CN109523118A - Risk data screening technique, device, computer equipment and storage medium - Google Patents
Risk data screening technique, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109523118A CN109523118A CN201811183701.8A CN201811183701A CN109523118A CN 109523118 A CN109523118 A CN 109523118A CN 201811183701 A CN201811183701 A CN 201811183701A CN 109523118 A CN109523118 A CN 109523118A
- Authority
- CN
- China
- Prior art keywords
- data
- risk
- risk data
- assumptions
- risks
- 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
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
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This application involves big data technical field, it is applied to financial industry, provides a kind of risk data screening technique, device, computer equipment and storage medium.Method includes: the affiliated data category according to the risk data of acquisition, determine the composition risks and assumptions of risk data index, obtain risk data index, multiple risk data indexs are inputted to default risk evaluation model respectively, obtain the risk data indicator combination that risk evaluation result difference is greater than setting range, and by the comparison of the corresponding risk data of risk data indicator combination, determines effective risks and assumptions, risk data is screened.Pass through risk data classification, determine the composition risks and assumptions of risk data index, preliminary screening has been carried out to risk data and has obtained risk data index, according to the risk evaluation result of multiple risk data indexs, determine effective risks and assumptions, the postsearch screening to risk data is realized, and then improves the validity of risk data, avoids invalid data risk of interferences assessment result.
Description
Technical field
This application involves big data technical fields, set more particularly to a kind of risk data screening technique, device, computer
Standby and storage medium.
Background technique
With Enterprise Diversification and the development of international operation, more and more enterprises, for the risk of enterprise itself
Control and early warning are increasingly valued, and since effective Risk-warning advantageously reduces business risk, reduce interests loss.Traditional wind
The characteristics of dangerous method for early warning is according to research object monitors the variation tendency of risk signal by collection related data information, and
The degree of strength that various risk status deviate threshold value of warning is evaluated, is issued warning signal to decision-making level and what is taken some countermeasures in advance is
System.The core technology of these methods is usually Expert Rules or machine learning algorithm.
However, traditional machine learning algorithm is often to screen according to industry experience when carrying out risk data screening
To risk data in there may be partial invalidity data, risk of interferences prediction results.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of risk data sieve that can be improved data validity
Choosing method, device, computer equipment and storage medium.
A kind of risk data screening technique, which comprises
Risk data to be screened is obtained, according to the affiliated data category of the risk data, determines risk data index
Composition risks and assumptions;
According to the risk data and the composition risks and assumptions, the risk data index is obtained;
Multiple risk data indexs are inputted to default risk evaluation model respectively, it is poor to obtain the risk evaluation result
The different risk data indicator combination greater than setting range, and obtain each risk data index pair in the risk data indicator combination
The composition risks and assumptions set answered;
Compare the corresponding risk data of the composition risks and assumptions set, effective risks and assumptions are determined according to comparison result;
According to effective risks and assumptions, the risk data is screened.
It is described in one of the embodiments, to obtain risk data to be screened, according to the affiliated number of the risk data
According to classification, determine that the composition risks and assumptions of risk data index include:
The risk data of positive sample and the risk data of negative sample are obtained, according to the affiliated data class of the risk data
Not, classify to the risk data;
According to preset evaluation parameter, the differentiation of the risk data of all categories for positive sample and negative sample is evaluated
Degree;
According to the discrimination evaluation result, the composition risks and assumptions of the risk data index are determined.
It is described in one of the embodiments, to obtain risk data to be screened, according to the affiliated number of the risk data
According to classification, before the composition risks and assumptions for determining risk data index, further includes:
Pending data is obtained, the normalized of data format is carried out to the pending data;
Data cleansing is carried out to the pending data of the normalized Jing Guo data format, obtains cleaning data;
Derivative calculation processing is carried out to the cleaning data, obtains derivative data;
According to preset threshold range, the cleaning data and the derivative data are screened, determine the risk number
According to.
The acquisition pending data in one of the embodiments, carries out data format to the pending data
Normalized includes:
Obtain the pending data draw in unstructured pending data, to the unstructured pending data into
Row keyword extraction and/or subject distillation;
According to extraction as a result, the unstructured pending data is converted to structural data.
It is described according to preset threshold range in one of the embodiments, to the cleaning data and the derivative data
It is screened, after determining the risk data, further includes:
According to the risk data, the corresponding warning information of the risk data is pushed.
It is described according to effective risks and assumptions in one of the embodiments, the risk data is carried out to screen it
Afterwards, further includes:
According to the valid data classification, effective risk data of enterprise to be analyzed is obtained, and updates the risk data
The composition risks and assumptions of index;
According to the composition risks and assumptions of effective risk data and the risk data index of update, determine described in
The risk data index of each dimension of enterprise to be analyzed;
Obtain the dimension class label that the risk data index of each dimension carries;
The risk data index is inputted default risk corresponding with the dimension class label in preset model group to comment
Estimate model, obtains the corresponding risk evaluation result of the risk data index;
According to the risk evaluation result of each default risk evaluation model, the integrated risk letter of enterprise to be analyzed is determined
Breath.
In one of the embodiments, it is described by the risk data index input preset model group in the dimension class
The corresponding default risk evaluation model of distinguishing label also wraps before obtaining the corresponding risk evaluation result of the risk data index
It includes:
Obtain the positive sample risk data and negative sample risk data of each dimension;
According to the positive sample risk data and negative sample risk data of each dimension, training obtains the default wind of each dimension
Dangerous assessment models;
According to the default risk evaluation model of each dimension, preset model group is constructed.
