CN109829629A - Generation method, device, computer equipment and the storage medium of risk analysis reports - Google Patents
Generation method, device, computer equipment and the storage medium of risk analysis reports Download PDFInfo
- Publication number
- CN109829629A CN109829629A CN201910012944.3A CN201910012944A CN109829629A CN 109829629 A CN109829629 A CN 109829629A CN 201910012944 A CN201910012944 A CN 201910012944A CN 109829629 A CN109829629 A CN 109829629A
- Authority
- CN
- China
- Prior art keywords
- risk
- resource
- case
- target
- analysis
- 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.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This application involves generation method, device, computer equipment and the storage mediums of a kind of risk analysis reports based on machine learning.This method comprises: receiving the risk analysis request that terminal is sent;Risk analysis requests to carry the target resource identifier of selected virtual resource;It determines the corresponding monitored object of target resource identifier, obtains the monitoring data of monitored object, monitoring data is inputted into risk analysis model, obtain corresponding risk score;When risk score is more than threshold value, target resource identifier is labeled as risk case;The corresponding similar cases of risk case are determined according to monitoring data;Multiple risk points based on similar cases identification risk case;Multiple risk points are connected, the corresponding risk clue of risk case is generated;Based on risk score, similar cases and risk clue, the corresponding risk analysis reports of target resource identifier are generated, risk analysis reports are fed back into terminal.Risk analysis efficiency and accuracy can be improved using this method.
Description
Technical field
This application involves field of computer technology, more particularly to a kind of generation method of risk analysis reports, device, meter
Calculate machine equipment and storage medium.
Background technique
Risk monitoring and control is necessary service link in a variety of industries.For example, needing in financial industry to virtual resource
Monitored object is monitored with the presence or absence of default risk.With the development of computing technique, market also occurs some for risk
The tool of monitoring, but these tools can only provide the mechanical comparison and simple credit rating function of risk indicator mostly, it is difficult to
Accomplish the risk point for finding monitored object in real time.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of risk for improving risk analysis efficiency and accuracy
Generation method, device, computer equipment and the storage medium of analysis report.
A kind of generation method of risk analysis reports, which comprises receive the risk analysis request that terminal is sent;Institute
State the target resource identifier that risk analysis request carries selected virtual resource;Determine the corresponding prison of the target resource identifier
Object is controlled, the monitoring data of the monitored object is obtained, monitoring data is inputted into risk analysis model, obtain corresponding risk and comment
Point;When the risk score is more than threshold value, the target resource identifier is labeled as risk case;According to the monitoring data
Determine the corresponding similar cases of the risk case;Multiple risk points of the risk case are identified based on the similar cases;
The multiple risk point is connected, the corresponding risk clue of the risk case is generated;Based on the risk score, similar cases
And risk clue, the corresponding risk analysis reports of the target resource identifier are generated, the risk analysis reports are fed back into institute
State terminal.
In one embodiment, before receiving the risk analysis request that terminal is sent, further includes: receive what terminal was sent
Resource acquisition request;The resource acquisition request carries Target Attribute values;The resource factor for obtaining multiple virtual resources, by mesh
The resource factor for marking attribute value and each virtual resource inputs regulation-control model, obtains the resource mark of each virtual resource;
Corresponding expert model is chosen according to resource mark, resource factor input expert model is obtained into respective virtual resource
Prediction attribute value;The virtual resource that screening prediction attribute value and Target Attribute values match, is denoted as target resource;According to described
The corresponding expert model of target resource, determines the property of the target resource;The resource information of the target resource is obtained,
The resource information and property are back to terminal;Make multiple targets of the terminal according to the property to push
Resource is further screened.
In one embodiment, described that monitoring data is inputted into risk analysis model, corresponding risk score is obtained, is wrapped
It includes: obtaining training sample, screen target sample in the training sample;The first submodel is carried out based on the training sample
Training, obtains health analysis model;The second submodel is trained based on the target sample, obtains bankruptcy analysis model;
The monitoring data is inputted into the health analysis model, obtains the health index of the monitored object;By the monitoring data
The bankruptcy analysis model is inputted, the bankruptcy index of the monitored object is obtained;Referred to based on the health index and the bankruptcy
Number, calculates the risk score.
In one embodiment, described that the corresponding similar cases of the risk case are determined according to the monitoring data, packet
It includes: determining the affiliated industry type of the corresponding monitored object of virtual resource;Obtain the risk data of the risk case;The risk
Data include the risk data of the risk data of the virtual resource, the risk data of industry type and monitored object;Institute
State the risk label that the risk case is extracted in risk data;Calculate the risk label and pre-stored multiple history cases
Case label similarity, by the similarity be more than preset value history case marker be similar cases.
In one embodiment, multiple risk points that the risk case is identified based on the similar cases, comprising:
Obtain the risk indicator of multiple timing nodes of the similar cases;Judge that the risk case refers to the presence or absence of identical risk
It marks and the risk case time sequencing of identical risk indicator occurs and whether the similar cases are consistent;If so, really
The identical risk indicator of the last one fixed timing node, is denoted as sign index;By timing node after the sign index
Different risk indicators is labeled as the risk point of the risk case.
In one embodiment, described that the multiple risk point is connected, generate the corresponding risk line of the risk case
Rope, comprising: determine the monitoring period of the risk case;When reaching in the monitoring period, the acquisition monitoring is returned
The step of monitoring data of object, analyzes the risk case in current monitor week if the risk score is still more than threshold value
The risk point of phase;By the risk point series connection in multiple monitoring periods, the corresponding risk clue of the discovery main body is generated.
A kind of generating means of risk analysis reports, described device include: risk score module, are sent for receiving terminal
Risk analysis request;The risk analysis request carries the target resource identifier of selected virtual resource;Determine the mesh
The corresponding monitored object of resource identification is marked, the monitoring data of the monitored object is obtained, monitoring data is inputted into risk analysis mould
Type obtains corresponding risk score;Clue collection module, for when the risk score is more than threshold value, the target to be provided
Source mark is labeled as risk case;The corresponding similar cases of the risk case are determined according to the monitoring data;Based on described
Similar cases identify multiple risk points of the risk case;The multiple risk point is connected, the risk case pair is generated
The risk clue answered;Report generation module generates the mesh for being based on the risk score, similar cases and risk clue
The corresponding risk analysis reports of resource identification are marked, the risk analysis reports are fed back into the terminal.
