system and method based on artificial intelligence Download PDFInfo
- Publication number
- CN117216801A CN117216801A CN202311467424.4A CN202311467424A CN117216801A CN 117216801 A CN117216801 A CN 117216801A CN 202311467424 A CN202311467424 A CN 202311467424A CN 117216801 A CN117216801 A CN 117216801A
- Authority
- CN
- China
- Prior art keywords
- data
- risk
- financial data
- level
- zhxs
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000006399 behavior Effects 0.000 claims abstract description 69
- 238000012545 processing Methods 0.000 claims abstract description 56
- 238000000605 extraction Methods 0.000 claims abstract description 44
- 238000013136 deep learning model Methods 0.000 claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 23
- 230000002159 abnormal effect Effects 0.000 claims abstract description 14
- 238000012502 risk assessment Methods 0.000 claims abstract description 13
- 238000005516 engineering process Methods 0.000 claims abstract description 8
- 238000007726 management method Methods 0.000 claims description 53
- 238000012544 monitoring process Methods 0.000 claims description 30
- 230000005856 abnormality Effects 0.000 claims description 19
- 238000004458 analytical method Methods 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 8
- 238000004140 cleaning Methods 0.000 claims description 7
- 238000012795 verification Methods 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 238000003672 processing method Methods 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000013475 authorization Methods 0.000 description 2
- 230000003542 behavioural effect Effects 0.000 description 2
- 230000010485 coping Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 230000002195 synergetic effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses an artificial intelligence-based enterprise financial data security management system and method, which relate to the technical field of enterprise financial data security management, and the invention provides an artificial intelligence-based enterprise financial data security management system and method, wherein after collection samples of enterprise financial data are collected for processing and feature extraction, a deep learning model is built, and training is carried out, a comprehensive classification coefficient Zhxs is obtained and compared with a standard threshold value, risk levels are judged, corresponding encryption processing is carried out, different security measures are adopted for risk data of different levels, and unauthorized access and data leakage are prevented; by establishing a deep learning model and analyzing the behavior mode of the encrypted risk data, abnormal activities and access intrusion behaviors can be monitored in real time. The risk behaviors are identified by utilizing an artificial intelligence technology, so that the problems that false alarms and false alarms may exist in abnormal detection and risk assessment in the financial data safety management system are reduced.
Description
Technical Field
The invention relates to the technical field of enterprise financial security management, in particular to an artificial intelligence-based enterprise financial data security management system and method.
Background
The existence of the enterprise financial data safety management system is used for solving the financial data safety and management challenges faced by enterprises, and providing a comprehensive and effective solution. Financial data is one of the most important, most sensitive assets of an enterprise. Protecting the security of financial data is critical to the sustainable development of an enterprise.
The existing enterprise financial data safety management system simply uploads the financial data of an enterprise to the system, and then transmits, accesses and modifies some authorities of management personnel, but a large amount of financial data is manually encrypted in the process of managing, submitting, storing and accessing, so that the system is relatively intelligent, and the problem that false alarm and false alarm may exist in abnormality detection and risk assessment in the financial data safety management system due to manual management is easy to cause.
Disclosure of Invention
(one) solving the technical problems
Aiming at the defects of the prior art, the invention provides an artificial intelligence-based enterprise financial data safety management system and method, which are characterized in that after collection samples of enterprise financial data are collected for processing and feature extraction, a deep learning model is built, intelligent model training is carried out, a comprehensive classification coefficient Zhxs is obtained, the comprehensive classification coefficient Zhxs is compared with a standard threshold value, the risk level is judged, corresponding encryption processing is carried out, and the safety and the protection level of the enterprise financial data are improved. Different security measures are adopted for risk data of different levels, so that unauthorized access and data leakage are prevented; by establishing a deep learning model and analyzing the behavior mode of the encrypted risk data, abnormal activities and access intrusion behaviors can be monitored in real time. The risk behaviors are identified by utilizing an artificial intelligence technology, corresponding reports and early warning notices are generated, enterprises are helped to find and cope with potential risks in time, and the problems that false alarms and missing alarms possibly exist in abnormal detection and risk assessment in a financial data safety management system are reduced.
(II) technical scheme
In order to achieve the above purpose, the invention is realized by the following technical scheme: an enterprise financial data safety management method based on artificial intelligence comprises the following steps,
s1, collecting and obtaining a collection sample of enterprise financial data, and processing the collection sample of the enterprise financial data;
s2, extracting features from the aggregate samples of the processed financial data to obtain feature financial data;
s3, establishing a deep learning model, classifying the processed characteristic financial data, and performing model training to obtain a comprehensive classification coefficient Zhxs;
s4, intelligent security encryption management, wherein the comprehensive classification coefficient Zhxs is compared with a standard threshold value, and if the comprehensive classification coefficient Zhxs exceeds the standard threshold value, the comprehensive classification coefficient Zhxs is judged to be core high-level risk data; if the comprehensive classification coefficient Zhxs is lower than the standard threshold value by 30%, judging that the comprehensive classification coefficient is middle-level risk data; if the comprehensive classification coefficient Zhxs is lower than the standard threshold value by 40%, judging that the comprehensive classification coefficient is low-level risk data; if the comprehensive classification coefficient Zhxs is lower than the standard threshold value by 50%, judging that the class data is public; and carrying out corresponding encryption processing on the corresponding level risk data;
s5, analyzing the behavior mode of the level risk data corresponding to encryption processing, establishing a reference, monitoring abnormal activities and accessing intrusion behaviors, identifying the risk behaviors through an artificial intelligence technology, generating a corresponding report, and sending an early warning notice.