A kind of risk data screening plant, described device include:
Risks and assumptions determining module is formed, for obtaining risk data to be screened, according to belonging to the risk data
Data category determines the composition risks and assumptions of risk data index;
Risk data index obtains module, is used for risk evaluation module, for according to the risk data and the composition
Risks and assumptions obtain the risk data index;
Risk evaluation module is obtained for multiple risk data indexs to be inputted to default risk evaluation model respectively
The risk evaluation result difference is greater than the risk data indicator combination of setting range, and obtains the risk data indicator combination
In the corresponding composition risks and assumptions set of each risk data index;
Effective risks and assumptions determining module, is used for the corresponding risk data of the composition risks and assumptions set, according to
Comparison result determines effective risks and assumptions;
Risk data screening module, for being screened to the risk data according to effective risks and assumptions.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processing
Device performs the steps of when executing the computer program
Risk data to be screened is obtained, according to the affiliated data category of the risk data, determines risk data index
Composition risks and assumptions;
According to the risk data and the composition risks and assumptions, the risk data index is obtained;
Multiple risk data indexs are inputted to default risk evaluation model respectively, it is poor to obtain the risk evaluation result
The different risk data indicator combination greater than setting range, and obtain each risk data index pair in the risk data indicator combination
The composition risks and assumptions set answered;
Compare the corresponding risk data of the composition risks and assumptions set, effective risks and assumptions are determined according to comparison result;
According to effective risks and assumptions, the risk data is screened.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
It is performed the steps of when row
Risk data to be screened is obtained, according to the affiliated data category of the risk data, determines risk data index
Composition risks and assumptions;
According to the risk data and the composition risks and assumptions, the risk data index is obtained;
Multiple risk data indexs are inputted to default risk evaluation model respectively, it is poor to obtain the risk evaluation result
The different risk data indicator combination greater than setting range, and obtain each risk data index pair in the risk data indicator combination
The composition risks and assumptions set answered;
Compare the corresponding risk data of the composition risks and assumptions set, effective risks and assumptions are determined according to comparison result;
According to effective risks and assumptions, the risk data is screened.
Above-mentioned risk data screening technique, device, computer equipment and storage medium pass through the affiliated data of risk data
Classification obtains risk data index to determine the composition risks and assumptions of risk data index, and multiple risk data indexs are distinguished
Default risk evaluation model is inputted, obtains the risk data indicator combination that risk evaluation result difference is greater than setting range, and obtain
The corresponding composition risks and assumptions set of each risk data index in risk data indicator combination is taken, according to composition risks and assumptions set
The otherness of corresponding risk data determines effective risks and assumptions, to screen to risk data.This programme passes through risk
Data category determines the composition risks and assumptions for inputting the risk data index of default risk evaluation model, to risk data
It has carried out preliminary screening and has obtained risk data index, multiple risk data indexs are inputted into default risk evaluation model, according to commenting
Estimate as a result, determining effective risks and assumptions, realize the postsearch screening to risk data, to improve for assessing risk situation
Risk data validity, avoid invalid data risk of interferences assessment result.
Detailed description of the invention
Fig. 1 is the application scenario diagram of one embodiment risk data screening method;
Fig. 2 is the flow diagram of another embodiment risk data screening method;
Fig. 3 is the flow diagram of another embodiment risk data screening step;
Fig. 4 is the flow diagram of another embodiment risk data screening method;
Fig. 5 is the structural block diagram of one embodiment risk data screening device;
Fig. 6 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.
Risk data screening technique provided by the present application passes through the affiliated number of risk data for being applied to financial industry
Risk data index is obtained to determine the composition risks and assumptions of risk data index according to classification, by multiple risk data indexs point
Risk evaluation model Shu Ru not be preset, the risk data indicator combination that risk evaluation result difference is greater than setting range is obtained, and
The corresponding composition risks and assumptions set of each risk data index in risk data indicator combination is obtained, according to forming risks and assumptions collection
The otherness for closing corresponding risk data determines effective risks and assumptions, screens to risk data.Calculating can specifically be passed through
Machine program realizes that computer program can load in terminal to the risk data screening technique of the application, and terminal can be with
But it is not limited to various personal computers, laptop, smart phone, tablet computer.
In one embodiment, as shown in Figure 1, providing a kind of risk data screening technique, comprising the following steps:
Step S200 obtains risk data to be screened, according to the affiliated data category of risk data, determines risk data
The composition risks and assumptions of index.
After risk data refers to the data progress preliminary screening to each dimension of acquisition, the satisfaction of acquisition presets risk threshold value
It is required that data as unit of sample companies, pass through web crawlers algorithm and existing associated storage data in embodiment
Lookup, obtain the macroeconomic datas of the sample companies, industrial and commercial information, customs's data, financial data, public sentiment data, association
The data of each dimensions such as relationship, law data, real estate data are required according to preset risk threshold value, to the data of each dimension
It is screened, obtains risk data.Wherein, preset risk threshold value requires to be set according to history promise breaking record data
Fixed, when the data of acquisition are more than that preset risk threshold value requires, there are certain default risks, are denoted as risk data.Each
The data information amount of sample companies is huge, first preliminary classification is carried out according to data of the data dimension to acquisition, then by each dimension
Data according to subdivision data category carry out secondary classification and determine class label, it is right according to the class label of secondary classification
Data of all categories are screened, and filtering does not meet the data of default risk threshold value, are evaluated filtered Various types of data,
Determine whether the data category can be used as the composing factor of risk data index.
Step S300 obtains risk data index according to risk data and composition risks and assumptions.
It is corresponding with the data category of risk data to form risks and assumptions, using the risk data of the data category as risk
Data target each group obtains risk data index at the corresponding data of risks and assumptions.Risk data index include vector form or
It is matrix form, composition risks and assumptions indicate the component of vector or matrix, and risk data indicates the specific of each component
Value.
Multiple risk data indexs are inputted default risk evaluation model respectively, obtain risk evaluation result by step S400
Difference is greater than the risk data indicator combination of setting range, and it is corresponding to obtain each risk data index in risk data indicator combination
Composition risks and assumptions set.
Default risk evaluation model risk data index for receiving input and the assessment logic according to setting, to risk
It assesses data and carries out assessment processing, obtain risk evaluation result.Default risk evaluation model can use the positive sample got
Data and negative sample data construct to obtain by XGboost algorithm or RForest algorithm as training sample.It is general next
It says, the default risk evaluation model of each dimension can be constructed respectively according to different dimensions.Different risk data indexs is corresponding not
Same risk evaluation result is compared multiple risk evaluation results of same default risk evaluation model output, obtains ratio
The risk evaluation result difference relatively obtained is greater than the comparison other of setting range, and by its corresponding risk data index group pair,
Form risk data indicator combination.Due to risk indicator combination it is corresponding be same dimension risk data, and its risk is commented
It is larger to estimate result difference, it can be by the way that composition risks and assumptions set be further analyzed, so that screening obtains effective wind
Dangerous data.
Step S500, comparative group close corresponding risk data at risk factor set, determine effective risk according to comparison result
The factor.