In one embodiment, described device further includes Screening germplasm module, for receiving the resource acquisition of terminal transmission
Request;The resource acquisition request carries Target Attribute values;The resource factor for obtaining multiple virtual resources, by Target Attribute values
Regulation-control model is inputted with the resource factor of each virtual resource, obtains the resource mark of each virtual resource;According to described
Resource mark chooses corresponding expert model, and resource factor input expert model is obtained the prediction category of respective virtual resource
Property value;The virtual resource that screening prediction attribute value and Target Attribute values match, is denoted as target resource;According to the target resource
Corresponding expert model determines the property of the target resource;The resource information for obtaining the target resource, by the money
Source information and property are back to terminal;Make the terminal according to the property to multiple target resources of push into one
Step screening.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processing
Device realizes the generation method of the risk analysis reports provided in any one embodiment of the application when executing the computer program
The step of.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
The step of generation method of the risk analysis reports provided in any one embodiment of the application is provided when row.
Generation method, device, computer equipment and the storage medium of above-mentioned risk analysis reports, the wind sent according to terminal
Dangerous analysis request can determine the monitored object for the target resource that user selectes;The monitoring data of monitored object is obtained, and will prison
It controls data and inputs risk analysis model, available corresponding risk score;When the risk score is more than threshold value, can incite somebody to action
The target resource identifier is labeled as risk case;It can determine that the risk case is corresponding similar according to the monitoring data
Case;It can identify to obtain multiple risk points of the risk case based on the similar cases;By the multiple risk point string
Connection, can be generated the corresponding risk clue of the risk case;It, can based on the risk score, similar cases and risk clue
To generate the corresponding risk analysis reports of the target resource identifier.Since risk analysis model can comprehensively consider kinds of risks
Factor carries out risk profile, improves risk analysis efficiency;The similar case of risk case is further determined that after obtaining risk score
Example, and the risk point being likely to occur based on similar cases predicting monitoring object in following multiple timing nodes, are based on above- mentioned information
The risk analysis reports of generation can be convenient the risk situation that user quickly understands selected virtual resource comprehensively, improve risk point
Analyse precision.
Detailed description of the invention
Fig. 1 is the application scenario diagram of the generation method of one embodiment risk analysis report;
Fig. 2 is the flow diagram of the generation method of one embodiment risk analysis report;
The flow diagram for the step of Fig. 3 is target resource screening in one embodiment;
Fig. 4 is the structural block diagram of the generating means of one embodiment risk analysis report;
Fig. 5 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
The generation method of risk analysis reports provided by the present application can be applied in application environment as shown in Figure 1.Its
In, terminal 102 is communicated with server 104 by network.Wherein, terminal 102 can be, but not limited to be various individual calculus
Machine, laptop, smart phone, tablet computer and portable wearable device, server 104 can use independent server
The either server cluster of multiple servers composition is realized.When user needs to carry out risk analysis to selected virtual resource
When, risk analysis request can be sent to server 104 by terminal 102.Server 104 inquires the target resource that user selectes
Corresponding monitored object is identified, and obtains the monitoring data of monitored object.Server calls risk analysis model, by monitoring data
Risk analysis model is inputted, corresponding risk score is obtained.Server compares whether risk score is more than threshold value.If so, clothes
Target resource identifier is labeled as risk case by business device 104.Server 104 determines the corresponding phase of risk case according to monitoring data
Like case, and multiple risk points based on similar cases identification risk case.Server 104 connects multiple risk points, generates
The corresponding risk clue of risk case.Server 104 is based on risk score, similar cases and risk clue, generates target resource
Corresponding risk analysis reports are identified, risk analysis reports are fed back into terminal 102.Above-mentioned risk analysis reports generating process,
Kinds of risks factor can be comprehensively considered based on risk analysis model and carries out risk profile, improve risk analysis efficiency;It is obtaining
The similar cases of risk case are further determined that after risk score, and based on similar cases predicting monitoring object when following multiple
It is comprehensively quick to can be convenient user based on the risk analysis reports that above- mentioned information generate for the risk point that intermediate node is likely to occur
The risk situation for solving selected virtual resource, improves risk analysis precision.
In one embodiment, it as shown in Fig. 2, providing a kind of generation method of risk analysis reports, answers in this way
For being illustrated for the server in Fig. 1, comprising the following steps:
Step 202, the risk analysis request that terminal is sent is received;Risk analysis request carries selected virtual resource
Target resource identifier.
It is mounted with that virtual resource obtains platform in terminal.When user needs to obtain virtual resource, can be based in terminal
Virtual resource obtains platform and selectes virtual resource, can also request to carry out risk analysis to virtual resource.Virtual resource obtains flat
Platform provides multiple analysis dimension options, as analyzed in comprehensive analysis, financial analysis, the analysis of public opinion, colleague, analyzing in same area
Deng.The virtual resource and analysis dimension that terminal is selected according to user generate risk analysis request, and risk analysis request is sent to
Server.
Step 204, it determines the corresponding monitored object of target resource identifier, obtains the monitoring data of monitored object, will monitor
Data input risk analysis model, obtain corresponding risk score.
Virtual resource can be stock, bond etc..The corresponding monitored object of virtual resource refers to the provider of virtual resource.
Due to the monitored object for different industries type, different risk monitoring and control index significance levels in need of consideration are different.Sufficiently examine
The different attribute feature for considering different industries monitored object, distinguishes monitored object industry, constructs for different industries type
Different risk analysis model.Risk analysis model can be machine learning model.
Monitoring data includes the data of multiple dimensions such as finance, area, industry, law and public sentiment.Different monitoring data point
It Ju You not corresponding data source, acquisition time and data type.Data type includes but is not limited to image, audio, text sum number
Word.Server pre-processes the monitoring data of different types of data.Specifically, for the data of digital form, such as enterprise
Financial data, as evaluation business risk quantitative target key data source, monitoring can be directly applied to after simple process
The generation of the factor.But the data of the data types such as text, image, audio are then needed by refining, quantification treatment, in data
Existing code table carries out unified and standardized processing.
Server calls respective risk analysis model carries out risk scanning to virtual resource.