Preferably, the step S1 includes:
s11, collecting a collection sample of enterprise financial data, including classified income, expense and profit data; unclassified data, requiring manual labeling of categories;
revenue includes sales revenue, service fees, and interest revenue for the business; the expenses include purchasing cost, personnel salary month report form, house renting cost and transportation cost of enterprises; profit includes net profit and gross profit for the business; unclassified data includes liabilities, stakeholder equity, tax, and financial statement classifications; liabilities include short-term liabilities, long-term liabilities, and bonds for the corporation; the stakeholder equity includes the stakeholder information, capital reserves, and the earnings of the business; tax includes tax of enterprises, tax declaration and tax adjustment; the financial statement includes an enterprise's liability statement, profit statement, and cash flow statement;
s12, processing the collected aggregate samples of the enterprise financial data, wherein the processing comprises data cleaning, missing value processing and normalization processing.
Preferably, the step S2 includes extracting features from a collection sample of financial data, including numerical feature extraction including total amount, average value and variance extraction, and text type feature extraction including keyword and word frequency information extraction, to obtain feature financial data;
total amount extraction is used to calculate the total amount of financial data, including calculating the total amount of revenue, the total amount of expenditure, and the total amount of profit;
average value extraction means for calculating financial data, including calculating average sales, evaluation costs, and average profits;
the variance extraction is used for calculating variances of the financial data and measuring the variation degree of the data to obtain a variance variation waveform table;
the text type feature extraction is used for processing text fields in the financial data, extracting text word segmentation and word stems, converting the text into a keyword list, and setting keywords as product names, provider names and customer names; the word frequency information extraction comprises the steps of counting the occurrence times of keywords in the financial data in the text, and calculating the word frequency of each keyword in a financial data sample set as a text type feature.
Preferably, the step S3 includes,
s31, extracting the total amount, the average value and the variance to be used as numerical characteristics, and extracting keywords and word frequency information to be used as text type characteristics; associating the features with classification tags of the financial data samples;
s3, establishing a deep learning model to perform classification task training, wherein the training comprises one of a decision tree, a support vector machine and a neural network;
s4, predicting the new characteristic financial data sample according to the deep learning model to obtain a classification result; and calculating the comprehensive classification coefficient Zhxs according to the classification result.
Preferably, the comprehensive classification coefficient Zhxs is obtained by the following formula:
in the method, in the process of the invention,、/>、/>、...、/>the characteristic value is extracted from the characteristic financial data and is set to be a numerical type characteristic or a text type characteristic; />、/>、/>、...、/>Is the weight of the corresponding feature; />Is a correction constant.
Preferably, said S4 comprises,
s41, setting a standard threshold value for judging risks of different levels; setting standard thresholds as 30%, 40% and 50%;
s42, comparing comprehensive classification coefficients Zhxs of the characteristic financial data samples, and judging risk levels according to different conditions:
if Zhxs exceeds a standard threshold, judging the data as core high-level risk data;
if Zhxs is lower than the standard threshold value by 30%, judging that the risk data is medium-grade;
if Zhxs is lower than the standard threshold value by 40%, judging that the risk data is low-level;
if Zhxs is lower than the standard threshold value by 50%, judging that the data is public level data, namely representing risk-free data;
s43, according to the judging result of the risk level, carrying out corresponding encryption method processing on the risk data of the corresponding level.
Preferably, the encryption method of S43 depends on different levels of risk data; the encryption processing method specifically comprises the following steps:
s431, core high-level risk data: data encryption is carried out by adopting three-bit letters and three-bit numbers as passwords, corresponding access rights and control rights are set at the same time, and identity verification and recording are carried out according to access time stamps;
s432, middle-level risk data: encrypting data by adopting six-bit pure letters as passwords, and setting corresponding access rights and control rights;
s433, low-level risk data: encrypting data by using a pure six-bit number as a password, and setting corresponding access rights;
s434, the encryption processing is not performed, the access authority is opened, but the backup record is performed on the access record.
Preferably, the step S5 includes:
s51, establishing a behavior pattern reference of risk data for each level through learning and analysis of a normal behavior pattern; behavior pattern references include access frequency, operation type, and data transfer pattern;
s52, based on the established behavior pattern standard, monitoring the encrypted level risk data in real time by using intelligent anomaly monitoring; monitoring abnormality of access frequency, abnormality of operation type and abnormality of data transmission, and judging the abnormality as first-level risk behavior;
s53, monitoring access intrusion behaviors by adopting a network flow sensor, a log sensor and an identity verification sensor, wherein the access intrusion behaviors comprise unauthorized access, abnormal login attempts and data leakage, and judging the access intrusion behaviors as secondary risk behaviors;
s54, combining the primary risk behavior and the secondary risk behavior, generating a corresponding risk report, analyzing the risk report when the situation with the risk report is identified, generating an early warning command notification, and sending the early warning command notification to a security team.
Preferably, the warning command notification includes one or more of mail, short message and instant message.
An enterprise financial data safety management system based on artificial intelligence comprises a data acquisition module, a data preprocessing module, a feature extraction module, a model training and classifying module, a risk level judging module, a backup module, an encryption module, a safety control module, an abnormality monitoring module and an early warning module;
the data acquisition module collects financial data of enterprises from different data sources and acquires a collection sample of the financial data of the enterprises;
the data preprocessing module is used for performing data cleaning, repeated value removal and missing value processing on the aggregate samples of the enterprise financial data;
the feature extraction module is used for extracting key features from the financial data, including numerical features and text type features;
the model training and classifying module is used for establishing a deep learning model or other machine learning models, inputting the processed characteristic financial data into the model, and calculating and obtaining a comprehensive classification coefficient Zhxs;
the risk level judging module is used for comparing the comprehensive classification coefficient Zhxs with a preset standard threshold value, judging the risk level of the financial data, and dividing the data into risks of different levels according to a set threshold value condition, wherein the risks comprise core high-level risks, medium-level risks, low-level risks and public-level data;
the backup module is used for periodically backing up the data corresponding to the core high-level risk and the medium-level risk;
the encryption module is used for carrying out encryption processing on the risk data of the corresponding level;
the security control module is used for setting corresponding access rights and control measures to ensure that only authorized personnel can access and process sensitive data;
the anomaly monitoring module is used for monitoring the behavior mode and the access activity of the encrypted risk data, identifying the anomaly behavior and the intrusion behavior, and generating a corresponding report comprising details of the anomaly behavior and a risk assessment result;
the early warning module is used for sending early warning notices to relevant responsible persons or safety teams according to the corresponding reports so as to take measures in time to deal with the risk event.
(III) beneficial effects
The invention provides an artificial intelligence-based enterprise financial data security management system and method. The beneficial effects are as follows:
(1) According to the enterprise financial data safety management system and method based on artificial intelligence, the numerical value characteristics and the text type characteristics are extracted from the collection sample of the financial data, so that the characteristic information of the financial data can be comprehensively and accurately expressed. The method is favorable for comprehensively and accurately expressing the characteristic information of the financial data, providing important statistical indexes and variation trends, and capturing key text information and word frequency analysis. By extracting comprehensive features, establishing a deep learning model for classification training and calculating a comprehensive classification coefficient Zhxs, the classification accuracy and flexibility of the financial data can be improved. This helps to better understand and analyze the classification of financial data, providing beneficial information and basis for subsequent risk assessment, encryption management, and behavioral monitoring.
(2) According to the enterprise financial data safety management system and method based on artificial intelligence, the safety and privacy protection level of data can be improved through customized encryption processing aiming at risks of different levels. The strong encryption method of the core high-level risk data can effectively prevent unauthorized access and data leakage and protect the security of sensitive information. The encryption processing of the medium-level and low-level risk data provides moderate data protection, and ensures that the data is not easy to be attacked maliciously in the transmission and storage processes. And the backup record is carried out on the access record, so that the data access behavior is monitored and tracked, and the traceability and the credibility of the data are improved. The access authority and the control authority in the encryption processing method are set, so that the access and the control of the risk data of different levels are more flexible. Corresponding authority levels can be set according to different business requirements and security strategies, so that only authorized personnel can access and operate risk data of the corresponding levels, and compliance and security of data access are improved. By performing customized encryption processing on the risk data according to the risk level determination result, the security, privacy protection level and access control capability of the data can be improved. This helps to protect the financial data of the enterprise from risks and threats and ensures confidentiality and integrity of the data during transmission, storage and processing.
(3) According to the enterprise financial data safety management system and method based on the artificial intelligence, the enterprise financial data safety management method based on the artificial intelligence can monitor and identify risk behaviors, generate corresponding risk reports, timely send early warning notices, and improve the coping capacity and the safety protection level of risks.
(4) According to the enterprise financial data safety management system based on the artificial intelligence, through the synergistic effect of all modules of the system, the enterprise financial data safety management method based on the artificial intelligence can provide effective risk management and protection measures, reduces and ensures the safety, the integrity and the usability of financial data, improves the intelligence based on an artificial intelligence technology, and reduces the problem of false alarm and false alarm caused by abnormal monitoring and risk assessment caused by management of the financial data by manual authorization of an administrator in the system.
Drawings
FIG. 1 is a schematic diagram of steps of an artificial intelligence based enterprise financial data security management method of the present invention;
FIG. 2 is a schematic diagram of an artificial intelligence based enterprise financial data security management system of the present invention;
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
The existence of the enterprise financial data safety management system is used for solving the financial data safety and management challenges faced by enterprises, and providing a comprehensive and effective solution. Financial data is one of the most important, most sensitive assets of an enterprise. Protecting the security of financial data is critical to the sustainable development of an enterprise.
The existing enterprise financial data safety management system simply uploads the financial data of an enterprise to the system, and then transmits, accesses and modifies some authorities of management personnel, but a large amount of financial data is manually encrypted in the process of managing, submitting, storing and accessing, so that the system is relatively intelligent, and the problem that false alarm and false alarm may exist in abnormality detection and risk assessment in the financial data safety management system due to manual management is easy to cause.
The invention provides an artificial intelligence based enterprise financial data security management method, please refer to fig. 1, comprising the following steps,
s1, collecting and obtaining a collection sample of enterprise financial data, and processing the collection sample of the enterprise financial data;
s2, extracting features from the aggregate samples of the processed financial data to obtain feature financial data;
s3, establishing a deep learning model, classifying the processed characteristic financial data, and performing model training to obtain a comprehensive classification coefficient Zhxs;
s4, intelligent security encryption management, wherein the comprehensive classification coefficient Zhxs is compared with a standard threshold value, and if the comprehensive classification coefficient Zhxs exceeds the standard threshold value, the comprehensive classification coefficient Zhxs is judged to be core high-level risk data; if the comprehensive classification coefficient Zhxs is lower than the standard threshold value by 30%, judging that the comprehensive classification coefficient is middle-level risk data; if the comprehensive classification coefficient Zhxs is lower than the standard threshold value by 40%, judging that the comprehensive classification coefficient is low-level risk data; if the comprehensive classification coefficient Zhxs is lower than the standard threshold value by 50%, judging that the class data is public; and carrying out corresponding encryption processing on the corresponding level risk data;
s5, analyzing the behavior mode of the level risk data corresponding to encryption processing, establishing a reference, monitoring abnormal activities and accessing intrusion behaviors, identifying the risk behaviors through an artificial intelligence technology, generating a corresponding report, and sending an early warning notice.
In the embodiment, after the collection samples of the enterprise financial data are collected through S1-S3 for processing and feature extraction, a deep learning model is built, intelligent model training is performed, the comprehensive classification coefficient Zhxs is obtained, the comprehensive classification coefficient Zhxs is compared with a standard threshold in S4, the risk level is judged, corresponding encryption processing is performed, and the safety and the protection level of the enterprise financial data are improved. Different security measures are adopted for risk data of different levels, so that unauthorized access and data leakage are prevented; by establishing a deep learning model and analyzing the behavior mode of the encrypted risk data, abnormal activities and access intrusion behaviors can be monitored in real time. The risk behaviors are identified by utilizing an artificial intelligence technology, corresponding reports and early warning notices are generated, enterprises are helped to find and cope with potential risks in time, and the problems that false alarms and missing alarms possibly exist in abnormal detection and risk assessment in a financial data safety management system are reduced.
Example 2
This embodiment is an explanation of embodiment 1, specifically, the step S1 includes:
s11, collecting a collection sample of enterprise financial data, including classified income, expense and profit data; unclassified data, requiring manual labeling of categories;
revenue includes sales revenue, service fees, and interest revenue for the business; the expenses include purchasing cost, personnel salary month report form, house renting cost and transportation cost of enterprises; profit includes net profit and gross profit for the business; unclassified data includes liabilities, stakeholder equity, tax, and financial statement classifications; liabilities include short-term liabilities, long-term liabilities, and bonds for the corporation; the stakeholder equity includes the stakeholder information, capital reserves, and the earnings of the business; tax includes tax of enterprises, tax declaration and tax adjustment; the financial statement includes an enterprise's liability statement, profit statement, and cash flow statement;
s12, processing the collected aggregate samples of the enterprise financial data, wherein the processing comprises data cleaning, missing value processing and normalization processing.
In this embodiment, through the data cleaning process, errors, repetitions, incompleteness, or inconsistent data in the data set sample may be removed. This helps to improve the accuracy and consistency of the data, avoiding erroneous results in subsequent processing and analysis. In financial data there may be missing values, i.e. some data items lack values or information. By missing value processing, missing data can be filled or estimated using suitable methods, making the data set sample more complete. This helps to improve the usability and reliability of the data. The financial data may have different units of measure and value ranges. Through normalization processing, the data can be converted into uniform scales, so that the comparability between different features is realized. This helps to improve the effectiveness of data analysis and model training and reduce the bias from different scales. Data cleansing, missing value processing and normalization processing help to improve the quality of financial data. Higher quality data set samples can provide more accurate and reliable features for model training and analysis, thereby improving the accuracy and reliability of subsequent risk assessment and decision making.
Example 3
The embodiment is an explanation of embodiment 1, specifically, the step S2 includes extracting features from a set sample of financial data, including numerical feature extraction and text type feature extraction, where the numerical feature extraction includes extracting total amount, average value and variance, and the text type feature extraction includes extracting keyword and word frequency information, to obtain feature financial data;
total amount extraction is used to calculate the total amount of financial data, including calculating the total amount of revenue, the total amount of expenditure, and the total amount of profit; through the sum, the average value and the variance in the numerical feature extraction, important statistical indexes of the financial data can be calculated. The aggregate reflects the overall status of revenue, expenditure and profit.
Average value extraction means for calculating financial data, including calculating average sales, evaluation costs, and average profits; the average provides the average level of financial data and the variance measures the degree of change in the data. These metrics and trends may help businesses understand the overall situation and trend of financial data.
The variance extraction is used for calculating variances of the financial data and measuring the variation degree of the data to obtain a variance variation waveform table;
the text type feature extraction is used for processing text fields in the financial data, extracting text word segmentation and word stems, converting the text into a keyword list, and setting keywords as product names, provider names and customer names; the word frequency information extraction comprises the steps of counting the occurrence times of keywords in the financial data in the text, and calculating the word frequency of each keyword in a financial data sample set as a text type feature.
In this embodiment, by extracting the numerical feature and the text type feature from the aggregate sample of the financial data, the feature information of the financial data can be comprehensively and accurately expressed. The method is favorable for comprehensively and accurately expressing the characteristic information of the financial data, providing important statistical indexes and variation trends, and capturing key text information and word frequency analysis. These features can provide a rich data basis and useful information for subsequent model training, classification, and analysis.
Example 4
This example is an explanation of example 3, and specifically, the step S3 includes,
s31, extracting the total amount, the average value and the variance to be used as numerical characteristics, and extracting keywords and word frequency information to be used as text type characteristics; associating the features with classification tags of the financial data samples; characteristic financial data with rich information can be obtained. Such feature extraction and correlation helps to improve the expressive power and classification accuracy of the data.
S3, establishing a deep learning model to perform classification task training, wherein the training comprises one of a decision tree, a support vector machine and a neural network; the deep learning model has strong learning and expression capability, can effectively learn the mode and the characteristics of the financial data, and improves the classification accuracy and the classification robustness.
S4, predicting the new characteristic financial data sample according to the deep learning model to obtain a classification result; and calculating the comprehensive classification coefficient Zhxs according to the classification result. And predicting the new characteristic financial data sample through the deep learning model to obtain a classification result, and further calculating the comprehensive classification coefficient Zhxs. The comprehensive classification coefficient combines the weight and the feature value of each feature, and comprehensively evaluates the classification result of the sample. Such a comprehensive classification factor can provide a quantified indicator that helps evaluate and compare the degree of classification of different samples.
The comprehensive classification coefficient Zhxs is obtained by the following formula:
in the method, in the process of the invention,、/>、/>、...、/>the characteristic value is extracted from the characteristic financial data and is set to be a numerical type characteristic or a text type characteristic; />、/>、/>、...、/>Is the weight of the corresponding feature; />Is a correction constant.
And predicting the new characteristic financial data sample through the deep learning model to obtain a classification result, and further calculating the comprehensive classification coefficient Zhxs. The comprehensive classification coefficient combines the weight and the feature value of each feature, and comprehensively evaluates the classification result of the sample. Such a comprehensive classification factor can provide a quantified indicator that helps evaluate and compare the degree of classification of different samples.
In this embodiment, by extracting the comprehensive features, establishing the deep learning model for classification training, and calculating the comprehensive classification coefficient Zhxs, the accuracy and flexibility of classification of the financial data can be improved. This helps to better understand and analyze the classification of financial data, providing beneficial information and basis for subsequent risk assessment, encryption management, and behavioral monitoring.
Example 5
This example is an explanation of example 1, and specifically, S4 includes,
s41, setting a standard threshold value for judging risks of different levels; setting standard thresholds as 30%, 40% and 50%;
s42, comparing comprehensive classification coefficients Zhxs of the characteristic financial data samples, and judging risk levels according to different conditions:
if Zhxs exceeds a standard threshold, judging the data as core high-level risk data;
if Zhxs is lower than the standard threshold value by 30%, judging that the risk data is medium-grade;
if Zhxs is lower than the standard threshold value by 40%, judging that the risk data is low-level;
if Zhxs is lower than the standard threshold value by 50%, judging that the data is public level data, namely representing risk-free data;
s43, according to the judging result of the risk level, carrying out corresponding encryption method processing on the risk data of the corresponding level. By setting a standard threshold and comparing the comprehensive classification coefficient Zhxs with the standard threshold, risk data of different levels can be accurately judged. This ensures a high degree of care and protection of the core high-level risk data, as well as proper handling of the medium-level, low-level and public-level data, improving the accuracy and effectiveness of risk management.
The encryption method of S43 is based on risk data of different levels; the encryption processing method specifically comprises the following steps:
s431, core high-level risk data: data encryption is carried out by adopting three-bit letters and three-bit numbers as passwords, corresponding access rights and control rights are set at the same time, and identity verification and recording are carried out according to access time stamps;
s432, middle-level risk data: encrypting data by adopting six-bit pure letters as passwords, and setting corresponding access rights and control rights;
s433, low-level risk data: encrypting data by using a pure six-bit number as a password, and setting corresponding access rights;
s434, the encryption processing is not performed, the access authority is opened, but the backup record is performed on the access record. And carrying out customized encryption processing on the risk data of the corresponding level according to the judging result of the risk level. The core high-level risk data adopts a stronger encryption method, such as three-bit letters and three-bit digital passwords, and sets strict access rights and control rights, and identity verification and recording are performed at the same time. The medium-level risk data is encrypted by adopting a six-bit pure letter password, and the low-level risk data is encrypted by adopting a pure six-bit digital password. For the disclosure level data, encryption processing is not performed, but the access record is backed up.
In this embodiment, the security and privacy protection level of the data may be improved by customizing the encryption process for different levels of risk. The strong encryption method of the core high-level risk data can effectively prevent unauthorized access and data leakage and protect the security of sensitive information. The encryption processing of the medium-level and low-level risk data provides moderate data protection, and ensures that the data is not easy to be attacked maliciously in the transmission and storage processes. And the backup record is carried out on the access record, so that the data access behavior is monitored and tracked, and the traceability and the credibility of the data are improved. The access authority and the control authority in the encryption processing method are set, so that the access and the control of the risk data of different levels are more flexible. Corresponding authority levels can be set according to different business requirements and security strategies, so that only authorized personnel can access and operate risk data of the corresponding levels, and compliance and security of data access are improved. By performing customized encryption processing on the risk data according to the risk level determination result, the security, privacy protection level and access control capability of the data can be improved. This helps to protect the financial data of the enterprise from risks and threats and ensures confidentiality and integrity of the data during transmission, storage and processing.
Example 6
This embodiment is an explanation of embodiment 1, specifically, the S5 includes:
s51, establishing a behavior pattern reference of risk data for each level through learning and analysis of a normal behavior pattern; behavior pattern references include access frequency, operation type, and data transfer pattern;
s52, based on the established behavior pattern standard, monitoring the encrypted level risk data in real time by using intelligent anomaly monitoring; monitoring abnormality of access frequency, abnormality of operation type and abnormality of data transmission, and judging the abnormality as first-level risk behavior; by monitoring the abnormality of the access frequency, the abnormality of the operation type and the abnormality of the data transmission, the first-level risk behavior can be found in time, and the sensitivity and the identification capability to risks are improved.
S53, monitoring access intrusion behaviors by adopting a network flow sensor, a log sensor and an identity verification sensor, wherein the access intrusion behaviors comprise unauthorized access, abnormal login attempts and data leakage, and judging the access intrusion behaviors as secondary risk behaviors; by identifying the intrusion behaviors, potential security threats can be timely found and prevented, and the security of enterprise financial data is protected.
S54, combining the primary risk behavior and the secondary risk behavior, generating a corresponding risk report, analyzing the risk report when the situation with the risk report is identified, generating an early warning command notification, and sending the early warning command notification to a security team.
The early warning command notification comprises one or more of mail, short message and instant message modes.
In the embodiment, the enterprise financial data safety management method based on artificial intelligence can realize monitoring and identification of risk behaviors, generate corresponding risk reports, timely send early warning notification and improve the coping capacity and the safety protection level of risks.