Risk data index is made of multiple risks and assumptions, and each risks and assumptions correspond to at-risk data, since risk is commented
Estimate and differs greatly existing for result, therefore corresponding risk data index is there is also difference, and the root of not reaching the same goal of risk data index
It is caused by it forms the corresponding risk data difference of risks and assumptions, by comparing the corresponding wind of composition risks and assumptions set on earth
Dangerous data can determine that specifically there is different risk datas is which, so that it is determined that effective risks and assumptions out, effective risk because
Son refers to easily distinguishable risk data index, leads to the important factor in order of different risk evaluation results.In embodiment, risk
Assessment result can be risk probability, and risk probability is converted according to setting output format, such as by risk probability with very
The scoring form output such as system, hundred-mark system, according to the risk score of each risk data index, the wind low for score height and score
Dangerous data target compares and analyzes, and determines the composition risks and assumptions wherein to differ greatly, for example, by a large amount of comparison statistics,
It was found that score it is high risk data index composition risks and assumptions corresponding with the financial data of risk data index that score is low it is poor
It is different larger, the corresponding risks and assumptions that form of financial data are determined for the influence degree of risk score height with this, group becomes a common practice
Effective selection of the dangerous factor, can be used for improving the performance of risk evaluation model.
Step S600 screens risk data according to effective risks and assumptions.
Risks and assumptions are corresponding with the data category of risk data, by effective risks and assumptions, can obtain risk data
Valid data classification effective risk data can quickly be screened, so that it is determined that risk is commented according to determining valid data classification
The effective input data for estimating model, obtains accurate risk evaluation result.
Above-mentioned risk data screening technique, by the affiliated data category of risk data, to determine risk data index
Risks and assumptions are formed, risk data index is obtained, multiple risk data indexs are inputted to default risk evaluation model respectively, are obtained
Risk evaluation result difference is greater than the risk data indicator combination of setting range, and obtains each risk in risk data indicator combination
The corresponding composition risks and assumptions set of data target, according to the otherness of the corresponding risk data of composition risks and assumptions set, really
Fixed effective risks and assumptions, to be screened to risk data.For this programme by risk data classification, determination is default for inputting
The composition risks and assumptions of the risk data index of risk evaluation model have carried out preliminary screening to risk data and have obtained risk data
Multiple risk data indexs are inputted default risk evaluation model and determine effective risks and assumptions according to assessment result by index, real
Show the postsearch screening to risk data, to improve the validity of the risk data for assessing risk situation, avoids nothing
Imitate data risk of interferences assessment result.
In one embodiment, as shown in Fig. 2, step S200, obtains risk data to be screened, according to risk data
Affiliated data category determines that the composition risks and assumptions of risk data index include:
Step S220 obtains the risk data of positive sample and the risk data of negative sample, according to the affiliated number of risk data
According to classification, classify to risk data.
Step S240 evaluates risk data of all categories for positive sample and negative sample according to preset evaluation parameter
Discrimination.
Step S260 determines the composition risks and assumptions of risk data index according to discrimination evaluation result.
Risk data is the related data of known business, can will be known according to the existing violation of agreement of the known business
Enterprise can be divided into positive sample and negative sample, and corresponding risk data is to positive sample data and negative sample data through being sieved
Choosing handle be more than given threshold data.The affiliated data category of risk data refer to the affiliated dimension of risk data with
And belong to specific data category in the dimension.For example, it is 0.8 that risk data, which is income-debt-to-equity ratio of the enterprise A in January,
The risk data is then divided into financial data dimension, specific data category is business revenue data.
In embodiment, preset evaluation parameter includes PD poor, AR value and IV value.Wherein PD difference is used for according to i-th
Enterprise's sample is divided into two class of positive sample and negative sample by the triggering for forming risks and assumptions, and PD difference is enterprise respectively in positive sample, negative
The difference of accounting in sample, briefly, PD difference are the discriminations to positive sample and negative sample in order to judge this classification standard.
Wherein, giRepresent the quantity of the positive sample of i-th of composition risks and assumptions differentiation;biRepresent i-th of composition risks and assumptions
The quantity of the negative sample of differentiation;GiIn the positive sample for representing the differentiation of i-th of composition risks and assumptions, the quantity of true positive sample;BiGeneration
In the negative sample that i-th of table composition risks and assumptions differentiate, the quantity of true negative sample.
AR value, when AR value is closer to 0, is illustrated for analyzing signal to the index of positive sample and negative sample discrimination,
The composing factor does not have a discrimination to positive sample and negative sample, the bigger explanation composing factor of the absolute value of AR value to positive sample and
Negative sample more has discrimination bigger.
Wherein: G represents the quantity of true positive sample;B represents the quantity of true negative sample.
Whether positive sample and the negative sample distribution that IV value is used to measure a variable under different enumerated values are variant, work as group
Discrimination at risk factor pair positive sample and negative sample is bigger, and corresponding IV value is higher.
, AR value poor by PD and IV value evaluate the composition risks and assumptions of risk data index, are joined according to evaluation
Several evaluation results is realized to the screening of composition risks and assumptions, screens out the composition weak to positive sample and negative sample separating capacity
Risks and assumptions.Specifically, composition risks and assumptions can be ranked up, certain data target has according to the power of separating capacity
100 composition risks and assumptions, the input data format of model are required to be to form the vector that risks and assumptions are 50, can be filtered out
50 strong composition risks and assumptions of separating capacity are realized as mode input data, that is, risk data index composition risks and assumptions
Form the screening of risks and assumptions.
In one embodiment, as shown in figure 3, step S200, obtains risk data to be screened, according to risk data
Affiliated data category, before the composition risks and assumptions for determining risk data index, further includes:
Step S120 obtains pending data, and the normalized of data format is carried out to pending data.
Step S140 carries out data cleansing to the pending data of the normalized Jing Guo data format, is cleaned
Data.
Step S160 carries out derivative calculation processing to cleaning data, obtains derivative data.
Step S180 screens cleaning data and derivative data, determines risk data according to preset threshold range.
In data mining, in the initial data of magnanimity there is it is a large amount of it is imperfect, inconsistent, have abnormal data, sternly
The execution efficiency of data mining is influenced again, in some instances it may even be possible to lead to the deviation of Result.The normalized of data format is
Refer to the structural data for converting the text data of non-structured data such as news etc to form.Data format is returned
One change processing, is conducive to the further data mining of data.Data cleansing refers to that mistake, redundancy and the data eliminated in data are made an uproar
Sound can do a simple statistical analysis after the normalized for carrying out data format first, as average value,
The frequency, variance etc. find abnormal point, noise etc. in data, to judge the quality of data, then carry out exceptional value to data
With missing values processing, duplicate removal processing and noise treatment.The derivative calculation processing of data is to be hidden in wherein to believe by algorithm search
The process of breath, derivative calculate of data includes: according to the corresponding classification of data and data processing need in one of the embodiments,
It asks, sets corresponding calculation formula, such as by taking financial data as an example, can be calculated by the data of acquisition with setting formula:
Sales dollar ratio, net profit, the velocity of liquid assets, turnover rate of fixed assets, total assets profit margin, cost profit
The derivative datas such as rate, operating profit ratio, income debt ratio, equity ratio.These data can not generally directly acquire to obtain, foundation
The acquisition data of each approach, and by deep derivative calculating, available comprehensive depth data is conducive to more intuitively
Carry out risk assessment.The risk data for existing in data and derivative data and influencing risk assessment is cleaned, a large amount of data with existing are passed through
Conformity calculation, the available suitable threshold range assessed whether as risk data, as preset threshold range.
In one embodiment, pending data is obtained, the normalized packet of data format is carried out to pending data
It includes:
The unstructured pending data in pending data stroke is obtained, keyword is carried out to unstructured pending data
Extraction and/or subject distillation.
According to extraction as a result, unstructured pending data is converted to structural data.
Unstructured data includes such as news content of text, by carrying out keyword and/or master to unstructured data
Topic is extracted, and when unstructured data is the text such as news data comprising Data subject, extracts descriptor, in embodiment,
When including multiple descriptor in theme of news, then the corresponding multi-threaded word extraction of progress.Get keyword and/or descriptor.When
When unstructured data is the text of public sentiment not comprising data subject etc, the keyword of text is extracted.Specific extraction side
Method can will be mentioned using TF-IDF keyword extracting method, Topic-model subject distillation method and RAKE keyword extraction etc.
Keyword, descriptor for taking out etc. arrange the structural data to show in tabular form.
In one embodiment, as shown in figure 3, step S180, according to preset threshold range, to cleaning data and generaton number
According to being screened, after determining risk data, further includes:
Step S190 pushes the corresponding warning information of risk data according to risk data.
According to preset threshold range, data are screened, when exceeding preset threshold range, the data is characterized and is in non-
Normal condition, pushes the risk data of the abnormal condition, and sends early warning information to relevant staff's counterpart terminal, so that
Staff can further judge the corresponding company of the data or enterprise, determine that the enterprise is really there is risk also
It is difference caused by the data obtained.Since risk data includes to clean data and pass through deeply to excavate obtained derivative data,
By increasing the function of early warning push, risk data is applied into other scenes, improves the utilization rate of data.It is pushed away by early warning
It send, relevant staff will be pushed to by deeply excavating the obtained corresponding various features of derivative data, be conducive to work people
Member not only increases the utilization rate of data using risk data progress Analysis of Policy Making, and enrich the decision of staff according to
According to improving the efficiency of decision-making.
In one embodiment, as shown in figure 4, step S600 screens risk data according to effective risks and assumptions
Later, further includes:
Step S720 obtains effective risk data of enterprise to be analyzed according to effective risks and assumptions, and updates risk data
The composition risks and assumptions of index.
Step S730, according to the composition risks and assumptions of effective risk data and the risk data index of update, determine to
Analyze the risk data index of each dimension of enterprise.
Step S740 obtains the dimension class label that the risk data index of each dimension carries.
Risk data index is inputted default risk corresponding with dimension class label in preset model group and commented by step S750
Estimate model, obtains the corresponding risk evaluation result of risk data index.
Step S760 determines the synthesis wind of enterprise to be analyzed according to the risk evaluation result of each default risk evaluation model
Dangerous information.
According to effective risks and assumptions, the composition risks and assumptions of risk data index are updated, when needing to carry out risk to enterprise
When analysis, effective risk data of each data dimension of enterprise to be analyzed is obtained by determining valid data classification, will be updated
Composition risks and assumptions matched with effective risk data, obtain the risk data index of each dimension of enterprise to be analyzed.The dimension of data
Degree includes macroeconomic data, industrial and commercial information, customs's data, financial data, public sentiment data, incidence relation, law data, premises
Produce data etc..The risk data index of each dimension all carries dimension class label, for distinguishing its dimension.The wind of each dimension
Dangerous data target has corresponding default risk evaluation model, and the default risk evaluation model of each dimension constitutes preset model group,
Model Group can be analyzed for each dimension data, respectively obtain the risk score of each dimension risk data as a result, to each dimension
The risk evaluation result of degree is weighted processing, obtains the integrated risk information of enterprise to be analyzed.Integrated risk information combines
The data of each dimension of enterprise to be analyzed, with multi-angle analyze the data of enterprise to be analyzed, improve assessment result has
Effect property.
In one embodiment, risk data index is inputted corresponding with dimension class label default in preset model group
Risk evaluation model, before obtaining the corresponding risk evaluation result of risk data index, further includes:
Obtain the positive sample risk data and negative sample risk data of each dimension.
According to the positive sample risk data and negative sample risk data of each dimension, the default risk that training obtains each dimension is commented
Estimate model.
According to the default risk evaluation model of each dimension, preset model group is constructed.
The building process of model is specially to arrange to the quantity and depth of decision-tree model, can pass through XGboost
Algorithm or RForest algorithm are realized.By taking the Model Group that the risk evaluation model of different dimensions constructs as an example, each dimension is obtained
Positive sample risk data and negative sample risk data, training obtains the default risk evaluation model of corresponding dimension, to construct
It obtains with dimension the Model Group that divides.It is appreciated that carrying out machine learning for different classes of data, different moulds is constructed
Type group is also based on business rule building Model Group in other embodiments, can also construct model based on data cases
Group, such as be divided into real estate industry, non-silver industry, banking, service trade, manufacturing industry etc..
It should be understood that although each step in the flow chart of Fig. 1-4 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 1-4
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, as shown in figure 5, providing a kind of risk data screening plant, comprising:
Risks and assumptions determining module 200 is formed, for obtaining risk data to be screened, according to the affiliated number of risk data
According to classification, the composition risks and assumptions of risk data index are determined;
Risk data index obtain module 300, be used for risk evaluation module, for according to risk data and composition risk because
Son obtains risk data index;
Risk evaluation module 400 obtains wind for multiple risk data indexs to be inputted to default risk evaluation model respectively
Dangerous assessment result difference is greater than the risk data indicator combination of setting range, and obtains each risk number in risk data indicator combination
According to the corresponding composition risks and assumptions set of index;
Effective risks and assumptions determining module 500 closes corresponding risk data at risk factor set for comparative group, according to than
Relatively result determines effective risks and assumptions;
Risk data screening module 600, for being screened to risk data according to effective risks and assumptions.
In one embodiment, risks and assumptions determining module 200 is formed, be also used to obtain the risk data of positive sample and is born
The risk data of sample classifies to risk data according to the affiliated data category of risk data, is joined according to preset evaluation
Number, evaluates risk data of all categories and determines risk according to discrimination evaluation result for the discrimination of positive sample and negative sample
The composition risks and assumptions of data target.
In one embodiment, risk data screening plant further includes data processing module, for obtaining number to be processed
According to the normalized of pending data progress data format, to the number to be processed of the normalized Jing Guo data format
According to data cleansing is carried out, cleaning data are obtained, derivative calculation processing is carried out to cleaning data, derivative data is obtained, according to default
Threshold range screens cleaning data and derivative data, determines risk data.
Data processing module in one embodiment is also used to obtain the unstructured number to be processed in pending data stroke
According to, keyword extraction and/or subject distillation are carried out to unstructured pending data, according to extract as a result, by it is unstructured to
Handle data transitions are structural data.
In one embodiment, risk data screening plant further includes warning information pushing module, for according to risk number
According to the corresponding warning information of push risk data.
In one embodiment, risk data screening plant further includes business risk analysis module, for according to significant figure
According to classification, effective risk data of enterprise to be analyzed is obtained, and updates the composition risks and assumptions of risk data index, according to effective
The composition risks and assumptions of risk data and the risk data index of update determine that the risk data of each dimension of enterprise to be analyzed refers to
Mark obtains the dimension class label that the risk data index of each dimension carries, risk data index is inputted in preset model group
Default risk evaluation model corresponding with dimension class label obtains the corresponding risk evaluation result of risk data index, according to
The risk evaluation result of each default risk evaluation model determines the integrated risk information of enterprise to be analyzed.
In one embodiment, risk data screening plant further includes that preset model group constructs module, for obtaining each dimension
The positive sample risk data and negative sample risk data of degree, according to the positive sample risk data of each dimension and negative sample risk number
According to training obtains the default risk evaluation model of each dimension, according to the default risk evaluation model of each dimension, constructs preset model
Group.
Above-mentioned risk data screening plant, by the affiliated data category of risk data, to determine risk data index
Risks and assumptions are formed, risk data index is obtained, multiple risk data indexs are inputted to default risk evaluation model respectively, are obtained
Risk evaluation result difference is greater than the risk data indicator combination of setting range, and obtains each risk in risk data indicator combination
The corresponding composition risks and assumptions set of data target, according to the otherness of the corresponding risk data of composition risks and assumptions set, really
Fixed effective risks and assumptions, to be screened to risk data.For this programme by risk data classification, determination is default for inputting
The composition risks and assumptions of the risk data index of risk evaluation model have carried out preliminary screening to risk data and have obtained risk data
Multiple risk data indexs are inputted default risk evaluation model and determine effective risks and assumptions according to assessment result by index, real
Show the postsearch screening to risk data, to improve the validity of the risk data for assessing risk situation, avoids nothing
Imitate data risk of interferences assessment result.
Specific about risk data screening plant limits the limit that may refer to above for risk data screening technique
Fixed, details are not described herein.Modules in above-mentioned risk data screening plant can fully or partially through software, hardware and its
Combination is to realize.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be with
It is stored in the memory in computer equipment in a software form, in order to which processor calls the above modules of execution corresponding
Operation.
In one embodiment, a kind of computer equipment is provided, which can be terminal, internal structure
Figure can be as shown in Figure 6.The computer equipment includes processor, the memory, network interface, display connected by system bus
Screen and input unit.Wherein, the processor of the computer equipment is for providing calculating and control ability.The computer equipment is deposited
Reservoir includes non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system and computer journey
Sequence.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The network interface of machine equipment is used to communicate with external terminal by network connection.When the computer program is executed by processor with
Realize a kind of risk data screening technique.The display screen of the computer equipment can be liquid crystal display or electric ink is shown
Screen, the input unit of the computer equipment can be the touch layer covered on display screen, be also possible on computer equipment shell
Key, trace ball or the Trackpad of setting can also be external keyboard, Trackpad or mouse etc..
It will be understood by those skilled in the art that structure shown in Fig. 6, 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.
In one embodiment, a kind of computer equipment, including memory and processor are provided, which is stored with
Computer program, the processor perform the steps of when executing computer program
Risk data to be screened is obtained, according to the affiliated data category of risk data, determines the group of risk data index
At risks and assumptions;
According to risk data and composition risks and assumptions, risk data index is obtained;
Multiple risk data indexs are inputted to default risk evaluation model respectively, acquisition risk evaluation result difference, which is greater than, to be set
Determine the risk data indicator combination of range, and obtains the corresponding composition risk of each risk data index in risk data indicator combination
Factor set;
Comparative group closes corresponding risk data at risk factor set, determines effective risks and assumptions according to comparison result;
According to effective risks and assumptions, risk data is screened.
In one embodiment, it is also performed the steps of when processor executes computer program
The risk data of positive sample and the risk data of negative sample are obtained, it is right according to the affiliated data category of risk data
Risk data is classified;
According to preset evaluation parameter, risk data of all categories is evaluated for the discrimination of positive sample and negative sample;
According to discrimination evaluation result, the composition risks and assumptions of risk data index are determined.
In one embodiment, it is also performed the steps of when processor executes computer program
Pending data is obtained, the normalized of data format is carried out to pending data;
Data cleansing is carried out to the pending data of the normalized Jing Guo data format, obtains cleaning data;
Derivative calculation processing is carried out to cleaning data, obtains derivative data;
According to preset threshold range, cleaning data and derivative data are screened, determine risk data.
In one embodiment, it is also performed the steps of when processor executes computer program
The unstructured pending data in pending data stroke is obtained, keyword is carried out to unstructured pending data
Extraction and/or subject distillation;
According to extraction as a result, unstructured pending data is converted to structural data.
In one embodiment, it is also performed the steps of when processor executes computer program
According to risk data, the corresponding warning information of risk data is pushed.
In one embodiment, it is also performed the steps of when processor executes computer program
According to valid data classification, effective risk data of enterprise to be analyzed is obtained, and updates the group of risk data index
At risks and assumptions;
According to the composition risks and assumptions of effective risk data and the risk data index of update, determine that enterprise to be analyzed is each
The risk data index of dimension;
Obtain the dimension class label that the risk data index of each dimension carries;
Risk data index is inputted into default risk evaluation model corresponding with dimension class label in preset model group, is obtained
Obtain the corresponding risk evaluation result of risk data index;
According to the risk evaluation result of each default risk evaluation model, the integrated risk information of enterprise to be analyzed is determined.
In one embodiment, it is also performed the steps of when processor executes computer program
Obtain the positive sample risk data and negative sample risk data of each dimension;
According to the positive sample risk data and negative sample risk data of each dimension, the default risk that training obtains each dimension is commented
Estimate model;
According to the default risk evaluation model of each dimension, preset model group is constructed.
The above-mentioned computer equipment for realizing risk data screening technique, by the affiliated data category of risk data,
It determines the composition risks and assumptions of risk data index, obtains risk data index, multiple risk data indexs are inputted respectively
Default risk evaluation model obtains the risk data indicator combination that risk evaluation result difference is greater than setting range, and obtains wind
The corresponding composition risks and assumptions set of each risk data index in the combination of dangerous data target, according to composition risks and assumptions set correspondence
Risk data otherness, effective risks and assumptions are determined, to screen to risk data.This programme passes through risk data
Classification determines the composition risks and assumptions for inputting the risk data index of default risk evaluation model, carries out to risk data
Preliminary screening obtains risk data index, and multiple risk data indexs are inputted default risk evaluation model, are tied according to assessment
Fruit determines effective risks and assumptions, realizes the postsearch screening to risk data, to improve the wind for assessing risk situation
The validity of dangerous data avoids invalid data risk of interferences assessment result.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
Risk data to be screened is obtained, according to the affiliated data category of risk data, determines the group of risk data index
At risks and assumptions;
According to risk data and composition risks and assumptions, risk data index is obtained;
Multiple risk data indexs are inputted to default risk evaluation model respectively, acquisition risk evaluation result difference, which is greater than, to be set
Determine the risk data indicator combination of range, and obtains the corresponding composition risk of each risk data index in risk data indicator combination
Factor set;
Comparative group closes corresponding risk data at risk factor set, determines effective risks and assumptions according to comparison result;
According to effective risks and assumptions, risk data is screened.
In one embodiment, it is also performed the steps of when computer program is executed by processor
The risk data of positive sample and the risk data of negative sample are obtained, it is right according to the affiliated data category of risk data
Risk data is classified;
According to preset evaluation parameter, risk data of all categories is evaluated for the discrimination of positive sample and negative sample;
According to discrimination evaluation result, the composition risks and assumptions of risk data index are determined.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Pending data is obtained, the normalized of data format is carried out to pending data;
Data cleansing is carried out to the pending data of the normalized Jing Guo data format, obtains cleaning data;
Derivative calculation processing is carried out to cleaning data, obtains derivative data;
According to preset threshold range, cleaning data and derivative data are screened, determine risk data.
In one embodiment, it is also performed the steps of when computer program is executed by processor
The unstructured pending data in pending data stroke is obtained, keyword is carried out to unstructured pending data
Extraction and/or subject distillation;
According to extraction as a result, unstructured pending data is converted to structural data.
In one embodiment, it is also performed the steps of when computer program is executed by processor
According to risk data, the corresponding warning information of risk data is pushed.
In one embodiment, it is also performed the steps of when computer program is executed by processor
According to valid data classification, effective risk data of enterprise to be analyzed is obtained, and updates the group of risk data index
At risks and assumptions;
According to the composition risks and assumptions of effective risk data and the risk data index of update, determine that enterprise to be analyzed is each
The risk data index of dimension;
Obtain the dimension class label that the risk data index of each dimension carries;
Risk data index is inputted into default risk evaluation model corresponding with dimension class label in preset model group, is obtained
Obtain the corresponding risk evaluation result of risk data index;
According to the risk evaluation result of each default risk evaluation model, the integrated risk information of enterprise to be analyzed is determined.
In one embodiment, it is also performed the steps of when computer program is executed by processor
Obtain the positive sample risk data and negative sample risk data of each dimension;
According to the positive sample risk data and negative sample risk data of each dimension, the default risk that training obtains each dimension is commented
Estimate model;
According to the default risk evaluation model of each dimension, preset model group is constructed.
The above-mentioned computer readable storage medium for realizing risk data screening technique passes through the affiliated number of risk data
Risk data index is obtained to determine the composition risks and assumptions of risk data index according to classification, by multiple risk data indexs point
Risk evaluation model Shu Ru not be preset, the risk data indicator combination that risk evaluation result difference is greater than setting range is obtained, and
The corresponding composition risks and assumptions set of each risk data index in risk data indicator combination is obtained, according to forming risks and assumptions collection
The otherness for closing corresponding risk data determines effective risks and assumptions, to screen to risk data.This programme passes through wind
Dangerous data category determines the composition risks and assumptions for inputting the risk data index of default risk evaluation model, to risk number
Risk data index is obtained according to preliminary screening has been carried out, multiple risk data indexs are inputted into default risk evaluation model, according to
Assessment result determines effective risks and assumptions, realizes the postsearch screening to risk data, to improve for assessing risk feelings
The validity of the risk data of condition avoids invalid data risk of interferences assessment result.
Those of ordinary skill in the art will appreciate that realize above-described embodiment risk data screening method in whole or
Part process is relevant hardware can be instructed to complete by computer program, computer program to can be stored in one non-
In volatile computer read/write memory medium, the computer program is when being executed, it may include such as the embodiment of above-mentioned each method
Process.Wherein, to memory, storage, database or other media used in each embodiment provided herein
Any reference 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.It is volatile
Property memory may include random access memory (RAM) or external cache.By way of illustration and not limitation, RAM
It is available in many forms, such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram
(DDRSDRAM), enhanced 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 risk data screening technique, which comprises
Risk data to be screened is obtained, according to the affiliated data category of the risk data, determines the group of risk data index
At risks and assumptions;
According to the risk data and the composition risks and assumptions, the risk data index is obtained;
Multiple risk data indexs are inputted to default risk evaluation model respectively, it is big to obtain the risk evaluation result difference
In the risk data indicator combination of setting range, and it is corresponding to obtain each risk data index in the risk data indicator combination
Form risks and assumptions set;
Compare the corresponding risk data of the composition risks and assumptions set, effective risks and assumptions are determined according to comparison result;
According to effective risks and assumptions, the risk data is screened.
2. the method according to claim 1, wherein described obtain risk data to be screened, according to the wind
The affiliated data category of dangerous data determines that the composition risks and assumptions of risk data index include:
The risk data of positive sample and the risk data of negative sample are obtained, it is right according to the affiliated data category of the risk data
The risk data is classified;
According to preset evaluation parameter, the risk data of all categories is evaluated for the discrimination of positive sample and negative sample;
According to the discrimination evaluation result, the composition risks and assumptions of the risk data index are determined.
3. the method according to claim 1, wherein described obtain risk data to be screened, according to the wind
The affiliated data category of dangerous data, before the composition risks and assumptions for determining risk data index, further includes:
Pending data is obtained, the normalized of data format is carried out to the pending data;
Data cleansing is carried out to the pending data of the normalized Jing Guo data format, obtains cleaning data;
Derivative calculation processing is carried out to the cleaning data, obtains derivative data;
According to preset threshold range, the cleaning data and the derivative data are screened, determine the risk data.
4. according to the method described in claim 3, it is characterized in that, the acquisition pending data, to the pending data
Carry out data format normalized include:
The unstructured pending data in the pending data stroke is obtained, the unstructured pending data is closed
Key word extracts and/or subject distillation;
According to extraction as a result, the unstructured pending data is converted to structural data.
5. according to the method described in claim 3, it is characterized in that, described according to preset threshold range, to the cleaning data
It is screened with the derivative data, after determining the risk data, further includes:
According to the risk data, the corresponding warning information of the risk data is pushed.
6. the method according to claim 1, wherein described according to effective risks and assumptions, to the risk
After data are screened, further includes:
According to effective risks and assumptions, effective risk data of enterprise to be analyzed is obtained, and updates the risk data index
Composition risks and assumptions;
According to the composition risks and assumptions of effective risk data and the risk data index of update, determine described wait divide
Analyse the risk data index of each dimension of enterprise;
Obtain the dimension class label that the risk data index of each dimension carries;
The risk data index is inputted into default risk assessment mould corresponding with the dimension class label in preset model group
Type obtains the corresponding risk evaluation result of the risk data index;
According to the risk evaluation result of each default risk evaluation model, the integrated risk information of enterprise to be analyzed is determined.
7. according to the method described in claim 6, it is characterized in that, described input preset model group for the risk data index
In default risk evaluation model corresponding with the dimension class label, obtain the corresponding risk assessment of the risk data index
As a result before, further includes:
Obtain the positive sample risk data and negative sample risk data of each dimension;
According to the positive sample risk data and negative sample risk data of each dimension, the default risk that training obtains each dimension is commented
Estimate model;
According to the default risk evaluation model of each dimension, preset model group is constructed.
8. a kind of risk data screening plant, which is characterized in that described device includes:
Risks and assumptions determining module is formed, for obtaining risk data to be screened, according to the affiliated data of the risk data
Classification determines the composition risks and assumptions of risk data index;
Risk data index obtains module, is used for risk evaluation module, for according to the risk data and the composition risk
The factor obtains the risk data index;
Risk evaluation module presets risk evaluation model for inputting multiple risk data indexs respectively, described in acquisition
Risk evaluation result difference is greater than the risk data indicator combination of setting range, and obtains each in the risk data indicator combination
The corresponding composition risks and assumptions set of risk data index;
Effective risks and assumptions determining module, is used for the corresponding risk data of the composition risks and assumptions set, according to comparing
As a result effective risks and assumptions are determined;
Risk data screening module, for being screened to the risk data according to effective risks and assumptions.
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 7 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 7 is realized when being executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811183701.8A CN109523118A (en) | 2018-10-11 | 2018-10-11 | Risk data screening technique, device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811183701.8A CN109523118A (en) | 2018-10-11 | 2018-10-11 | Risk data screening technique, device, computer equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109523118A true CN109523118A (en) | 2019-03-26 |
Family
ID=65772265
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811183701.8A Pending CN109523118A (en) | 2018-10-11 | 2018-10-11 | Risk data screening technique, device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109523118A (en) |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110188980A (en) * | 2019-04-15 | 2019-08-30 | 深圳壹账通智能科技有限公司 | Business risk methods of marking, device, computer equipment and storage medium |
CN110210559A (en) * | 2019-05-31 | 2019-09-06 | 北京小米移动软件有限公司 | Object screening technique and device, storage medium |
CN110276519A (en) * | 2019-05-14 | 2019-09-24 | 平安科技(深圳)有限公司 | Risk portrait method for building up, device, computer equipment and storage medium |
CN110490720A (en) * | 2019-07-08 | 2019-11-22 | 深圳壹账通智能科技有限公司 | Financial data analysis and early warning method, apparatus, computer equipment and storage medium |
CN110619568A (en) * | 2019-09-17 | 2019-12-27 | 王文斌 | Risk assessment report generation method, device, equipment and storage medium |
CN110838366A (en) * | 2019-10-15 | 2020-02-25 | 平安科技(深圳)有限公司 | Method and device for predicting disease risk |
CN111310784A (en) * | 2020-01-14 | 2020-06-19 | 支付宝(杭州)信息技术有限公司 | Resource data processing method and device |
CN111461892A (en) * | 2020-03-31 | 2020-07-28 | 支付宝(杭州)信息技术有限公司 | Method and device for selecting derived variables of risk identification model |
CN111861734A (en) * | 2020-07-31 | 2020-10-30 | 重庆富民银行股份有限公司 | Test evaluation system and method for three-party data source |
CN111882428A (en) * | 2020-07-27 | 2020-11-03 | 上海应用技术大学 | Commercial bank liquidity risk assessment method and device |
CN111967999A (en) * | 2020-08-27 | 2020-11-20 | 平安养老保险股份有限公司 | Investigation processing method, device, computer equipment and storage medium |
CN112465648A (en) * | 2020-10-21 | 2021-03-09 | 湖南天设信息科技有限公司 | Risk data evaluation method and device, computer equipment and storage medium |
CN112926833A (en) * | 2021-01-28 | 2021-06-08 | 北京安九信息技术有限公司 | Detection method and detection system for health condition of enterprise |
WO2022052546A1 (en) * | 2020-09-11 | 2022-03-17 | 长鑫存储技术有限公司 | Public opinion data processing system and method, computer storage medium, and electronic device |
CN114596011A (en) * | 2022-05-09 | 2022-06-07 | 未来地图(深圳)智能科技有限公司 | Enterprise data processing method based on artificial intelligence and related device |
CN114997750A (en) * | 2022-08-03 | 2022-09-02 | 广东知得失网络科技有限公司 | Risk information pushing method, system, equipment and medium |
CN116841650A (en) * | 2023-08-31 | 2023-10-03 | 腾讯科技(深圳)有限公司 | Sample construction method, device, equipment and storage medium |
CN117853081A (en) * | 2024-02-06 | 2024-04-09 | 一智科技(成都)有限公司 | Worker's early warning system |
US11989743B2 (en) | 2020-09-11 | 2024-05-21 | Changxin Memory Technologies, Inc. | System and method for processing public sentiment, computer storage medium and electronic device |
-
2018
- 2018-10-11 CN CN201811183701.8A patent/CN109523118A/en active Pending
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110188980A (en) * | 2019-04-15 | 2019-08-30 | 深圳壹账通智能科技有限公司 | Business risk methods of marking, device, computer equipment and storage medium |
CN110276519A (en) * | 2019-05-14 | 2019-09-24 | 平安科技(深圳)有限公司 | Risk portrait method for building up, device, computer equipment and storage medium |
CN110210559B (en) * | 2019-05-31 | 2021-10-08 | 北京小米移动软件有限公司 | Object screening method and device and storage medium |
CN110210559A (en) * | 2019-05-31 | 2019-09-06 | 北京小米移动软件有限公司 | Object screening technique and device, storage medium |
CN110490720A (en) * | 2019-07-08 | 2019-11-22 | 深圳壹账通智能科技有限公司 | Financial data analysis and early warning method, apparatus, computer equipment and storage medium |
CN110619568A (en) * | 2019-09-17 | 2019-12-27 | 王文斌 | Risk assessment report generation method, device, equipment and storage medium |
CN110838366A (en) * | 2019-10-15 | 2020-02-25 | 平安科技(深圳)有限公司 | Method and device for predicting disease risk |
CN111310784A (en) * | 2020-01-14 | 2020-06-19 | 支付宝(杭州)信息技术有限公司 | Resource data processing method and device |
CN111461892A (en) * | 2020-03-31 | 2020-07-28 | 支付宝(杭州)信息技术有限公司 | Method and device for selecting derived variables of risk identification model |
CN111882428A (en) * | 2020-07-27 | 2020-11-03 | 上海应用技术大学 | Commercial bank liquidity risk assessment method and device |
CN111861734A (en) * | 2020-07-31 | 2020-10-30 | 重庆富民银行股份有限公司 | Test evaluation system and method for three-party data source |
CN111861734B (en) * | 2020-07-31 | 2024-05-03 | 重庆富民银行股份有限公司 | Test evaluation system and method for three-party data source |
CN111967999A (en) * | 2020-08-27 | 2020-11-20 | 平安养老保险股份有限公司 | Investigation processing method, device, computer equipment and storage medium |
WO2022052546A1 (en) * | 2020-09-11 | 2022-03-17 | 长鑫存储技术有限公司 | Public opinion data processing system and method, computer storage medium, and electronic device |
US11989743B2 (en) | 2020-09-11 | 2024-05-21 | Changxin Memory Technologies, Inc. | System and method for processing public sentiment, computer storage medium and electronic device |
CN112465648A (en) * | 2020-10-21 | 2021-03-09 | 湖南天设信息科技有限公司 | Risk data evaluation method and device, computer equipment and storage medium |
CN112926833A (en) * | 2021-01-28 | 2021-06-08 | 北京安九信息技术有限公司 | Detection method and detection system for health condition of enterprise |
CN114596011A (en) * | 2022-05-09 | 2022-06-07 | 未来地图(深圳)智能科技有限公司 | Enterprise data processing method based on artificial intelligence and related device |
CN114997750A (en) * | 2022-08-03 | 2022-09-02 | 广东知得失网络科技有限公司 | Risk information pushing method, system, equipment and medium |
CN114997750B (en) * | 2022-08-03 | 2022-10-25 | 广东知得失网络科技有限公司 | Risk information pushing method, system, equipment and medium |
CN116841650A (en) * | 2023-08-31 | 2023-10-03 | 腾讯科技(深圳)有限公司 | Sample construction method, device, equipment and storage medium |
CN116841650B (en) * | 2023-08-31 | 2023-11-21 | 腾讯科技(深圳)有限公司 | Sample construction method, device, equipment and storage medium |
CN117853081A (en) * | 2024-02-06 | 2024-04-09 | 一智科技(成都)有限公司 | Worker's early warning system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109523118A (en) | Risk data screening technique, device, computer equipment and storage medium | |
US10943186B2 (en) | Machine learning model training method and device, and electronic device | |
CN108648074B (en) | Loan assessment method, device and equipment based on support vector machine | |
Wehkamp et al. | Governance and deforestation—a meta-analysis in economics | |
WO2019218699A1 (en) | Fraud transaction determining method and apparatus, computer device, and storage medium | |
CN111461216B (en) | Case risk identification method based on machine learning | |
Keramati et al. | A proposed classification of data mining techniques in credit scoring | |
CN110490720A (en) | Financial data analysis and early warning method, apparatus, computer equipment and storage medium | |
CN109360105A (en) | Product risks method for early warning, device, computer equipment and storage medium | |
CN110866819A (en) | Automatic credit scoring card generation method based on meta-learning | |
Wang et al. | Multi-classification assessment of bank personal credit risk based on multi-source information fusion | |
Schreyer et al. | Adversarial learning of deepfakes in accounting | |
CN112488716B (en) | Abnormal event detection system | |
CN108492001A (en) | A method of being used for guaranteed loan network risk management | |
Soviany | The benefits of using artificial intelligence in payment fraud detection: A case study | |
CN114298176A (en) | Method, device, medium and electronic equipment for detecting fraudulent user | |
CN117455417B (en) | Automatic iterative optimization method and system for intelligent wind control approval strategy | |
CN112801775A (en) | Client credit evaluation method and device | |
CN114841801A (en) | Credit wind control method and device based on user behavior characteristics | |
CN115908022A (en) | Abnormal transaction risk early warning method and system based on network modeling | |
Zheng et al. | [Retracted] Using an Optimized Learning Vector Quantization‐(LVQ‐) Based Neural Network in Accounting Fraud Recognition | |
CN117556225B (en) | Pedestrian credit data risk management system | |
Torres et al. | A proposal for online analysis and identification of fraudulent financial transactions | |
Hsin et al. | Feature engineering and resampling strategies for fund transfer fraud with limited transaction data and a time-inhomogeneous modi operandi | |
CN111784360A (en) | Anti-fraud prediction method and system based on network link backtracking |
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 |