In one embodiment, monitoring data is inputted into risk analysis model, obtains corresponding risk score, comprising: obtain
Training sample is taken, target sample is screened in training sample;The first submodel is trained based on training sample, obtains health
Analysis model;The second submodel is trained based on target sample, obtains bankruptcy analysis model;Monitoring data is inputted into health
Analysis model obtains the health index of monitored object;Monitoring data is inputted into bankruptcy analysis model, obtains the bankruptcy of monitored object
Index;Based on health index and bankruptcy index, calculation risk scoring.
If the analysis dimension that user selectes is " financial analysis ", the corresponding risk analysis model of server calls.Risk point
Analysis model includes health analysis model and bankruptcy analysis model.Wherein, health analysis model include debt paying ability, operation ability,
The health analysis submodel of the monitoring angle of profitability, business growth ability and the multiple dimensions of quality of getting a profit.Monitoring analysis submodel
It can be and obtained based on LGBM model (Light gradient boosting machine, Fast Field boosting algorithm) training
's.Server extracts monitored object respectively in the monitor control index of multiple monitoring angles, by each monitoring angle in monitoring data
Monitor control index input corresponding health analysis submodel, obtain corresponding health scoring.Server is by multiple monitoring angles
Corresponding health scoring input scoring transformation model, obtains corresponding health index.Scoring transformation model is also possible to be based on
LGBM model training obtains, and can also be and is obtained based on other model trainings, with no restriction to this.
Bankruptcy analysis model can be the bad sample data based on bankrupt enterprise to GBDT model (Gradient
Boosting Decision Tree, gradient promote decision tree) be trained.Compared to health analysis model, bankruptcy point
The sample data used when analysing model training is more concentrated.Server will extract obtained monitor control index input bankruptcy analysis mould
Type obtains the high bankruptcy index of monitored object.It is corresponding to calculate target resource identifier according to health index and bankruptcy index for server
Finance scoring.
If the selected analysis dimension of user is " comprehensive analysis ", server calls each analysis dimension in the manner described above
Risk analysis model calculates the subitem score of corresponding analysis dimension, such as finance scoring.Based on multiple subitem scores, monitoring is calculated
The risk score of object.
Step 206, when risk score is more than threshold value, target resource identifier is labeled as risk case.
Server compares whether risk score is more than threshold value.If so, indicate monitored object default risk with higher, clothes
The target resource identifier is labeled as risk case by business device.
Step 208, the corresponding similar cases of risk case are determined according to monitoring data.
Server has been pre-stored a variety of history cases in case library and the risk portrait of each history case (is denoted as bad
Sample portrait).Bad sample portrait includes multiple risk labels.Risk label is for characterizing what history case occurred in which subject
Problem.With time change, risk means are also possible to change.It, can be to corresponding in order to improve similar cases matching accuracy
Bad sample portrait also carries out dynamic update.
The similarity that server is drawn a portrait by calculating the risk of history case with the risk of current risk case portrait, will
Wherein one or more history cases are determined as the similar cases of risk case.Similar cases refer to and current risk case tool
There is the history case of similar risk feature.Multiple monitor control indexes of the server based on monitored object, generate the multiple of monitored object
Risk label draws a portrait (being denoted as portrait to be matched) using the risk that multiple risk labels generate monitored object.Risk analysis model
It further include whitewashing analysis model.It whitewashes analysis model and is applied not only to predicting monitoring object with the presence or absence of risk behavior, also pass through phase
Forecasting risk clue is matched like case.Server calls whitewash the cosine that analysis model calculates portrait to be matched with bad sample portrait
Similitude obtains similarity.If similarity is more than threshold value, corresponding history case marker is similar cases by server.
Step 210, multiple risk points based on similar cases identification risk case.
Each bad sample portrait is associated with the risk indicator of multiple timing nodes.Server is drawn according to the bad sample to match
As the risk indicator of associated multiple timing nodes predicts the risk clue of monitored object.
In one embodiment, multiple risk points based on similar cases identification risk case, comprising: obtain similar cases
Multiple timing nodes risk indicator;Judge that risk case occurs with the presence or absence of identical risk indicator and risk case
Whether the time sequencing of identical risk indicator is consistent with similar cases;If so, determining the identical of the last one timing node
Risk indicator is denoted as sign index;By the different risk indicators of timing node after sign index labeled as risk case
Risk point.
Different time nodes may be similar from the Risk mode of different history cases.In other words, as time goes by when
The similar cases of preceding risk case may change.Server is from " risk label identical with similar cases " and " identical
Two angles of the time sequencing of risk label " generate risk clue.It specifically may determine that risk case whether there is and similar case
Whether the time sequencing and similar cases that the identical risk indicator of example and same risk index occur are consistent.If it exists to similar case
The time of occurrence of the identical risk indicator of example and same risk index sequence is consistent with similar cases, then server by the last one
The identical risk indicator (being denoted as sign index) of timing node is labeled as a risk point.For example, bad sample companies A has 6
A abnormal index, enterprise B have had already appeared wherein 5 kinds of abnormal indexes and have then predicted to be possible to the 6th kind of abnormal index occur, so as to
Abnormal index in the 6th to be labeled as to a risk point of enterprise B.
Step 212, multiple risk points are connected, generates the corresponding risk clue of risk case.
In one embodiment, multiple risk points are connected, generates the corresponding risk clue of risk case, comprising: determine
The monitoring period of risk case;When monitor the period reach when, return obtain monitored object monitoring data the step of, if risk is commented
Dividing still is more than threshold value, risk point of the analysis risk case in the current monitor period;The risk point in multiple monitoring periods is connected,
Generate the corresponding risk clue of discovery main body.
What the monitoring period can be dynamically determined according to risk score or the industry type of monitored object etc., it is also possible to pre-
If fixed value, it is without limitation.Server in the manner described above determine monitored object it is each monitoring the period risk point,
And multiple risk points are connected sequentially in time, obtain the corresponding risk clue of monitored object.
Step 214, it is based on risk score, similar cases and risk clue, generates the corresponding risk point of target resource identifier
Risk analysis reports are fed back to terminal by analysis report.
Server can determine the corresponding risk of monitored object by clustering to risk score.Each risk
Classification is described with corresponding classification.Indicating risk is carried out according to risk score and the corresponding classification description of affiliated risk.