Example 6
Referring to fig. 2, an artificial intelligence-based enterprise financial data security management system includes a data acquisition module, a data preprocessing module, a feature extraction module, a model training and classifying module, a risk level judging module, a backup module, an encryption module, a security control module, an anomaly monitoring module and an early warning module;
the data acquisition module collects financial data of enterprises from different data sources and acquires a collection sample of the financial data of the enterprises;
the data preprocessing module is used for performing data cleaning, repeated value removal and missing value processing on the aggregate samples of the enterprise financial data;
the feature extraction module is used for extracting key features from the financial data, including numerical features and text type features;
the model training and classifying module is used for establishing a deep learning model or other machine learning models, inputting the processed characteristic financial data into the model, and calculating and obtaining a comprehensive classification coefficient Zhxs;
the risk level judging module is used for comparing the comprehensive classification coefficient Zhxs with a preset standard threshold value, judging the risk level of the financial data, and dividing the data into risks of different levels according to a set threshold value condition, wherein the risks comprise core high-level risks, medium-level risks, low-level risks and public-level data;
the backup module is used for periodically backing up the data corresponding to the core high-level risk and the medium-level risk;
the encryption module is used for carrying out encryption processing on the risk data of the corresponding level;
the security control module is used for setting corresponding access rights and control measures to ensure that only authorized personnel can access and process sensitive data;
the anomaly monitoring module is used for monitoring the behavior mode and the access activity of the encrypted risk data, identifying the anomaly behavior and the intrusion behavior, and generating a corresponding report comprising details of the anomaly behavior and a risk assessment result;
the early warning module is used for sending early warning notices to relevant responsible persons or safety teams according to the corresponding reports so as to take measures in time to deal with the risk event.
Through the synergistic effect of the modules, the enterprise financial data safety management method based on artificial intelligence can provide effective risk management and protection measures, reduce and ensure the safety, integrity and usability of financial data, improve the intelligence based on artificial intelligence technology, and reduce the problem of false alarm and false omission caused by abnormal monitoring and risk assessment due to the management of financial data by the artificial authorization of an administrator in a system.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (10)
1. An artificial intelligence-based enterprise financial data security management method is characterized in that: comprises the steps of,
s1, collecting and obtaining a collection sample of enterprise financial data, and processing the collection sample of the enterprise financial data;
s2, extracting features from the aggregate samples of the processed financial data to obtain feature financial data;
s3, establishing a deep learning model, classifying the processed characteristic financial data, and performing model training to obtain a comprehensive classification coefficient Zhxs;
s4, intelligent security encryption management, wherein the comprehensive classification coefficient Zhxs is compared with a standard threshold value, and if the comprehensive classification coefficient Zhxs exceeds the standard threshold value, the comprehensive classification coefficient Zhxs is judged to be core high-level risk data; if the comprehensive classification coefficient Zhxs is lower than the standard threshold value by 30%, judging that the comprehensive classification coefficient is middle-level risk data; if the comprehensive classification coefficient Zhxs is lower than the standard threshold value by 40%, judging that the comprehensive classification coefficient is low-level risk data; if the comprehensive classification coefficient Zhxs is lower than the standard threshold value by 50%, judging that the class data is public; and carrying out corresponding encryption processing on the corresponding level risk data;
s5, analyzing the behavior mode of the level risk data corresponding to encryption processing, establishing a reference, monitoring abnormal activities and accessing intrusion behaviors, identifying the risk behaviors through an artificial intelligence technology, generating a corresponding report, and sending an early warning notice.
2. An artificial intelligence based enterprise financial data security management method in accordance with claim 1, wherein: the step S1 comprises the following steps:
s11, collecting a collection sample of enterprise financial data, including classified income, expense and profit data; unclassified data, requiring manual labeling of categories;
revenue includes sales revenue, service fees, and interest revenue for the business; the expenses include purchasing cost, personnel salary month report form, house renting cost and transportation cost of enterprises; profit includes net profit and gross profit for the business; unclassified data includes liabilities, stakeholder equity, tax, and financial statement classifications; liabilities include short-term liabilities, long-term liabilities, and bonds for the corporation; the stakeholder equity includes the stakeholder information, capital reserves, and the earnings of the business; tax includes tax of enterprises, tax declaration and tax adjustment; the financial statement includes an enterprise's liability statement, profit statement, and cash flow statement;
s12, processing the collected aggregate samples of the enterprise financial data, wherein the processing comprises data cleaning, missing value processing and normalization processing.
3. An artificial intelligence based enterprise financial data security management method in accordance with claim 1, wherein: the S2 step comprises the steps of extracting features from a collection sample of financial data, wherein the feature extraction comprises numerical feature extraction and text type feature extraction, the numerical feature extraction comprises total extraction, average value extraction and variance extraction, and the text type feature extraction comprises keyword extraction and word frequency information extraction, so that feature financial data are obtained;
total amount extraction is used to calculate the total amount of financial data, including calculating the total amount of revenue, the total amount of expenditure, and the total amount of profit;
average value extraction means for calculating financial data, including calculating average sales, evaluation costs, and average profits;
the variance extraction is used for calculating variances of the financial data and measuring the variation degree of the data to obtain a variance variation waveform table;
the text type feature extraction is used for processing text fields in the financial data, extracting text word segmentation and word stems, converting the text into a keyword list, and setting keywords as product names, provider names and customer names; the word frequency information extraction comprises the steps of counting the occurrence times of keywords in the financial data in the text, and calculating the word frequency of each keyword in a financial data sample set as a text type feature.
4. An artificial intelligence based enterprise financial data security management method in accordance with claim 1, wherein: the step S3 includes the steps of,
s31, extracting the total amount, the average value and the variance to be used as numerical characteristics, and extracting keywords and word frequency information to be used as text type characteristics; associating the features with classification tags of the financial data samples;
s3, establishing a deep learning model to perform classification task training, wherein the training comprises one of a decision tree, a support vector machine and a neural network;
s4, predicting the new characteristic financial data sample according to the deep learning model to obtain a classification result; and calculating the comprehensive classification coefficient Zhxs according to the classification result.
5. The artificial intelligence based enterprise financial data security management method of claim 4, wherein: the comprehensive classification coefficient Zhxs is obtained by the following formula:
in the method, in the process of the invention,、/>、/>、...、/>the characteristic value is extracted from the characteristic financial data and is set to be a numerical type characteristic or a text type characteristic; />、/>、/>、...、/>Is the weight of the corresponding feature; />Is a correction constant.
6. An artificial intelligence based enterprise financial data security management method in accordance with claim 1, wherein: the step S4 includes the step of,
s41, setting a standard threshold value for judging risks of different levels; setting standard thresholds as 30%, 40% and 50%;
s42, comparing comprehensive classification coefficients Zhxs of the characteristic financial data samples, and judging risk levels according to different conditions:
if Zhxs exceeds a standard threshold, judging the data as core high-level risk data;
if Zhxs is lower than the standard threshold value by 30%, judging that the risk data is medium-grade;
if Zhxs is lower than the standard threshold value by 40%, judging that the risk data is low-level;
if Zhxs is lower than the standard threshold value by 50%, judging that the data is public level data, namely representing risk-free data;
s43, according to the judging result of the risk level, carrying out corresponding encryption method processing on the risk data of the corresponding level.
7. The artificial intelligence based enterprise financial data security management method of claim 6, wherein: the encryption method of S43 is based on risk data of different levels; the encryption processing method specifically comprises the following steps:
s431, core high-level risk data: data encryption is carried out by adopting three-bit letters and three-bit numbers as passwords, corresponding access rights and control rights are set at the same time, and identity verification and recording are carried out according to access time stamps;
s432, middle-level risk data: encrypting data by adopting six-bit pure letters as passwords, and setting corresponding access rights and control rights;
s433, low-level risk data: encrypting data by using a pure six-bit number as a password, and setting corresponding access rights;
s434, the encryption processing is not performed, the access authority is opened, but the backup record is performed on the access record.
8. An artificial intelligence based enterprise financial data security management method in accordance with claim 1, wherein: the step S5 comprises the following steps:
s51, establishing a behavior pattern reference of risk data for each level through learning and analysis of a normal behavior pattern; behavior pattern references include access frequency, operation type, and data transfer pattern;
s52, based on the established behavior pattern standard, monitoring the encrypted level risk data in real time by using intelligent anomaly monitoring; monitoring abnormality of access frequency, abnormality of operation type and abnormality of data transmission, and judging the abnormality as first-level risk behavior;
s53, monitoring access intrusion behaviors by adopting a network flow sensor, a log sensor and an identity verification sensor, wherein the access intrusion behaviors comprise unauthorized access, abnormal login attempts and data leakage, and judging the access intrusion behaviors as secondary risk behaviors;
s54, combining the primary risk behavior and the secondary risk behavior, generating a corresponding risk report, analyzing the risk report when the situation with the risk report is identified, generating an early warning command notification, and sending the early warning command notification to a security team.
9. An artificial intelligence based enterprise financial data security management method in accordance with claim 8, wherein: the early warning command notification comprises one or more of mail, short message and instant message modes.
10. An artificial intelligence-based enterprise financial data security management system, which is characterized in that: the system comprises a data acquisition module, a data preprocessing module, a feature extraction module, a model training and classifying module, a risk level judging module, a backup module, an encryption module, a safety control module, an abnormality monitoring module and an early warning module;
the data acquisition module collects financial data of enterprises from different data sources and acquires a collection sample of the financial data of the enterprises;
the data preprocessing module is used for performing data cleaning, repeated value removal and missing value processing on the aggregate samples of the enterprise financial data;
the feature extraction module is used for extracting key features from the financial data, including numerical features and text type features;
the model training and classifying module is used for establishing a deep learning model or other machine learning models, inputting the processed characteristic financial data into the model, and calculating and obtaining a comprehensive classification coefficient Zhxs;
the risk level judging module is used for comparing the comprehensive classification coefficient Zhxs with a preset standard threshold value, judging the risk level of the financial data, and dividing the data into risks of different levels according to a set threshold value condition, wherein the risks comprise core high-level risks, medium-level risks, low-level risks and public-level data;
the backup module is used for periodically backing up the data corresponding to the core high-level risk and the medium-level risk;
the encryption module is used for carrying out encryption processing on the risk data of the corresponding level;
the security control module is used for setting corresponding access rights and control measures to ensure that only authorized personnel can access and process sensitive data;
the anomaly monitoring module is used for monitoring the behavior mode and the access activity of the encrypted risk data, identifying the anomaly behavior and the intrusion behavior, and generating a corresponding report comprising details of the anomaly behavior and a risk assessment result;
the early warning module is used for sending early warning notices to relevant responsible persons or safety teams according to the corresponding reports so as to take measures in time to deal with the risk event.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311467424.4A CN117216801A (en) | 2023-11-07 | 2023-11-07 | Enterprise financial data safety management system and method based on artificial intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311467424.4A CN117216801A (en) | 2023-11-07 | 2023-11-07 | Enterprise financial data safety management system and method based on artificial intelligence |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117216801A true CN117216801A (en) | 2023-12-12 |
Family
ID=89039251
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311467424.4A Pending CN117216801A (en) | 2023-11-07 | 2023-11-07 | Enterprise financial data safety management system and method based on artificial intelligence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117216801A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117455245A (en) * | 2023-12-22 | 2024-01-26 | 赛飞特工程技术集团有限公司 | Intelligent risk assessment system for enterprise safety production |
CN117592092A (en) * | 2024-01-19 | 2024-02-23 | 山东铭云信息技术有限公司 | Secret checking method and system for database content |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112150263A (en) * | 2020-10-09 | 2020-12-29 | 信阳农林学院 | Enterprise financial risk early warning system |
CN113177728A (en) * | 2021-05-20 | 2021-07-27 | 珠海华发金融科技研究院有限公司 | Enterprise operation and financial risk management and control method and system |
CN113989017A (en) * | 2021-10-20 | 2022-01-28 | 中国农业银行股份有限公司河北省分行 | Post-loan risk early warning device and method for public clients |
CN114048436A (en) * | 2021-11-11 | 2022-02-15 | 北京道口金科科技有限公司 | Construction method and construction device for forecasting enterprise financial data model |
CN114331735A (en) * | 2021-12-16 | 2022-04-12 | 普洛斯科技(重庆)有限公司 | Financial risk model training method and device and financial risk prediction method and device |
CN115293598A (en) * | 2022-08-10 | 2022-11-04 | 山东财经大学 | Enterprise financial management risk identification method based on financial big data |
CN115423594A (en) * | 2022-09-29 | 2022-12-02 | 东方星野数字科技(北京)有限公司 | Enterprise financial risk assessment method, device, equipment and storage medium |
CN116402630A (en) * | 2023-06-09 | 2023-07-07 | 深圳市迪博企业风险管理技术有限公司 | Financial risk prediction method and system based on characterization learning |
CN116739811A (en) * | 2023-06-28 | 2023-09-12 | 威海海洋职业学院 | Enterprise financial information intelligent management system and method for self-adaptive risk control |
CN116777652A (en) * | 2023-05-24 | 2023-09-19 | 航天科工网络信息发展有限公司 | Risk evaluation model-based financial analysis method |
-
2023
- 2023-11-07 CN CN202311467424.4A patent/CN117216801A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112150263A (en) * | 2020-10-09 | 2020-12-29 | 信阳农林学院 | Enterprise financial risk early warning system |
CN113177728A (en) * | 2021-05-20 | 2021-07-27 | 珠海华发金融科技研究院有限公司 | Enterprise operation and financial risk management and control method and system |
CN113989017A (en) * | 2021-10-20 | 2022-01-28 | 中国农业银行股份有限公司河北省分行 | Post-loan risk early warning device and method for public clients |
CN114048436A (en) * | 2021-11-11 | 2022-02-15 | 北京道口金科科技有限公司 | Construction method and construction device for forecasting enterprise financial data model |
CN114331735A (en) * | 2021-12-16 | 2022-04-12 | 普洛斯科技(重庆)有限公司 | Financial risk model training method and device and financial risk prediction method and device |
CN115293598A (en) * | 2022-08-10 | 2022-11-04 | 山东财经大学 | Enterprise financial management risk identification method based on financial big data |
CN115423594A (en) * | 2022-09-29 | 2022-12-02 | 东方星野数字科技(北京)有限公司 | Enterprise financial risk assessment method, device, equipment and storage medium |
CN116777652A (en) * | 2023-05-24 | 2023-09-19 | 航天科工网络信息发展有限公司 | Risk evaluation model-based financial analysis method |
CN116402630A (en) * | 2023-06-09 | 2023-07-07 | 深圳市迪博企业风险管理技术有限公司 | Financial risk prediction method and system based on characterization learning |
CN116739811A (en) * | 2023-06-28 | 2023-09-12 | 威海海洋职业学院 | Enterprise financial information intelligent management system and method for self-adaptive risk control |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117455245A (en) * | 2023-12-22 | 2024-01-26 | 赛飞特工程技术集团有限公司 | Intelligent risk assessment system for enterprise safety production |
CN117592092A (en) * | 2024-01-19 | 2024-02-23 | 山东铭云信息技术有限公司 | Secret checking method and system for database content |
CN117592092B (en) * | 2024-01-19 | 2024-04-05 | 山东铭云信息技术有限公司 | Secret checking method and system for database content |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Holton | Identifying disgruntled employee systems fraud risk through text mining: A simple solution for a multi-billion dollar problem | |
CN117216801A (en) | Enterprise financial data safety management system and method based on artificial intelligence | |
US12034755B2 (en) | Computationally assessing and remediating security threats | |
US20220327541A1 (en) | Systems and methods of generating risk scores and predictive fraud modeling | |
Zhang et al. | Lab | |
Lewis et al. | DIGITAL AUDITING: Modernizing the Government Financial Statement Audit Approach. | |
CN117094184B (en) | Modeling method, system and medium of risk prediction model based on intranet platform | |
CN112637108A (en) | Internal threat analysis method and system based on anomaly detection and emotion analysis | |
CN117132114A (en) | Enterprise internal risk management precaution device system | |
CA3214663A1 (en) | Systems and methods of generating risk scores and predictive fraud modeling | |
CN117421761B (en) | Database data information security monitoring method | |
CN117370548A (en) | User behavior risk identification method, device, electronic equipment and medium | |
Mihailescu et al. | Unveiling Threats: Leveraging User Behavior Analysis for Enhanced Cybersecurity | |
Heidenreich | How to design a method for measuring IT security in micro enterprises for IT security level measuring? A literature analysis | |
Kearns et al. | Developing a forensic continuous audit model | |
CN114003969A (en) | Risk assessment method based on block chain technology | |
Otero | Optimization methodology for change management controls using Grey Systems Theory | |
He et al. | Modeling and management of cyber risk: a cross-disciplinary review | |
CN118036080B (en) | Data security treatment method and system based on big data technology | |
Kubigenova et al. | Prospects for Information Security in Big Data Technology | |
Vedapuri et al. | Corporate Tax Compliance and Fraud Prevention System by Principal Component Analysis and Auto-Encoder | |
Kubigenova et al. | VIEWS ON BIG DATA TECHNOLOGY INFORMATION SECURITY | |
CN118279067A (en) | Information data management method based on process mining technology | |
Wong et al. | Learning System Security Compliance for Banking | |
Samokhvalov et al. | Methodological Approach to Assessing Information Security of Critical Infrastructure Objects. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |