CN116485185A - Enterprise risk analysis system and method based on comparison data - Google Patents
Enterprise risk analysis system and method based on comparison data Download PDFInfo
- Publication number
- CN116485185A CN116485185A CN202310469686.8A CN202310469686A CN116485185A CN 116485185 A CN116485185 A CN 116485185A CN 202310469686 A CN202310469686 A CN 202310469686A CN 116485185 A CN116485185 A CN 116485185A
- Authority
- CN
- China
- Prior art keywords
- data
- enterprise
- wind control
- rule
- risk
- 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
- 238000012502 risk assessment Methods 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 title claims abstract description 32
- 230000004927 fusion Effects 0.000 claims abstract description 45
- 238000012549 training Methods 0.000 claims abstract description 36
- 230000002265 prevention Effects 0.000 claims abstract description 30
- 238000011217 control strategy Methods 0.000 claims abstract description 17
- 238000013075 data extraction Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 69
- 239000013598 vector Substances 0.000 claims description 38
- 230000014509 gene expression Effects 0.000 claims description 35
- 230000011218 segmentation Effects 0.000 claims description 15
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 8
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 5
- 238000012545 processing Methods 0.000 abstract description 5
- 230000006870 function Effects 0.000 description 15
- 238000001514 detection method Methods 0.000 description 7
- 238000007726 management method Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 230000003472 neutralizing effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 238000012954 risk control Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data processing, and discloses an enterprise risk analysis system, method, equipment and storage medium based on comparison data. The system comprises a wind control rule model generation module, a wind control rule generation module, a wind control data extraction module, a contrast calculation module and an enterprise risk determination module, wherein the wind control rule model can be obtained by training the enterprise data characteristics according to a regression model; generating a fusion wind control rule according to the wind control rule model; extracting wind control data in the fusion wind control rule; and calculating the contrast of the wind control data and the real-time target enterprise data, calculating a risk value of the target enterprise according to the contrast and the rule triggering threshold value, determining the risk of the target enterprise according to the risk value, and generating a risk prevention and control strategy according to the risk of the target enterprise. The invention can improve the data security when the enterprise risk analysis is carried out.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an enterprise risk analysis system and method based on comparison data.
Background
Along with the development of enterprise internationalization and service diversification, the service types and service data faced by enterprises are more and more complex, and the enterprise risk is continuously increased, so that in order to bring risk prevention to the enterprises, real-time analysis is needed to be carried out on the data generated by the enterprises so as to carry out enterprise risk prevention.
The existing enterprise risk analysis technology is based on big data technology, and the enterprise risk is analyzed through a traditional Internet and machine system wind control method. In practical application, the traditional wind control method lacks a dynamic, comprehensive and accurate processing means according to enterprise production operation face data, and cannot comprehensively analyze enterprise risk events, so that data security is lower when enterprise risk analysis is performed.
Disclosure of Invention
The invention provides an enterprise risk analysis system, an enterprise risk analysis method, electronic equipment and a computer readable storage medium based on comparison data, and mainly aims to solve the problem of low data security during enterprise risk analysis.
In order to achieve the above object, the present invention provides an enterprise risk analysis system based on comparison data, which is characterized in that the system comprises a wind control rule model generation module, a wind control rule generation module, a wind control data extraction module, a contrast calculation module and an enterprise risk determination module, wherein,
The wind control rule model generation module is used for acquiring enterprise sample data, extracting enterprise data characteristics of the enterprise sample data, and training the enterprise data characteristics by utilizing a preset regression model to obtain a wind control rule model;
the wind control rule generation module is used for determining rule probability of the enterprise sample data according to the wind control rule model and generating a first wind control rule and a second wind control rule of a preset target enterprise according to the rule probability;
the wind control data extraction module is configured to perform rule fusion on the first wind control rule and the second wind control rule by using a preset data fusion algorithm to obtain a fused wind control rule, and extract wind control data in the fused wind control rule by using a preset probability algorithm, where the method is specifically configured to, when calculating a data risk probability in the fused wind control rule by using the wind control rule model:
and determining wind control data in the fusion wind control rule according to the data risk probability by using the probability algorithm, wherein the probability algorithm is as follows:
wherein,,for the wind control data, < >>Predicting a model theta of a stroke control rule under a sample feature vector x to be +. >Time probability (I)>Predicting a model theta of a stroke control rule under a sample feature vector x to be +.>Time probability, arg min is a minimum function;
the contrast computing module is used for acquiring real-time target enterprise data and computing the contrast between the wind control data and the real-time target enterprise data through a preset contrast algorithm;
the enterprise risk determination module is used for calculating a risk value of the target enterprise according to the contrast and a preset rule trigger threshold through a preset risk assessment algorithm, determining a risk of the target enterprise according to the risk value, and generating a risk prevention and control strategy according to the risk of the target enterprise.
Optionally, the wind control rule model generation module is specifically configured to, when extracting the enterprise data feature of the enterprise sample data:
dividing the enterprise sample data according to preset data attributes to obtain divided sample data;
vector conversion is carried out on the divided sample data one by using a preset vector conversion model, so that an enterprise sample vector is obtained;
the enterprise sample vector is treated as the enterprise data feature.
Optionally, the wind control rule model generating module is specifically configured to, when training the enterprise data feature by using a preset regression model to obtain a wind control rule model:
Training the enterprise data features by using the regression model to obtain training parameters, wherein the regression model is as follows:
z＝w T x+b
wherein p (y=1|x) is the probability of the sample feature vector x when the prediction type y=1, x is the sample feature vector in the enterprise data feature, z is a training value, T is a transposed symbol, w is a first training parameter, and b is a second training parameter;
updating the training parameters through a preset loss function to obtain optimal training parameters;
and assigning the optimal training parameters to the regression model to obtain the wind control rule model.
Optionally, the wind control rule generation module is specifically configured to, when determining the rule probability of the enterprise sample data according to the wind control rule model:
acquiring feature score values of the enterprise data features;
dividing the enterprise sample data into a first target rule set when the feature segmentation value is smaller than or equal to a preset feature segmentation threshold value;
dividing the enterprise sample data into a second target rule set when the feature segmentation value is larger than a preset feature segmentation threshold value;
determining a first rule probability of the first target rule set according to the wind control rule model, and determining a second rule probability of the second target rule set by using the wind control rule model;
And collecting the first rule probability and the second rule probability as rule probabilities of the enterprise sample data.
Optionally, when the wind control rule generating module generates the first wind control rule and the second wind control rule of the preset target enterprise according to the rule probability, the wind control rule generating module is specifically configured to:
sequencing the rule probabilities according to a sequencing order from big to small to obtain a rule probability sequence;
selecting the highest target rule set in the rule probability sequence as the first wind control rule;
and selecting the next highest target rule set in the rule probability sequence as the second wind control rule.
Optionally, when the wind control data extraction module performs rule fusion on the first wind control rule and the second wind control rule by using a preset data fusion algorithm to obtain a fused wind control rule, the wind control data extraction module is specifically configured to:
acquiring a first rule logic expression in the first wind control rule;
acquiring a second rule logic expression in the second wind control rule;
and fusing the first rule logic expression and the second rule logic expression by using the data fusion algorithm to obtain the fused wind control rule, wherein the data fusion algorithm is as follows:
G＝R∪Q
Wherein G is the fusion wind control rule, R is the first rule logic expression, and Q is the second rule logic expression.
Optionally, the contrast calculating module is specifically configured to, when calculating the contrast between the wind control data and the real-time target enterprise data through a preset contrast algorithm:
performing core semantic extraction on the wind control data to obtain wind control data semantics, and performing core semantic extraction on the real-time target enterprise data to obtain real-time enterprise data semantics;
generating a semantic tree according to the wind control data semantics and the real-time enterprise data semantics;
calculating semantic weights among each edge in the semantic tree by using a preset weight algorithm, wherein the weight algorithm is as follows:
wherein w is i,j For the semantic weight, c, between the wind control data semantic i and the real-time enterprise data semantic j 1 As a first weight factor, c 2 For the second weight factor, d is the depth of the semantic tree, k i The method is characterized in that the method is used for layer numbering of wind control data semantics i, alpha is a weight adjustment parameter, pi is a circumference rate, sin is a sine function, log is a log function, and k is a j Layer numbering of real-time enterprise data semantics j, m being the total number of nodes of the semantic tree;
Calculating the contrast of the wind control data and the real-time target enterprise data according to the semantic weight by using the following contrast algorithm:
wherein s is u,v For the contrast ratio of the wind control data u and the real-time target enterprise data v, delta is a contrast adjustment parameter, w i,j And m is the total number of nodes of the semantic tree, and is the semantic weight between the data semantic i and the data semantic j.
Optionally, the enterprise risk determination module is specifically configured to, when calculating the risk value of the target enterprise according to the contrast and the preset rule trigger threshold through a preset risk assessment algorithm:
acquiring a rule value of the real-time target enterprise data;
when the rule value is larger than a preset rule trigger threshold, calculating a risk value of the target enterprise according to the contrast by using the risk assessment algorithm, wherein the risk assessment algorithm is as follows:
wherein F is the risk value, s is the contrast, ε k Risk compensation coefficient, τ, for the kth target risk indicator k And n is the number of target risk indexes, and the severity of the kth target risk index is the severity of the kth target risk index.
Optionally, the enterprise risk determination module is specifically configured to, when generating a risk prevention and control policy according to the target enterprise risk:
Determining a risk level of the target enterprise risk according to the risk value;
and generating enterprise risk prevention and control strategies corresponding to the risk grades through a preset risk strategy library.
In order to solve the above problems, the present invention further provides an operation method of an enterprise risk analysis system based on comparison data, the method comprising:
acquiring enterprise sample data, extracting enterprise data characteristics of the enterprise sample data, and training the enterprise data characteristics by using a preset regression model to obtain a wind control rule model;
determining rule probability of the enterprise sample data according to the wind control rule model, and generating a first wind control rule and a second wind control rule of a preset target enterprise according to the rule probability;
performing rule fusion on the first wind control rule and the second wind control rule by using a preset data fusion algorithm to obtain a fused wind control rule, and extracting wind control data in the fused wind control rule by using a preset probability algorithm;
acquiring real-time target enterprise data, and calculating the contrast between the wind control data and the real-time target enterprise data through a preset contrast algorithm;
and calculating a risk value of the target enterprise according to the contrast and a preset rule triggering threshold through a preset risk assessment algorithm, determining the risk of the target enterprise according to the risk value, and generating a risk prevention and control strategy according to the risk of the target enterprise.
According to the embodiment of the invention, through the enterprise data characteristics of the enterprise sample data, the regression model is further utilized to train the enterprise data characteristics to obtain the wind control rule model, so that the wind control rule model is beneficial to generating enterprise wind control rules corresponding to the enterprise data according to the wind control rule model, different enterprise wind control rules are fused to obtain the fused wind control rules, the comprehensive wind control rules are beneficial to obtaining, and the accuracy of enterprise risk detection is ensured; and extracting wind control data in the fusion wind control rule, comparing the wind control data with real-time target enterprise data according to the minimum risk of the wind control data to obtain data contrast, determining a risk value of the real-time target enterprise data based on the data contrast, determining the risk of the target enterprise according to the risk value, generating a risk prevention and control strategy according to the risk of the target enterprise, performing risk prevention and control on the target enterprise according to the risk prevention and control strategy, and improving the operation and data safety of the target enterprise. Therefore, the enterprise risk analysis system and method based on comparison data can solve the problem of lower data security when a user performs enterprise risk analysis.
Drawings
FIG. 1 is a functional block diagram of an enterprise risk analysis system based on comparison data according to an embodiment of the present invention;
Fig. 2 is a flowchart illustrating an operation method of an enterprise risk analysis system based on comparison data according to an embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. 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.
The terminology used in the embodiments of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise, the "plurality" generally includes at least two.
The words "if", as used herein, may be interpreted as "at … …" or "at … …" or "in response to a determination" or "in response to a detection", depending on the context. Similarly, the phrase "if determined" or "if detected (stated condition or event)" may be interpreted as "when determined" or "in response to determination" or "when detected (stated condition or event)" or "in response to detection (stated condition or event), depending on the context.
In addition, the sequence of steps in the method embodiments described below is only an example and is not strictly limited.
In practice, the server-side devices deployed by the comparison data-based enterprise risk analysis system may be comprised of one or more devices. The enterprise risk analysis system based on comparison data can be implemented as follows: service instance, virtual machine, hardware device. For example, the comparison data-based enterprise risk analysis system may be implemented as a business instance deployed on one or more devices in a cloud node. Briefly, the comparison data-based enterprise risk analysis system may be understood as a software deployed on cloud nodes to provide a comparison data-based enterprise risk analysis system for each user. Alternatively, the comparison data-based enterprise risk analysis system may also be implemented as a virtual machine deployed on one or more devices in the cloud node. The virtual machine is provided with application software for managing each user side. Or, the enterprise risk analysis system based on the comparison data can also be realized as a service end formed by a plurality of hardware devices of the same or different types, and one or more hardware devices are arranged for providing the enterprise risk analysis system based on the comparison data for each user end.
In an implementation form, the enterprise risk analysis system and the user side based on comparison data are mutually adapted. Namely, the enterprise risk analysis system based on the comparison data is used as an application installed on the cloud service platform, and the user side is used as a client side for establishing communication connection with the application; or the enterprise risk analysis system based on comparison data is realized as a website, and the user side is realized as a webpage; and then or the enterprise risk analysis system based on comparison data is realized as a cloud service platform, and the user side is realized as an applet in the instant messaging application.
Referring to fig. 1, a functional block diagram of an enterprise risk analysis system based on comparison data according to an embodiment of the present invention is shown.
The enterprise risk analysis system 100 based on comparison data of the present invention may be disposed in a cloud server, and in implementation form, may be used as one or more service devices, may also be used as an application installed on a cloud (e.g., a server of a mobile service operator, a server cluster, etc.), or may also be developed as a website. Depending on the functions implemented, the comparison data-based enterprise risk analysis system 100 may include a wind control rule model generation module 101, a wind control rule generation module 102, a wind control data extraction module 103, a contrast calculation module 104, and an enterprise risk determination module 105. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the enterprise risk analysis system based on the comparison data, each module can be independently realized and called with other modules. A call herein is understood to mean that a module may connect to a plurality of modules of another type and provide corresponding services to the plurality of modules to which it is connected. For example, the sharing evaluation module can call the same information acquisition module to acquire the information acquired by the information acquisition module based on the characteristics, and in the enterprise risk analysis system based on the comparison data provided by the embodiment of the invention, the application range of the enterprise risk analysis system architecture based on the comparison data can be adjusted by adding the module and directly calling the module without modifying the program code, so that the cluster-type horizontal expansion is realized, and the aim of rapidly and flexibly expanding the enterprise risk analysis system based on the comparison data is fulfilled. In practical applications, the modules may be disposed in the same device or different devices, or may be service instances disposed in virtual devices, for example, in a cloud server.
The following description is directed to various components of the enterprise risk analysis system based on the comparison data and specific workflows, respectively, in conjunction with specific embodiments:
The wind control rule model generating module 101 is configured to obtain enterprise sample data, extract enterprise data features of the enterprise sample data, and train the enterprise data features by using a preset regression model to obtain a wind control rule model;
in the embodiment of the present invention, the enterprise sample data is used to evaluate the overall status of an enterprise, including enterprise basic information (registration information, operation scope, operation mode, enterprise scale, etc.), industry environment (industry competition status, market trend, policy regulation, etc.), enterprise management (organization architecture, decision level, internal control, risk management, etc.), financial status (asset liability list, profit list, cash flow list), so that the risk situation of the enterprise is evaluated according to factors such as enterprise information, industry environment, management system, financial status, etc. in the enterprise sample data, and possible risks and coping strategies are listed. In addition, each basic enterprise data contains a label, the labels are risky and risky, and the enterprise sample data is { (x) 1 ,y 1 ),(x 2 ,y 2 ),…,(x n ,y n )}。
In detail, the enterprise sample data may be acquired through a computer sentence (e.g., java sentence, python sentence, etc.) having a data grabbing function.
Further, the characteristic extraction is carried out on the enterprise sample data to obtain enterprise data characteristic vectors of the enterprise sample data, and the enterprise data characteristic vectors are used for training a regression model to obtain a wind control rule model.
In the embodiment of the present invention, the enterprise data features are data features obtained by performing vector conversion on each divided sample data, such asRepresenting enterprise data characteristics ∈ ->Representing the feature vector contained in the first divided sample data,/i>Then->Divided sample data representing basic information of an enterprise, then (x 1 ,y 1 ) Feature vector (x) representing registered information in basic information of enterprise 2 ,y 2 ) And a feature vector representing the business scope in the basic information of the enterprise.
In the embodiment of the present invention, the wind control rule model generation module 101 is specifically configured to:
dividing the enterprise sample data according to preset data attributes to obtain divided sample data;
vector conversion is carried out on the divided sample data one by using a preset vector conversion model, so that an enterprise sample vector is obtained;
the enterprise sample vector is treated as the enterprise data feature.
In detail, the data attribute is a data category to which the enterprise sample data belongs, including enterprise basic information, an industry environment, a management system and a financial condition, and if the enterprise sample data is registration information, an operation range, an operation mode, an enterprise scale and the like, the enterprise sample data is divided into the enterprise basic information; the enterprise sample data is an asset liability statement, a profit statement, a cash flow statement, etc., and is divided into financial conditions, and thus, divided sample data can be obtained.
Specifically, the vector conversion model includes, but is not limited to, a word2vec model and a Bert model, and the vector conversion model may be used to perform vector conversion on the divided sample data one by one to obtain an enterprise sample vector corresponding to each divided sample data, where the enterprise sample vector is used as the enterprise data feature.
Further, the data with risks and without risks in the enterprise sample data are classified through the enterprise data characteristics, so that a risk wind control rule model is generated and used for preventing and controlling the wind control data according to the wind control rule, and the safety of the enterprise data is guaranteed.
In the embodiment of the invention, the wind control rule model is used for training the risk prevention and control rule of the enterprise risk according to the enterprise risk data so as to generate the wind control rule model, and the wind control rule model is used for performing risk prevention and control on the enterprise data according to the wind control rule screened out by the wind control rule model.
In the embodiment of the present invention, when the wind control rule model generating module 101 trains the enterprise data feature by using a preset regression model, the wind control rule model is specifically configured to:
training the enterprise data features by using the regression model to obtain training parameters, wherein the regression model is as follows:
z＝w T x+b
Wherein p (y=1|x) is the probability of the sample feature vector x when the prediction type y=1, x is the sample feature vector in the enterprise data feature, z is a training value, T is a transposed symbol, w is a first training parameter, and b is a second training parameter;
updating the training parameters through a preset loss function to obtain optimal training parameters;
and assigning the optimal training parameters to the regression model to obtain the wind control rule model.
In detail, w is the same column vector as the x dimension in the regression model, and w T x is a scalar and b is a bias value, and the enterprise data feature is input into the regression model to obtain the formula z=w T The w parameter and the b parameter in x+b are continuously updated through a preset loss function and a preset gradient descent algorithm to obtain an optimal training parameter, and the optimal training parameter is input into a formula z=w T And a mathematical expression of the wind control rule model can be obtained in x+b.
Illustratively, the formula z=w in the wind control rule model T x+b isTaking enterprise data characteristics corresponding to enterprise sample data as x T ＝{x 1 ,x 2 ,x 3 Vector, then according to the formula z=w T x+b can determine the value of z and thus from Risk data probabilities in enterprise sample data are determined, where y=1 indicates risk and y=0 indicates no risk.
Further, according to the wind control rule model, the risk data probability in the enterprise sample data can be determined, and further the wind control rule is generated according to the risk data probability, so that the comprehensiveness of prevention and control of enterprise risk data is improved.
The wind control rule generating module 102 is configured to determine a rule probability of the enterprise sample data according to the wind control rule model, and generate a first wind control rule and a second wind control rule of a preset target enterprise according to the rule probability;
in the embodiment of the invention, the rule probability refers to classifying enterprise sample data into different rule sets according to different characteristic values, and further calculating the rule probability of each rule set according to the wind control rule model.
In the embodiment of the present invention, when determining the rule probability of the enterprise sample data according to the wind control rule model, the wind control rule generation module 102 is specifically configured to:
acquiring feature score values of the enterprise data features;
dividing the enterprise sample data into a first target rule set when the feature segmentation value is smaller than or equal to a preset feature segmentation threshold value;
Dividing the enterprise sample data into a second target rule set when the feature segmentation value is larger than a preset feature segmentation threshold value;
determining a first rule probability of the first target rule set according to the wind control rule model, and determining a second rule probability of the second target rule set by using the wind control rule model;
and collecting the first rule probability and the second rule probability as rule probabilities of the enterprise sample data.
In detail, the feature score includes mean, variance and standard deviation, and the quantized value corresponding to each enterprise data feature can be calculated to obtain the mean, variance and standard deviation Then all +.>Corresponding x in (a) 1 The quantized values calculate the mean, variance and standard deviation according to a mathematical calculation formula, i.e. the mean is r= (x) 1 +…+x n ) N, where n is the number of all vectors, and similarly, the variance and standard deviation are calculated according to the variance calculation formula and standard deviation calculation formula.
For example, when the mean value is 5, the variance is 9, and the standard deviation is 3 in the feature segmentation threshold, the mean value is taken as a first level, the variance is taken as a second level, and the standard deviation is taken asFor the third layer, whenThe feature score is {1,8,4}, +. >The feature score is {2, 10,2}, -for }>The feature score is {6, 17,5}, -for }>The feature score value is {7,6,1}, i.e. will +.>Dividing into a first target rule set, and when the average value of at least two characteristic segmentation values is greater than the characteristic segmentation threshold value, adding ∈>Divided into a second set of target rules. Until->And (3) completing all the division to obtain different first target rule sets and second target rule sets.
Specifically, a first rule probability of the first target rule set is calculated using the wind control rule model, i.eThe calculated probability is used as a first rule probability, and a second rule probability of the second target rule set is calculated by utilizing the wind control rule model, namely +.>The obtained probability is taken as a second rule probability, and the first rule probability is collectedAnd the second rule probability is the rule probability of the enterprise sample data, namely the rule probability is { P } 11 ,P 12 ,...,P 21, ...,P 2n },P 11 For the first target rule set probability in the first rule probability, P 21, For the second target rule set probability in the second rule probability, P 2n Is the nth target rule set probability in the second rule probabilities.
Further, wind control rules of the target enterprises are generated according to the rule probabilities, and further risks of the target enterprises can be prevented and controlled according to the wind control rules, so that data security of the enterprises is improved.
In the embodiment of the present invention, when the wind control rule generating module 102 generates the first wind control rule and the second wind control rule of the preset target enterprise according to the rule probability, the wind control rule generating module is specifically configured to:
sequencing the rule probabilities according to a sequencing order from big to small to obtain a rule probability sequence;
selecting the highest target rule set in the rule probability sequence as the first wind control rule;
and selecting the next highest target rule set in the rule probability sequence as the second wind control rule.
In detail, the obtained rule probabilities are ordered in order from big to small to obtain a rule probability sequence, for example, the rule probability is { P } 11 ,P 12 ,...,P 21, ...,P 2n Sorting according to the order from big to small to obtain a rule probability sequence of { P } 21, ,P 11, ,...,P 12, ...,P 2n Then P is selected in the rule probability sequence 21, Corresponding target rule setThe corresponding logic expression is used as the first wind control rule, and P is selected from a rule probability sequence 11, Corresponding target rule setThe corresponding logical expression is used as the second wind And (5) controlling rules.
Further, the first wind control rule and the second wind control rule are fused, so that a more comprehensive wind control rule can be obtained, and the accuracy of enterprise risk prevention and control is improved.
In the embodiment of the invention, the essence of the fusion wind control rule is a rule set formed by logic operations such as AND, OR AND the like on a plurality of variables, so that the logic expression of the first wind control rule AND the logic expression of the second wind control rule are required to be combined to obtain a more comprehensive wind control rule.
The wind control data extraction module 103 is configured to perform rule fusion on the first wind control rule and the second wind control rule by using a preset data fusion algorithm to obtain a fused wind control rule, and extract wind control data in the fused wind control rule by using a preset probability algorithm;
in the embodiment of the present invention, when the wind control data extraction module 103 performs rule fusion on the first wind control rule and the second wind control rule by using a preset data fusion algorithm to obtain a fused wind control rule, the method is specifically used for:
acquiring a first rule logic expression in the first wind control rule;
acquiring a second rule logic expression in the second wind control rule;
and fusing the first rule logic expression and the second rule logic expression by using the data fusion algorithm to obtain the fused wind control rule, wherein the data fusion algorithm is as follows:
G＝R∪Q
Wherein G is the fusion wind control rule, R is the first rule logic expression, and Q is the second rule logic expression.
In detail, the logic expression of the first wind control rule and the logic expression of the second wind control rule are merged through a data fusion algorithm, and the obtained rule union set is used as a fusion wind control rule. If the logic expression of the first wind control rule is P= { R is less than or equal to 5 U.J is less than or equal to 9 or R is less than or equal to 5 U.H is less than or equal to 3 or J is less than or equal to 9 U.H is less than or equal to 3}, the logic expression of the second wind control rule is Q= { R>5∪J>9 or R>5∪H>3 or J>9∪H>3, if the selected logic expression in the first rule logic expression is P= { R is less than or equal to 5 U.J is less than or equal to 9}, the selected logic expression in the second rule logic expression is Q= { R>5∪H>3, then fuse the wind control rule G= { R ε R + ∪H>3 U.J.gtoreq.p., wherein R is the mean, J is the variance, H is the standard deviation, R + And if any one of R, J, H is absent in the fusion wind control rule, J is more than or equal to rho, so that the fusion accuracy is improved, and the applicability of the fusion rule is ensured.
Further, according to sample data corresponding to the fusion wind control rule, wind control data with minimum risk is extracted, and further wind control data is compared with enterprise data acquired in real time to determine enterprise risk.
In the embodiment of the present invention, when the wind control data extraction module 103 extracts the wind control data in the fused wind control rule by using a preset probability algorithm, the wind control data extraction module is specifically configured to:
calculating the probability of data risk in the fusion wind control rule by using the wind control rule model;
and determining wind control data in the fusion wind control rule according to the data risk probability by using the probability algorithm, wherein the probability algorithm is as follows:
wherein,,for the wind control data, < >>Predicting a model theta of a stroke control rule under a sample feature vector x to be +.>Time probability (I)>Predicting a model theta of a stroke control rule under a sample feature vector x to be +.>Time probability, arg min is a minimum function.
In detail, the wind control rule model is utilized to calculate the probability of the target data in the fusion wind control rule under the condition of risk and risk-free, and in addition, the data with the same probability exists, so that the probability of the same data under the condition of risk and risk-free is required to calculate the minimum value of the same data, and the data corresponding to the minimum value is wind control data, namely the wind control data with the minimum risk.
Specifically, in the probability algorithmIndicating that there is a risk of- >Indicating no risk, re-dividing the enterprise sample data by fusing the wind control rules, and if yes +.>The data risk probability of (2) isThe probability of (2) is 0.5 +.>The probability of (2) is 0.5 +.>The data risk probability of (2) is->The probability of (2) is 0, < >>The probability of (1),>the data risk probability of (2) isThe probability of (1),>the probability of (2) is 0, < >>The data risk probability of (2) isThe probability of (2) is 0, < >>The probability of 1 is thatObtained->Is->
Further, the wind control data is the data with the minimum risk, and the wind control data of the target enterprise data acquired in real time is compared to determine the risk degree of the real-time target enterprise data.
The contrast calculating module 104 is configured to obtain real-time target enterprise data, and calculate a contrast between the wind control data and the real-time target enterprise data through a preset contrast algorithm;
in the embodiment of the present invention, the real-time target enterprise data is real-time data generated in the enterprise operation process, such as financial data, operation data, management data, etc., where the real-time target enterprise data may be obtained through computer sentences (such as Java sentences, python sentences, etc.) having a data grabbing function.
Further, the wind control data is compared with the real-time target enterprise data to determine whether the real-time target enterprise data has risks, so that the safe operation of the enterprise is further ensured.
In the embodiment of the invention, the contrast is the similarity or the distinction between the wind control data and the real-time target enterprise data.
In the embodiment of the present invention, when the contrast calculating module 104 calculates the contrast between the wind control data and the real-time target enterprise data through a preset contrast algorithm, the contrast calculating module is specifically configured to:
performing core semantic extraction on the wind control data to obtain wind control data semantics, and performing core semantic extraction on the real-time target enterprise data to obtain real-time enterprise data semantics;
generating a semantic tree according to the wind control data semantics and the real-time enterprise data semantics;
calculating semantic weights among each edge in the semantic tree by using a preset weight algorithm, wherein the weight algorithm is as follows:
wherein w is i,j For the semantic weight, c, between the wind control data semantic i and the real-time enterprise data semantic j 1 As a first weight factor, c 2 For the second weight factor, d is the depth of the semantic tree, k i The method is characterized in that the method is used for layer numbering of wind control data semantics i, alpha is a weight adjustment parameter, pi is a circumference rate, sin is a sine function, log is a log function, and k is a j Layer numbering of real-time enterprise data semantics j, m being the total number of nodes of the semantic tree;
Calculating the contrast of the wind control data and the real-time target enterprise data according to the semantic weight by using the following contrast algorithm:
wherein s is u,v For the contrast ratio of the wind control data u and the real-time target enterprise data v, delta is a contrast adjustment parameter, w i,j And m is the total number of nodes of the semantic tree, and is the semantic weight between the data semantic i and the data semantic j.
In detail, a pre-constructed semantic analysis model performs core semantic extraction on the wind control data and the real-time target enterprise data to obtain control data semantics and real-time enterprise data semantics. Wherein the semantic analysis model includes, but is not limited to, an NLP (Natural Language Processing ) model, HMM (Hidden Markov Model, hidden markov model). And carrying out convolution, pooling and other operations on the wind control data and the real-time target enterprise data by utilizing a pre-constructed semantic analysis model so as to extract low-dimensional feature expressions of the wind control data and the real-time target enterprise data, mapping the extracted low-dimensional feature expressions to a pre-constructed high-dimensional space to obtain high-dimensional feature expressions of the low-dimensional features, and selectively outputting the high-dimensional feature expressions by utilizing a preset activation function to obtain control data semantics and real-time enterprise data semantics.
Specifically, a root node is set, wind control data semantics and real-time enterprise data semantics are in one-to-one correspondence, and an N-dimensional semantic tree is generated, wherein if the wind control data semantics are F= { X 1 ,X 2 ,...,X N Real-time enterprise data semantics are s= { Y } 1 ,Y 2 ,...,Y N X is to 1 ，Y 1 As a subtree in the semantic tree, X will be 2 ，Y 2 As a subtree in the semantic tree, thereby obtaining an N-dimensional semantic tree, and further calculating each X according to a weight algorithm N ，Y N Edge weight between, wherein c in the weight algorithm 1 As a first weight factor, c 2 For the second weight factor is custom set, then c 1 +c 2 The weight adjustment parameter is defined by =1, and is generally 0.5, i.e. the weight adjustment parameter can be used for semantic weightAnd adjusting the calculation error of the weight, and adjusting the semantic weight through the weight adjustment parameter when the calculated semantic weight is too large, so that the semantic weight ensures a controllable range and meets the applicability of the semantic weight.
Further, the contrast ratio of the wind control data and the real-time target enterprise data is calculated according to a contrast ratio algorithm, wherein delta is a contrast adjustment parameter, and is generally 0.5, so that the situation that the contrast ratio is too large or too small is prevented, and therefore, the risk value of the target enterprise can be determined according to the contrast ratio, and the larger the contrast ratio is, the larger the risk of the target enterprise is, the smaller the contrast ratio is, and the smaller the risk of the target enterprise is.
The enterprise risk determination module 105 is configured to calculate, according to a preset risk evaluation algorithm, a risk value of the target enterprise according to the contrast and a preset rule trigger threshold, determine a risk of the target enterprise according to the risk value, and generate a risk prevention and control policy according to the risk of the target enterprise.
In the embodiment of the invention, the rule triggering threshold refers to a limiting value of the real-time target enterprise data triggering fusion wind control rule, and only if the rule value of the real-time target enterprise data reaches the rule triggering threshold, the risk brought by the real-time target enterprise data to the enterprise is necessarily determined, the risk detection efficiency can be improved by setting the rule triggering threshold, and for some data with extremely small risk, direct filtering processing can be selected without detecting the risk degree of the data, so that the risk detection efficiency of the enterprise is improved.
In the embodiment of the present invention, when the risk value of the target enterprise is calculated by the enterprise risk determination module 105 according to the contrast and the preset rule trigger threshold through a preset risk assessment algorithm, the method is specifically used for:
acquiring a rule value of the real-time target enterprise data;
when the rule value is larger than a preset rule trigger threshold, calculating a risk value of the target enterprise according to the contrast by using the risk assessment algorithm, wherein the risk assessment algorithm is as follows:
Wherein F is the risk value, s is the contrast, ε k Risk compensation coefficient, τ, for the kth target risk indicator k And n is the number of target risk indexes, and the severity of the kth target risk index is the severity of the kth target risk index.
In detail, the rule value is a measure for measuring whether the real-time target enterprise data has risk, wherein the weight of the real-time target enterprise data can be determined by using a hierarchical analysis method, and then the weight is used as the rule value of the real-time target enterprise data, and the real-time target enterprise data can be subjectively scored from the importance degree, the important time node and the enterprise data requirement degree of the real-time target enterprise data to obtain the weight of the real-time target enterprise data, and then the rule value of the real-time target enterprise data is obtained.
Specifically, when the rule value is greater than a preset rule threshold, calculating an influence value of real-time target enterprise data on a target enterprise through the risk assessment algorithm, wherein epsilon in the risk assessment algorithm k The risk compensation coefficient is a value set for each target risk index, and can be used for further determining the risk value of the real-time target enterprise data when the fusion wind control rule cannot completely meet the risk detection of the real-time target enterprise data, and τ k The severity of the kth target risk index is determined by using a hierarchical analysis method, wherein the target risk index comprises indexes such as enterprise information, industry environment, management system, financial condition and the like.
Further, according to comparison of the real-time target enterprise data and the wind control data, the risk value of the real-time target enterprise data can be determined, and then a prevention and control strategy of the target enterprise for risk prevention and control is generated according to the risk value, so that operation and data security of the target enterprise are guaranteed.
In the embodiment of the invention, the risk prevention and control strategy refers to selecting different prevention and control strategies from a risk strategy library according to different risk grade degrees, wherein the risk strategy library comprises a risk avoiding strategy, a risk controlling strategy, a risk dispersing and neutralizing strategy, a risk bearing strategy and a risk transferring strategy.
In the embodiment of the present invention, when the enterprise risk determination module 105 generates a risk prevention and control policy according to the target enterprise risk, the method is specifically used for:
determining a risk level of the target enterprise risk according to the risk value;
and generating enterprise risk prevention and control strategies corresponding to the risk grades through a preset risk strategy library.
In detail, when the risk value is {5,30}, the risk level is low, when the risk value is {31,60}, the risk level is medium, and when the risk value is {61,100}, the risk level is high, so that different enterprise risk prevention and control strategies can be determined according to different risk levels.
Specifically, when the risk level is low, a risk avoidance policy or a risk control policy may be selected from the risk policy library; when the risk level is a middle level, a dispersion and neutralization risk strategy can be selected from a risk strategy library; when the risk level is high, a risk bearing strategy and a risk transferring strategy can be selected from a risk strategy library, wherein a risk avoiding strategy refers to a simple method which is most feasible for avoiding risks when losses caused by risks cannot be offset by profits possibly obtained by the project, and a risk controlling strategy refers to reducing losses caused by risks, namely controlling risks. The risk bearing strategy can only bear the loss caused by the risk by itself when the risk cannot be avoided and the risk cannot be completely controlled or dispersed and neutralized.
Referring to fig. 2, a flow chart of an operation method of the enterprise risk analysis system based on comparison data according to an embodiment of the invention is shown. In this embodiment, the method for operating the enterprise risk analysis system based on comparison data includes:
s1, acquiring enterprise sample data, extracting enterprise data characteristics of the enterprise sample data, and training the enterprise data characteristics by using a preset regression model to obtain an air control rule model;
S2, determining rule probability of the enterprise sample data according to the wind control rule model, and generating a first wind control rule and a second wind control rule of a preset target enterprise according to the rule probability;
s3, carrying out rule fusion on the first wind control rule and the second wind control rule by using a preset data fusion algorithm to obtain a fused wind control rule, and extracting wind control data in the fused wind control rule by using a preset probability algorithm;
s4, acquiring real-time target enterprise data, and calculating the contrast between the wind control data and the real-time target enterprise data through a preset contrast algorithm;
s5, calculating a risk value of the target enterprise according to the contrast and a preset rule triggering threshold through a preset risk assessment algorithm, determining the risk of the target enterprise according to the risk value, and generating a risk prevention and control strategy according to the risk of the target enterprise.
According to the embodiment of the invention, through the enterprise data characteristics of the enterprise sample data, the regression model is further utilized to train the enterprise data characteristics to obtain the wind control rule model, so that the wind control rule model is beneficial to generating enterprise wind control rules corresponding to the enterprise data according to the wind control rule model, different enterprise wind control rules are fused to obtain the fused wind control rules, the comprehensive wind control rules are beneficial to obtaining, and the accuracy of enterprise risk detection is ensured; and extracting wind control data in the fusion wind control rule, comparing the wind control data with real-time target enterprise data according to the minimum risk of the wind control data to obtain data contrast, determining a risk value of the real-time target enterprise data based on the data contrast, determining the risk of the target enterprise according to the risk value, generating a risk prevention and control strategy according to the risk of the target enterprise, performing risk prevention and control on the target enterprise according to the risk prevention and control strategy, and improving the operation and data safety of the target enterprise. Therefore, the enterprise risk analysis system and method based on comparison data can solve the problem of lower data security when a user performs enterprise risk analysis.
In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods may be implemented in other ways. For example, the system embodiments described above are merely illustrative, e.g., the division of the modules is merely a logical function division, and other manners of division may be implemented in practice.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. Multiple units or systems as set forth in the system claims may also be implemented by means of one unit or system in software or hardware. The terms first, second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.
Claims (10)
1. An enterprise risk analysis system based on comparison data is characterized by comprising an air control rule model generation module, an air control rule generation module, an air control data extraction module, a contrast calculation module and an enterprise risk determination module, wherein,
The wind control rule model generation module is used for acquiring enterprise sample data, extracting enterprise data characteristics of the enterprise sample data, and training the enterprise data characteristics by utilizing a preset regression model to obtain a wind control rule model;
the wind control rule generation module is used for determining rule probability of the enterprise sample data according to the wind control rule model and generating a first wind control rule and a second wind control rule of a preset target enterprise according to the rule probability;
the wind control data extraction module is configured to perform rule fusion on the first wind control rule and the second wind control rule by using a preset data fusion algorithm to obtain a fused wind control rule, and extract wind control data in the fused wind control rule by using a preset probability algorithm, where the method is specifically configured to, when calculating a data risk probability in the fused wind control rule by using the wind control rule model:
and determining wind control data in the fusion wind control rule according to the data risk probability by using the probability algorithm, wherein the probability algorithm is as follows:
wherein,,for the wind control data, < >>Predicting a model theta of a stroke control rule under a sample feature vector x to be +. >Time probability (I)>Predicting a model theta of a stroke control rule under a sample feature vector x to be +.>Time probability, arg min is a minimum function;
the contrast computing module is used for acquiring real-time target enterprise data and computing the contrast between the wind control data and the real-time target enterprise data through a preset contrast algorithm;
the enterprise risk determination module is used for calculating a risk value of the target enterprise according to the contrast and a preset rule trigger threshold through a preset risk assessment algorithm, determining a risk of the target enterprise according to the risk value, and generating a risk prevention and control strategy according to the risk of the target enterprise.
2. The comparison data based enterprise risk analysis system of claim 1, wherein the wind control rule model generation module, when extracting enterprise data features of the enterprise sample data, is specifically configured to:
dividing the enterprise sample data according to preset data attributes to obtain divided sample data;
vector conversion is carried out on the divided sample data one by using a preset vector conversion model, so that an enterprise sample vector is obtained;
the enterprise sample vector is treated as the enterprise data feature.
3. The system for analyzing enterprise risk based on comparison data of claim 1, wherein the wind control rule model generation module is specifically configured to, when training the enterprise data features by using a preset regression model to obtain a wind control rule model:
training the enterprise data features by using the regression model to obtain training parameters, wherein the regression model is as follows:
z＝w T x+b
wherein p (y=1|x) is the probability of the sample feature vector x when the prediction type y=1, x is the sample feature vector in the enterprise data feature, z is a training value, T is a transposed symbol, w is a first training parameter, and b is a second training parameter;
updating the training parameters through a preset loss function to obtain optimal training parameters;
and assigning the optimal training parameters to the regression model to obtain the wind control rule model.
4. The comparison data based enterprise risk analysis system of claim 1, wherein the wind control rule generation module is operable, when determining the rule probabilities for the enterprise sample data from the wind control rule model, to:
acquiring feature score values of the enterprise data features;
Dividing the enterprise sample data into a first target rule set when the feature segmentation value is smaller than or equal to a preset feature segmentation threshold value;
dividing the enterprise sample data into a second target rule set when the feature segmentation value is larger than a preset feature segmentation threshold value;
determining a first rule probability of the first target rule set according to the wind control rule model, and determining a second rule probability of the second target rule set by using the wind control rule model;
and collecting the first rule probability and the second rule probability as rule probabilities of the enterprise sample data.
5. The system for analyzing risk of enterprises based on comparison data according to claim 1, wherein the wind control rule generating module is configured to, when generating the first wind control rule and the second wind control rule of the preset target enterprise according to the rule probability:
sequencing the rule probabilities according to a sequencing order from big to small to obtain a rule probability sequence;
selecting the highest target rule set in the rule probability sequence as the first wind control rule;
and selecting the next highest target rule set in the rule probability sequence as the second wind control rule.
6. The enterprise risk analysis system based on comparison data of claim 1, wherein the wind control data extraction module is configured to, when performing rule fusion on the first wind control rule and the second wind control rule by using a preset data fusion algorithm to obtain a fused wind control rule:
acquiring a first rule logic expression in the first wind control rule;
acquiring a second rule logic expression in the second wind control rule;
and fusing the first rule logic expression and the second rule logic expression by using the data fusion algorithm to obtain the fused wind control rule, wherein the data fusion algorithm is as follows:
G＝R∪Q
wherein G is the fusion wind control rule, R is the first rule logic expression, and Q is the second rule logic expression.
7. The comparison data-based enterprise risk analysis system of claim 1, wherein the contrast calculation module is configured to, when calculating the contrast of the wind control data and the real-time target enterprise data by a preset contrast algorithm:
performing core semantic extraction on the wind control data to obtain wind control data semantics, and performing core semantic extraction on the real-time target enterprise data to obtain real-time enterprise data semantics;
Generating a semantic tree according to the wind control data semantics and the real-time enterprise data semantics;
calculating semantic weights among each edge in the semantic tree by using a preset weight algorithm, wherein the weight algorithm is as follows:
wherein w is i,j For the semantic weight, c, between the wind control data semantic i and the real-time enterprise data semantic j 1 As a first weight factor, c 2 For the second weight factor, d is the depth of the semantic tree, k i The method is characterized in that the method is used for layer numbering of wind control data semantics i, alpha is a weight adjustment parameter, pi is a circumference rate, sin is a sine function, log is a log function, and k is a j Layer numbering of real-time enterprise data semantics j, m being the total number of nodes of the semantic tree;
calculating the contrast of the wind control data and the real-time target enterprise data according to the semantic weight by using the following contrast algorithm:
wherein s is u,v For the contrast ratio of the wind control data u and the real-time target enterprise data v, delta is a contrast adjustment parameter, w i,j And m is the total number of nodes of the semantic tree, and is the semantic weight between the data semantic i and the data semantic j.
8. The comparison data-based enterprise risk analysis system of claim 1, wherein the enterprise risk determination module is configured to, when calculating the risk value of the target enterprise according to the contrast and a preset rule trigger threshold by a preset risk assessment algorithm:
Acquiring a rule value of the real-time target enterprise data;
when the rule value is larger than a preset rule trigger threshold, calculating a risk value of the target enterprise according to the contrast by using the risk assessment algorithm, wherein the risk assessment algorithm is as follows:
wherein F is the risk value, s is the contrast, ε k Risk compensation coefficient, τ, for the kth target risk indicator k And n is the number of target risk indexes, and the severity of the kth target risk index is the severity of the kth target risk index.
9. The comparison data-based enterprise risk analysis system of claim 1, the enterprise risk determination module, when generating a risk prevention policy from the target enterprise risk, is specifically configured to:
determining a risk level of the target enterprise risk according to the risk value;
and generating enterprise risk prevention and control strategies corresponding to the risk grades through a preset risk strategy library.
10. An operation method of an enterprise risk analysis system based on comparison data is characterized in that the method is suitable for the enterprise risk analysis system based on the comparison data, the system comprises a wind control rule model generation module, a wind control rule generation module, a wind control data extraction module, a contrast calculation module and an enterprise risk determination module, and the method comprises the following steps:
Acquiring enterprise sample data, extracting enterprise data characteristics of the enterprise sample data, and training the enterprise data characteristics by using a preset regression model to obtain a wind control rule model;
determining rule probability of the enterprise sample data according to the wind control rule model, and generating a first wind control rule and a second wind control rule of a preset target enterprise according to the rule probability;
performing rule fusion on the first wind control rule and the second wind control rule by using a preset data fusion algorithm to obtain a fused wind control rule, and extracting wind control data in the fused wind control rule by using a preset probability algorithm;
acquiring real-time target enterprise data, and calculating the contrast between the wind control data and the real-time target enterprise data through a preset contrast algorithm;
and calculating a risk value of the target enterprise according to the contrast and a preset rule triggering threshold through a preset risk assessment algorithm, determining the risk of the target enterprise according to the risk value, and generating a risk prevention and control strategy according to the risk of the target enterprise.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310469686.8A CN116485185A (en) | 2023-04-23 | 2023-04-23 | Enterprise risk analysis system and method based on comparison data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310469686.8A CN116485185A (en) | 2023-04-23 | 2023-04-23 | Enterprise risk analysis system and method based on comparison data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116485185A true CN116485185A (en) | 2023-07-25 |
Family
ID=87226549
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310469686.8A Pending CN116485185A (en) | 2023-04-23 | 2023-04-23 | Enterprise risk analysis system and method based on comparison data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116485185A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116977062A (en) * | 2023-08-04 | 2023-10-31 | 江苏臻云技术有限公司 | Risk label management system and method for financial business |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110443458A (en) * | 2019-07-05 | 2019-11-12 | 深圳壹账通智能科技有限公司 | Methods of risk assessment, device, computer equipment and storage medium |
| CN112801431A (en) * | 2021-04-13 | 2021-05-14 | 广东浩迪智云技术有限公司 | Enterprise operation risk assessment method and system based on artificial intelligence analysis |
| CN114612251A (en) * | 2022-03-14 | 2022-06-10 | 平安科技（深圳）有限公司 | Risk assessment method, device, equipment and storage medium |
| CN115082225A (en) * | 2022-07-21 | 2022-09-20 | 天津金城银行股份有限公司 | Enterprise financing risk assessment method and device |
-
2023
- 2023-04-23 CN CN202310469686.8A patent/CN116485185A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110443458A (en) * | 2019-07-05 | 2019-11-12 | 深圳壹账通智能科技有限公司 | Methods of risk assessment, device, computer equipment and storage medium |
| CN112801431A (en) * | 2021-04-13 | 2021-05-14 | 广东浩迪智云技术有限公司 | Enterprise operation risk assessment method and system based on artificial intelligence analysis |
| CN114612251A (en) * | 2022-03-14 | 2022-06-10 | 平安科技（深圳）有限公司 | Risk assessment method, device, equipment and storage medium |
| CN115082225A (en) * | 2022-07-21 | 2022-09-20 | 天津金城银行股份有限公司 | Enterprise financing risk assessment method and device |
Non-Patent Citations (1)
| Title |
|---|
| 刘东海;: "基于多因素分析的ERP实施风险管理的研究", 科技广场, no. 03 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116977062A (en) * | 2023-08-04 | 2023-10-31 | 江苏臻云技术有限公司 | Risk label management system and method for financial business |
| CN116977062B (en) * | 2023-08-04 | 2024-01-23 | 江苏臻云技术有限公司 | Risk label management system and method for financial business |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108717408B (en) | Sensitive word real-time monitoring method, electronic equipment, storage medium and system | |
| CN113094200B (en) | Application program fault prediction method and device | |
| CN111340148B (en) | Training method of business classification model, business classification method and terminal | |
| US20210097343A1 (en) | Method and apparatus for managing artificial intelligence systems | |
| CN113722493B (en) | Text classification data processing method, apparatus and storage medium | |
| CN113778894B (en) | Method, device, equipment and storage medium for constructing test cases | |
| CN111080117A (en) | Method and device for constructing equipment risk label, electronic equipment and storage medium | |
| CN117271767B (en) | Operation and maintenance knowledge base establishing method based on multiple intelligent agents | |
| CN116485185A (en) | Enterprise risk analysis system and method based on comparison data | |
| CN116402630A (en) | Financial risk prediction method and system based on characterization learning | |
| CN115225336A (en) | Vulnerability availability calculation method and device for network environment | |
| CN117235608B (en) | Risk detection method, risk detection device, electronic equipment and storage medium | |
| CN112364011A (en) | Online data model management device, method and system | |
| CN114816962B (en) | ATTENTION-LSTM-based network fault prediction method | |
| CN114969334B (en) | Abnormal log detection method and device, electronic equipment and readable storage medium | |
| CN116795978A (en) | Complaint information processing method and device, electronic equipment and medium | |
| CN116644148A (en) | Keyword recognition method and device, electronic equipment and storage medium | |
| CN115660695A (en) | Customer service personnel label portrait construction method and device, electronic equipment and storage medium | |
| Hamad et al. | Sentiment analysis of restaurant reviews in social media using naïve bayes | |
| US11397853B2 (en) | Word extraction assistance system and word extraction assistance method | |
| CN114741515A (en) | Social network user attribute prediction method and system based on graph generation | |
| Ibrishimova et al. | Automating incident classification using sentiment analysis and machine learning | |
| CN114528908A (en) | Network request data classification model training method, classification method and storage medium | |
| CN114186646A (en) | Block chain abnormal transaction identification method and device, storage medium and electronic equipment | |
| CN117291615B (en) | Visual contrast analysis method and device for overcoming anti-fraud based on network payment |
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 |