A score is provided compared to simple, it is explanatory that business can be improved based on natural language progress indicating risk.
Server is based on indicating risk, similar cases and associated bad sample portrait, risk clue and generates risk analysis
Report.In another embodiment, risk analysis reports are also shown the health son scoring that Risk Analysis Process obtains.Hold
Readily understood, health son scoring can be showed in risk analysis reports in the form of the charts such as radar map or histogram.
In the present embodiment, is requested according to the risk analysis that terminal is sent, can determine the prison for the target resource that user selectes
Control object;The monitoring data of monitored object is obtained, and monitoring data is inputted into risk analysis model, available corresponding risk
Scoring;When risk score is more than threshold value, target resource identifier can be labeled as risk case;It can be true according to monitoring data
Determine the corresponding similar cases of risk case;It can identify to obtain multiple risk points of risk case based on similar cases;It will be multiple
Risk point series connection, can be generated the corresponding risk clue of risk case;It, can based on risk score, similar cases and risk clue
To generate the corresponding risk analysis reports of target resource identifier.Since risk analysis model can comprehensively consider kinds of risks factor
Risk profile is carried out, risk analysis efficiency is improved;The similar cases of risk case are further determined that after obtaining risk score, and
Based on the risk point that similar cases predicting monitoring object is likely to occur in following multiple timing nodes, generated based on above- mentioned information
Risk analysis reports can be convenient the risk situation that user quickly understands selected virtual resource comprehensively, and it is accurate to improve risk analysis
Degree.
In one embodiment, as shown in figure 3, further including target before receiving the risk analysis request that terminal is sent
The step of Screening germplasm, specifically includes:
Step 302, the resource acquisition request that terminal is sent is received;Resource acquisition request carries Target Attribute values.
Step 304, the resource factor for obtaining multiple virtual resources, by the resource of Target Attribute values and each virtual resource because
Son input regulation-control model obtains the resource mark of each virtual resource.
Step 306, corresponding expert model is chosen according to resource mark, resource factor input expert model is obtained accordingly
The prediction attribute value of virtual resource.
Step 308, the virtual resource that screening prediction attribute value and Target Attribute values match, is denoted as target resource.
Step 310, according to the corresponding expert model of target resource, the property of target resource is determined.
Step 312, resource information and property are back to terminal by the resource information for obtaining target resource;Make terminal
It is further screened according to multiple target resources of the property to push.
When user needs to obtain virtual resource, platform can be obtained in virtual resource and set Target Attribute values.Target category
Property value can be the acquisition condition of virtual resource set by user.According to virtual resource difference, corresponding Target Attribute values can be with
It is different.For example, corresponding Target Attribute values can be prospective earnings when virtual resource is the financial products such as stock or security
Rate, greateset risk rate etc..Target Attribute values, which can be, obtains the given multiple gear sections of platform or gear in virtual resource
It selectes and obtains in value.
Server has been pre-stored the resource information of the virtual resource in multiple transaction in virtualization pool.Server is also pre-
Store the Mixture of expert model for screening the virtual resource for meeting Target Attribute values set by user.Mixture of expert model packet
Include regulation-control model and multiple expert models.Wherein, regulation-control model chooses which expert model calculates virtual resource for determining
Prediction attribute value.Regulation-control model can be the resource information of multiple virtual resources based on history cycle to EM algorithm
What (Expectation Maximization Algorithm, expectation-maximization algorithm) training obtained.
Assuming that Mixture of expert model includes k expert model.Each expert model is a neural network model.It is different
Expert model be good at processing different data sources data.Data source can be virtual resource provider.Each expert model tool
There is corresponding pattern number.If the resource of a virtual resource is labeled as i (1≤i≤k), i-th of expert model is selected.Root
It can be the data source of identification resource information according to resource mark, and then corresponding expert model can be selected according to data source.
The virtual resource that screening server prediction attribute value and Target Attribute values match, is denoted as target resource.Server
According to the corresponding expert model of target resource, the property of target resource is determined.Different expert models can reflect virtual money
The different property in source.Property refer to resource factor and predict attribute value between relationship, such as resource factor because
The higher linear relationship of the more big corresponding prediction attribute value of subvalue, or prediction attribute value is with resource factor presentation normal distribution
Etc. relationships.Server is by using different expert models, it can be determined that different virtual resources different moments follow which kind of because
The rule of sub-feature expression.Generalized linear regression can be used to characterize in property rule, can also be expanded into multiple-factor, i.e.,
In space using hyperplane as the expression of multiple-factor income characteristic.Server obtains the resource information of target resource, by resource
Information and property are back to terminal.Terminal can further screen multiple target resources of push according to property.
In the present embodiment, user need to only set the Target Attribute values for the virtual resource that expectation obtains, based on preset virtual
Resource acquisition strategy automatic screening meets the target resource set of Target Attribute values, precisely reduces virtual resource and screens range, mentions
High virtual resource obtains efficiency.The property of target resource is further provided, user can be assisted according to personal preference and row
Industry experience etc. carries out postsearch screening, so that the acquisition of virtual resource is more personalized, to can also be improved virtual resource acquisition
Accuracy.
In one embodiment, the corresponding similar cases of risk case are determined according to monitoring data, comprising: determine virtual money
The affiliated industry type of the corresponding monitored object in source;Obtain the risk data of risk case;Risk data includes the wind of virtual resource
The risk data of dangerous data, the risk data of industry type and monitored object;The wind of risk case is extracted in risk data
Dangerous label;Similarity is more than default by the similarity of calculation risk label and the case label of pre-stored multiple history cases
The history case marker of value is similar cases.
Server compares whether risk score is more than first threshold.When risk score is less than or equal to first threshold, table
Show that the risk for the virtual resource that user selectes is weaker, in order to improve risk analysis efficiency, server directly returns risk score
It is back to terminal.When risk score is more than first threshold, indicate that the risk for the virtual resource that user selectes is stronger, in order to improve
Risk analysis accuracy, the virtual resource that server selectes user are labeled as risk case, and further identify risk case
Similar cases.
Server determines that the corresponding monitored object of virtual resource that user selectes and its industry type (are denoted as target industry class
Type).Server can use crawler technology and crawl target line belonging to risk data relevant to virtual resource and virtual resource
The relevant risk data of industry type and risk data relevant to monitored object.For example, server can be by virtual resource
The name of title, the title of target industry type and monitored object is referred to as crawling keyword, crawls from targeted website and includes
The news public feelings information of keyword, and to news public feelings information carry out denoising, by news public feelings information advertisement noise,
Dirty word noise etc. filters out one by one, obtains risk data only comprising body content.
Risk data relevant to virtual resource includes the wind of the risk data of virtual resource, the issuer of virtual resource
The risk data of dangerous data and the affiliated industry of virtual resource is realized to the comprehensive risk identification of virtual resource progress, no longer only
Risk identification is carried out from the financial data of virtual resource and credit rating, is obviously improved the accurate of virtual resource default risk identification
Rate.
Server extracts the risk label of risk case in risk data.Specifically, the data type of risk data
It can be image, audio, text or number etc..When risk data is text or the text being converted to based on image or audio
When, server based on data amount and separator split text.Specifically, server calculates the data volume of text, inspection
Whether measured data amount is more than preset value.When data volume is more than preset value, server obtains preset target data amount, according to mesh
Mark data volume determines the fractionation position of text.Server detection splits whether position is located between adjacent separator.When fractionation position
When setting at a separator, text is split as multiple short sentences in fractionation position by server.When fractionation position is positioned at adjacent
When between separator, text is split as multiple short sentences at any one separator in adjacent separator by server.
Server has been pre-stored the corresponding regular expression of a variety of case types.Regular expression includes one or more wind
Dangerous keyword.Server carries out canonical matching to each short sentence that fractionation obtains respectively according to preset multiple regular expressions.
The risk keyword that the regular expression of successful match includes is respectively labeled as risk label by server.
The case of the preset similarity evaluating model calculation risk label of server calls and pre-stored multiple history cases
The similarity of example label.Specifically, server obtains history case pond.It include the history of a variety of case types in history case pond
Case.Each history case has different case labels, and each case label is provided with risk class.Default case type tool
Body may include financial case, legal case, capital case and operation case etc..By taking financial case type as an example, financial case
The corresponding case label of example type may include the labels such as the Change of Capital Structure, poor fluidity and achievement loss.Server can
With in advance by known promise breaking virtual resource as an example, to promise breaking virtual resource different case types monitoring data carry out
Analysis obtains the case label under different case types and risk class is arranged to case label.
Server matches the risk label of acquisition with the case label in history case pond, when risk label and case
Example tag match success when, then will be written with risk tag match successful story label and its risk class to match record
In.Server according to the risk class of case label and case label recorded in matching record calculate each history case with
The similarity of current risk case.
In one embodiment, the similarity of calculation risk label and the case label of pre-stored multiple history cases,
It include: the term vector that risk label is obtained using default term vector model;All term vectors are input to preparatory trained SVM
In model (Support Vector Machine, support vector machines), the confidence level of term vector Yu each case label is calculated;It will
The highest case label of confidence level is determined as the case label with risk tag match;Obtain the risk of the case label to match
Grade;The similarity of corresponding history case and risk case is calculated according to risk class.
Server obtains the corresponding case label of all history cases, removes duplicate case label, obtains case label
Table.The number that the corresponding case label of each history case of server statistics occurs in case label list, to generate case mark
Sign matrix.Server is calculated in the case where virtual resource has default risk, each case mark after obtaining case label matrix
Existing probability value is checked out, the probability value of acquisition is quantified as the corresponding risk class of each case label.For example, case label
The probability value that " poor fluidity " occurs is 80% to 89%, then the corresponding risk class of case label " poor fluidity " is set as 8.
Server by utilizing presets the term vector that term vector model obtains case keyword.Default term vector model can be
Word2vec model;Server can use the term vector that word2vec model obtains each risk label, and by the word of acquisition
Vector is input to preparatory trained SVM model, and the confidence that risk label matches from each different case labels is calculated
Degree.What the risk data that SVM model can use history case was obtained as training data training.Server is maximum by confidence level
Case label be determined as the case label with case Keywords matching.Pass through the determining case with risk tag match of SVM model
Example label, effectively improves the accuracy of user's intent classifier.Similarity is reached the history case marker of second threshold by server
For similar cases.
In the present embodiment, compared to one score value of user feedback is simply given, depth risk further is carried out to virtual resource
Analysis carries out similar cases push to user, facilitates user to be more clear and accurately understands virtual resource risk place, Jin Erbian
In more accurate screening virtual resource.
It should be understood that although each step in the flow chart of Fig. 2 and Fig. 3 is successively shown according to the instruction of arrow,
But these steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly state otherwise herein, these
There is no stringent sequences to limit for the execution of step, these steps can execute in other order.Moreover, in Fig. 2 and Fig. 3
At least part step may include that perhaps these sub-steps of multiple stages or stage are not necessarily same to multiple sub-steps
One moment executed completion, but can execute at different times, and the execution in these sub-steps or stage sequence is also not necessarily
Be successively carry out, but can at least part of the sub-step or stage of other steps or other steps in turn or
Alternately execute.
In one embodiment, as shown in figure 4, providing a kind of generating means of risk analysis reports, comprising: risk is commented
Sub-module 402, clue collection module 404 and report generation module 406, in which:
Risk score module 402, for receiving the risk analysis request of terminal transmission;Risk analysis request carries selected
Virtual resource target resource identifier;It determines the corresponding monitored object of target resource identifier, obtains the monitoring number of monitored object
According to monitoring data is inputted risk analysis model, obtains corresponding risk score.
Clue collection module 404, for when risk score is more than threshold value, target resource identifier to be labeled as risk case
Example;The corresponding similar cases of risk case are determined according to monitoring data;Multiple risks based on similar cases identification risk case
Point;Multiple risk points are connected, the corresponding risk clue of risk case is generated.
Report generation module 406 generates target resource identifier for being based on risk score, similar cases and risk clue
Risk analysis reports are fed back to terminal by corresponding risk analysis reports.
In one embodiment, which further includes Screening germplasm module 408, for receiving the resource acquisition of terminal transmission
Request;Resource acquisition request carries Target Attribute values;The resource factor for obtaining multiple virtual resources, by Target Attribute values and often
The resource factor of a virtual resource inputs regulation-control model, obtains the resource mark of each virtual resource;It is marked and is chosen according to resource
Resource factor input expert model is obtained the prediction attribute value of respective virtual resource by corresponding expert model;Screening prediction belongs to
Property the value and virtual resource that matches of Target Attribute values, be denoted as target resource;According to the corresponding expert model of target resource, determine
The property of target resource;The resource information for obtaining target resource, is back to terminal for resource information and property;Make end
It is further screened according to multiple target resources of the property to push at end.
In one embodiment, risk score module 402 is also used to obtain training sample, and target is screened in training sample
Sample;The first submodel is trained based on training sample, obtains health analysis model;Based on target sample to the second submodule
Type is trained, and obtains bankruptcy analysis model;Monitoring data is inputted into health analysis model, the health for obtaining monitored object refers to
Number;Monitoring data is inputted into bankruptcy analysis model, obtains the bankruptcy index of monitored object;Based on health index and bankruptcy index,
Calculation risk scoring.
In one embodiment, clue collection module 404 is also used to determine the affiliated row of the corresponding monitored object of virtual resource
Industry type;Obtain the risk data of risk case;Risk data includes the risk number of the risk data of virtual resource, industry type
Accordingly and the risk data of monitored object;The risk label of risk case is extracted in risk data;Calculation risk label and pre-
The history case marker that similarity is more than preset value is similar case by the similarity of the case label of multiple history cases of storage
Example.
In one embodiment, clue collection module 404 is also used to obtain the risk of multiple timing nodes of similar cases
Index;Judge that risk case the time of identical risk indicator occurs with the presence or absence of identical risk indicator and risk case
Whether sequence is consistent with similar cases;If so, determining the identical risk indicator of the last one timing node, it is denoted as sign and refers to
Mark;The different risk indicators of timing node after sign index are labeled as to the risk point of risk case.
In one embodiment, clue collection module 404 is also used to determine the monitoring period of risk case;When the monitoring period
When arrival, return obtain monitored object monitoring data the step of, if risk score is still more than threshold value, analysis risk case exist
The risk point in current monitor period;By the risk point series connection in multiple monitoring periods, the corresponding risk clue of discovery main body is generated.
The specific restriction of generating means about risk analysis reports may refer to above for risk analysis reports
The restriction of generation method, details are not described herein.Modules in the generating means of above-mentioned risk analysis reports can whole or portion
Divide and is realized by software, hardware and combinations thereof.Above-mentioned each module can be embedded in the form of hardware or independently of computer equipment
In processor in, can also be stored in a software form in the memory in computer equipment, in order to processor calling hold
The corresponding operation of the above modules of row.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 5.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is for storing multiple history cases and the corresponding risk portrait of each history case.The computer equipment
Network interface be used to communicate with external terminal by network connection.To realize one when the computer program is executed by processor
The generation method of kind risk analysis reports.
It will be understood by those skilled in the art that structure shown in Fig. 5, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
A kind of computer readable storage medium is stored thereon with computer program, when computer program is executed by processor
The step of generation method of the risk analysis reports provided in any one embodiment of the application is provided.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Instruct relevant hardware to complete by computer program, computer program to can be stored in a non-volatile computer readable
It takes in storage medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, this Shen
Please provided by any reference used in each embodiment to memory, storage, database or other media, may each comprise
Non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
Above embodiments only express the several embodiments of the application, and the description thereof is more specific and detailed, but can not
Therefore it is construed as limiting the scope of the patent.It should be pointed out that for those of ordinary skill in the art,
Under the premise of not departing from the application design, various modifications and improvements can be made, these belong to the protection scope of the application.
Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of generation method of risk analysis reports, which comprises
Receive the risk analysis request that terminal is sent;The risk analysis request carries the target resource of selected virtual resource
Mark;
It determines the corresponding monitored object of the target resource identifier, obtains the monitoring data of the monitored object, by monitoring data
Risk analysis model is inputted, corresponding risk score is obtained;
When the risk score is more than threshold value, the target resource identifier is labeled as risk case;
The corresponding similar cases of the risk case are determined according to the monitoring data;
Multiple risk points of the risk case are identified based on the similar cases;
The multiple risk point is connected, the corresponding risk clue of the risk case is generated;
Based on the risk score, similar cases and risk clue, the corresponding risk analysis report of the target resource identifier is generated
It accuses, the risk analysis reports is fed back into the terminal.
2. the method according to claim 1, wherein being gone back before receiving the risk analysis request that terminal is sent
Include:
Receive the resource acquisition request that terminal is sent;The resource acquisition request carries Target Attribute values;
The resource factor of Target Attribute values and each virtual resource is inputted and is adjusted by the resource factor for obtaining multiple virtual resources
Model is controlled, the resource mark of each virtual resource is obtained;
Corresponding expert model is chosen according to resource mark, resource factor input expert model is obtained into respective virtual
The prediction attribute value of resource;
The virtual resource that screening prediction attribute value and Target Attribute values match, is denoted as target resource;
According to the corresponding expert model of the target resource, the property of the target resource is determined;
The resource information and property are back to terminal by the resource information for obtaining the target resource;Make the terminal
It is further screened according to multiple target resources of the property to push.
3. being obtained the method according to claim 1, wherein described input risk analysis model for monitoring data
Corresponding risk score, comprising:
Training sample is obtained, screens target sample in the training sample;
The first submodel is trained based on the training sample, obtains health analysis model;
The second submodel is trained based on the target sample, obtains bankruptcy analysis model;
The monitoring data is inputted into the health analysis model, obtains the health index of the monitored object;
The monitoring data is inputted into the bankruptcy analysis model, obtains the bankruptcy index of the monitored object;
Based on the health index and the bankruptcy index, the risk score is calculated.
4. the method according to claim 1, wherein described determine the risk case according to the monitoring data
Corresponding similar cases, comprising:
Determine the affiliated industry type of the corresponding monitored object of the virtual resource;
Obtain the risk data of the risk case;The risk data includes the risk data of the virtual resource, industry class
The risk data of type and the risk data of monitored object;
The risk label of the risk case is extracted in the risk data;
The similarity is more than by the similarity for calculating the case label of the risk label and pre-stored multiple history cases
The history case marker of preset value is similar cases.
5. the method according to claim 1, wherein described identify the risk case based on the similar cases
Multiple risk points, comprising:
Obtain the risk indicator of multiple timing nodes of the similar cases;
Judge that the risk case identical risk indicator occurs with the presence or absence of identical risk indicator and the risk case
Time sequencing and the similar cases it is whether consistent;
If so, determining the identical risk indicator of the last one timing node, it is denoted as sign index;
The different risk indicators of timing node after the sign index are labeled as to the risk point of the risk case.
6. generating the wind the method according to claim 1, wherein described connect the multiple risk point
The corresponding risk clue of dangerous case, comprising:
Determine the monitoring period of the risk case;
When reaching in the monitoring period, the step of returning to the monitoring data for obtaining the monitored object, if the risk
Scoring is more than still threshold value, analyzes the risk case in the risk point in current monitor period;
By the risk point series connection in multiple monitoring periods, the corresponding risk clue of the discovery main body is generated.
7. a kind of generating means of risk analysis reports, which is characterized in that described device includes:
Risk score module, for receiving the risk analysis request of terminal transmission;The risk analysis request carries selected
The target resource identifier of virtual resource;It determines the corresponding monitored object of the target resource identifier, obtains the monitored object
Monitoring data is inputted risk analysis model by monitoring data, obtains corresponding risk score;
Clue collection module, for when the risk score is more than threshold value, the target resource identifier to be labeled as risk case
Example;The corresponding similar cases of the risk case are determined according to the monitoring data;The wind is identified based on the similar cases
Multiple risk points of dangerous case;The multiple risk point is connected, the corresponding risk clue of the risk case is generated;
Report generation module generates the target resource identifier for being based on the risk score, similar cases and risk clue
The risk analysis reports are fed back to the terminal by corresponding risk analysis reports.
8. device according to claim 7, which is characterized in that described device further include:
Screening germplasm module, for receiving the resource acquisition request of terminal transmission;The resource acquisition request carries target category
Property value;The resource factor for obtaining multiple virtual resources inputs the resource factor of Target Attribute values and each virtual resource
Regulation-control model obtains the resource mark of each virtual resource;Corresponding expert model is chosen according to resource mark, it will be described
Resource factor input expert model obtains the prediction attribute value of respective virtual resource;Screening prediction attribute value and Target Attribute values phase
Matched virtual resource, is denoted as target resource;According to the corresponding expert model of the target resource, the target resource is determined
Property;The resource information and property are back to terminal by the resource information for obtaining the target resource;Make described
Terminal is further screened according to multiple target resources of the property to push.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the step of processor realizes any one of claims 1 to 6 the method when executing the computer program.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any one of claims 1 to 6 is realized when being executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910012944.3A CN109829629B (en) | 2019-01-07 | 2019-01-07 | Risk analysis report generation method, apparatus, computer device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910012944.3A CN109829629B (en) | 2019-01-07 | 2019-01-07 | Risk analysis report generation method, apparatus, computer device and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109829629A true CN109829629A (en) | 2019-05-31 |
CN109829629B CN109829629B (en) | 2023-08-22 |
Family
ID=66860768
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910012944.3A Active CN109829629B (en) | 2019-01-07 | 2019-01-07 | Risk analysis report generation method, apparatus, computer device and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109829629B (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110377519A (en) * | 2019-07-22 | 2019-10-25 | 中国工商银行股份有限公司 | Performance capability test method, device, equipment and the storage medium of big data system |
CN110619568A (en) * | 2019-09-17 | 2019-12-27 | 王文斌 | Risk assessment report generation method, device, equipment and storage medium |
CN110659799A (en) * | 2019-08-14 | 2020-01-07 | 深圳壹账通智能科技有限公司 | Attribute information processing method and device based on relational network, computer equipment and storage medium |
CN110766559A (en) * | 2019-10-30 | 2020-02-07 | 海腾保险代理有限公司 | Label configuration method and device |
CN111198945A (en) * | 2019-12-03 | 2020-05-26 | 泰康保险集团股份有限公司 | Data processing method, device, medium and electronic equipment |
CN111736953A (en) * | 2020-06-23 | 2020-10-02 | 深圳市云智融科技有限公司 | Virtual resource delivery method and device, computer equipment and storage medium |
CN111770095A (en) * | 2020-06-29 | 2020-10-13 | 百度在线网络技术(北京)有限公司 | Detection method, device, equipment and storage medium |
CN112434082A (en) * | 2020-11-25 | 2021-03-02 | 平安普惠企业管理有限公司 | Operation and maintenance resource management method, device, equipment and medium |
CN112613569A (en) * | 2020-12-29 | 2021-04-06 | 北京百度网讯科技有限公司 | Image recognition method, and training method and device of image classification model |
CN113283760A (en) * | 2021-05-31 | 2021-08-20 | 浙江环玛信息科技有限公司 | Case flow analysis report generation method and system |
CN115168686A (en) * | 2022-07-25 | 2022-10-11 | 江苏金农股份有限公司 | Public opinion clue processing method based on feature words |
CN115964582A (en) * | 2022-11-03 | 2023-04-14 | 太平洋电信股份有限公司 | Network security risk assessment method and system |
CN116401679A (en) * | 2023-06-08 | 2023-07-07 | 张家港金典软件有限公司 | Security management method and system based on enterprise software vulnerability recognition |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110055852A1 (en) * | 2009-08-31 | 2011-03-03 | Bank Of America Corporation | Event processing for detection of suspicious financial activity |
US20140279641A1 (en) * | 2013-03-15 | 2014-09-18 | Alert Enterprise | Identity and asset risk score intelligence and threat mitigation |
US20160125546A1 (en) * | 2014-11-04 | 2016-05-05 | Hartford Fire Insurance Company | System to administer risk transfer and property salvage recovery |
CN106681910A (en) * | 2016-12-07 | 2017-05-17 | 中国工商银行股份有限公司 | Method and device for generating test cases on basis of mainframe code analysis |
CN106952155A (en) * | 2017-03-08 | 2017-07-14 | 深圳前海纵腾金融科技服务有限公司 | A kind of collection method and device based on credit scoring |
-
2019
- 2019-01-07 CN CN201910012944.3A patent/CN109829629B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110055852A1 (en) * | 2009-08-31 | 2011-03-03 | Bank Of America Corporation | Event processing for detection of suspicious financial activity |
US20140279641A1 (en) * | 2013-03-15 | 2014-09-18 | Alert Enterprise | Identity and asset risk score intelligence and threat mitigation |
US20160125546A1 (en) * | 2014-11-04 | 2016-05-05 | Hartford Fire Insurance Company | System to administer risk transfer and property salvage recovery |
CN106681910A (en) * | 2016-12-07 | 2017-05-17 | 中国工商银行股份有限公司 | Method and device for generating test cases on basis of mainframe code analysis |
CN106952155A (en) * | 2017-03-08 | 2017-07-14 | 深圳前海纵腾金融科技服务有限公司 | A kind of collection method and device based on credit scoring |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110377519B (en) * | 2019-07-22 | 2023-06-23 | 中国工商银行股份有限公司 | Performance capacity test method, device and equipment of big data system and storage medium |
CN110377519A (en) * | 2019-07-22 | 2019-10-25 | 中国工商银行股份有限公司 | Performance capability test method, device, equipment and the storage medium of big data system |
CN110659799A (en) * | 2019-08-14 | 2020-01-07 | 深圳壹账通智能科技有限公司 | Attribute information processing method and device based on relational network, computer equipment and storage medium |
WO2021027317A1 (en) * | 2019-08-14 | 2021-02-18 | 深圳壹账通智能科技有限公司 | Relationship network-based attribute information processing method and device, computer apparatus, and storage medium |
CN110619568A (en) * | 2019-09-17 | 2019-12-27 | 王文斌 | Risk assessment report generation method, device, equipment and storage medium |
CN110766559A (en) * | 2019-10-30 | 2020-02-07 | 海腾保险代理有限公司 | Label configuration method and device |
CN110766559B (en) * | 2019-10-30 | 2023-09-29 | 海腾保险代理有限公司 | Label configuration method and device |
CN111198945A (en) * | 2019-12-03 | 2020-05-26 | 泰康保险集团股份有限公司 | Data processing method, device, medium and electronic equipment |
CN111736953A (en) * | 2020-06-23 | 2020-10-02 | 深圳市云智融科技有限公司 | Virtual resource delivery method and device, computer equipment and storage medium |
CN111770095B (en) * | 2020-06-29 | 2023-04-18 | 百度在线网络技术(北京)有限公司 | Detection method, device, equipment and storage medium |
CN111770095A (en) * | 2020-06-29 | 2020-10-13 | 百度在线网络技术(北京)有限公司 | Detection method, device, equipment and storage medium |
CN112434082A (en) * | 2020-11-25 | 2021-03-02 | 平安普惠企业管理有限公司 | Operation and maintenance resource management method, device, equipment and medium |
CN112613569A (en) * | 2020-12-29 | 2021-04-06 | 北京百度网讯科技有限公司 | Image recognition method, and training method and device of image classification model |
CN112613569B (en) * | 2020-12-29 | 2024-04-09 | 北京百度网讯科技有限公司 | Image recognition method, training method and device for image classification model |
CN113283760A (en) * | 2021-05-31 | 2021-08-20 | 浙江环玛信息科技有限公司 | Case flow analysis report generation method and system |
CN115168686A (en) * | 2022-07-25 | 2022-10-11 | 江苏金农股份有限公司 | Public opinion clue processing method based on feature words |
CN115168686B (en) * | 2022-07-25 | 2023-05-12 | 江苏金农股份有限公司 | Public opinion clue processing method based on feature words |
CN115964582A (en) * | 2022-11-03 | 2023-04-14 | 太平洋电信股份有限公司 | Network security risk assessment method and system |
CN115964582B (en) * | 2022-11-03 | 2023-09-19 | 太平洋电信股份有限公司 | Network security risk assessment method and system |
CN116401679A (en) * | 2023-06-08 | 2023-07-07 | 张家港金典软件有限公司 | Security management method and system based on enterprise software vulnerability recognition |
CN116401679B (en) * | 2023-06-08 | 2023-09-05 | 张家港金典软件有限公司 | Security management method and system based on enterprise software vulnerability recognition |
Also Published As
Publication number | Publication date |
---|---|
CN109829629B (en) | 2023-08-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109829629A (en) | Generation method, device, computer equipment and the storage medium of risk analysis reports | |
CN109829628A (en) | Method for prewarning risk, device and computer equipment based on big data | |
CN109165840B (en) | Risk prediction processing method, risk prediction processing device, computer equipment and medium | |
CN109272396B (en) | Customer risk early warning method, device, computer equipment and medium | |
WO2020135535A1 (en) | Recommendation model training method and related apparatus | |
CN108876600A (en) | Warning information method for pushing, device, computer equipment and medium | |
CN104781837B (en) | System and method for forming predictions using event-based sentiment analysis | |
CN109783730A (en) | Products Show method, apparatus, computer equipment and storage medium | |
CN109543925B (en) | Risk prediction method and device based on machine learning, computer equipment and storage medium | |
CN109858010B (en) | Method and device for recognizing new words in field, computer equipment and storage medium | |
CN110377632B (en) | Litigation result prediction method, litigation result prediction device, litigation result prediction computer device and litigation result prediction storage medium | |
WO2017190610A1 (en) | Target user orientation method and device, and computer storage medium | |
CN111783993A (en) | Intelligent labeling method and device, intelligent platform and storage medium | |
CN110362798B (en) | Method, apparatus, computer device and storage medium for judging information retrieval analysis | |
CN109800354B (en) | Resume modification intention identification method and system based on block chain storage | |
US20220284067A1 (en) | Method for pushing information, electronic device | |
CN110674131A (en) | Financial statement data processing method and device, computer equipment and storage medium | |
CN109801151A (en) | Financial fraud risk monitoring and control method, apparatus, computer equipment and storage medium | |
CN111190946A (en) | Report generation method and device, computer equipment and storage medium | |
CN114399396A (en) | Insurance product recommendation method and device, computer equipment and storage medium | |
CN116415017A (en) | Advertisement sensitive content auditing method and system based on artificial intelligence | |
CN111242710A (en) | Business classification processing method and device, service platform and storage medium | |
CN113706211A (en) | Advertisement click rate prediction method and system based on neural network | |
CN112288279A (en) | Business risk assessment method and device based on natural language processing and linear regression | |
CN110310012B (en) | Data analysis method, device, equipment and computer readable storage medium |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |