CN113052422A - Wind control model training method and user credit evaluation method - Google Patents
Wind control model training method and user credit evaluation method Download PDFInfo
- Publication number
- CN113052422A CN113052422A CN201911383610.3A CN201911383610A CN113052422A CN 113052422 A CN113052422 A CN 113052422A CN 201911383610 A CN201911383610 A CN 201911383610A CN 113052422 A CN113052422 A CN 113052422A
- Authority
- CN
- China
- Prior art keywords
- user
- sample data
- data
- original sample
- credit
- 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
- 238000000034 method Methods 0.000 title claims abstract description 74
- 238000012549 training Methods 0.000 title claims abstract description 44
- 238000011156 evaluation Methods 0.000 title abstract description 27
- 230000006399 behavior Effects 0.000 claims abstract description 94
- 238000012271 agricultural production Methods 0.000 claims abstract description 19
- 238000004140 cleaning Methods 0.000 claims description 33
- 238000004422 calculation algorithm Methods 0.000 claims description 27
- 238000000605 extraction Methods 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 10
- 238000007637 random forest analysis Methods 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 description 18
- 238000004458 analytical method Methods 0.000 description 16
- 238000004891 communication Methods 0.000 description 16
- 230000008569 process Effects 0.000 description 12
- 238000012545 processing Methods 0.000 description 11
- 238000010276 construction Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000013178 mathematical model Methods 0.000 description 7
- 238000009360 aquaculture Methods 0.000 description 6
- 244000144974 aquaculture Species 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000007726 management method Methods 0.000 description 5
- 230000002093 peripheral effect Effects 0.000 description 5
- 230000009471 action Effects 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 238000012954 risk control Methods 0.000 description 4
- 238000012216 screening Methods 0.000 description 4
- 238000007619 statistical method Methods 0.000 description 4
- 230000003542 behavioural effect Effects 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000013210 evaluation model Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 238000005086 pumping Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 238000012550 audit Methods 0.000 description 1
- 238000013475 authorization Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000009395 breeding Methods 0.000 description 1
- 230000001488 breeding effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007635 classification algorithm Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000004836 empirical method Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000003621 irrigation water Substances 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000001915 proofreading effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000013077 scoring method Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/258—Data format conversion from or to a database
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0609—Buyer or seller confidence or verification
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Accounting & Taxation (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Mining & Mineral Resources (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Marine Sciences & Fisheries (AREA)
- Animal Husbandry (AREA)
- Agronomy & Crop Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a wind control model training method and a user credit evaluation method. The method comprises the following steps: determining target sample data according to the acquired original sample data; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user; and training to obtain a wind control model based on target sample data, so that the credit evaluation accuracy of the user is improved.
Description
Technical Field
The invention belongs to the technical field of big data, and particularly relates to a wind control model training method and a user credit evaluation method.
Background
Risk control has wide application value in most of the interconnected and financial companies in the modern society.
In the traditional risk control, related service personnel and safety experts set wind control rules according to conditions such as historical experience and service, and credit scores of users are determined based on data such as telecommunication calls and short message services. With the rapid development of communication technology and the attack of internet technology, the number of telecommunication calls and short message services is gradually reduced, so that the original data analysis dimension cannot completely depict the user behavior.
Disclosure of Invention
The embodiment of the invention provides a wind control model training method and a user credit evaluation method, which can solve the problem of low credit accuracy of a user evaluated based on the current risk evaluation mode.
In a first aspect, a method for training a wind control model is provided, where the method includes:
determining target sample data according to the acquired original sample data; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user;
and training to obtain a wind control model based on the target sample data.
In a possible implementation manner, determining target sample data according to the acquired original sample data includes:
carrying out data cleaning on original sample data; the data cleansing includes at least one of: converting original sample data containing null value fields into first preset fields; converting the field format of original sample data into a preset field format; converting first data in original sample data into a second preset field;
and performing feature extraction on the original sample data subjected to data cleaning to obtain target sample data.
In a possible implementation manner, performing feature extraction on original sample data after data cleaning to obtain target sample data includes:
extracting original sample data with time characteristics and position characteristics in the original sample data after data cleaning;
determining a first weight of each original sample in original sample data with time characteristics and position characteristics according to a random forest algorithm;
determining a second weight of each original sample in the original sample data with the time characteristic and the position characteristic according to an elastic network algorithm;
and taking the original sample data with the first weight larger than a first preset value and the second weight larger than a second preset value as target sample data.
In one possible implementation, the method further includes:
inputting target sample data into a wind control model, and determining a credit score of a user;
determining a predicted credit granting result of the user according to the credit score;
and adjusting parameters of the wind control model according to the predicted credit granting result and the real credit granting result of the user to obtain the optimized wind control model.
In a second aspect, a method for evaluating user credit is provided, the method comprising:
acquiring user behavior information;
inputting user behavior information into the first aspect or a wind control model trained by any possible implementation manner of the first aspect, and determining a user credit score;
and determining the credit granting result of the user according to the credit score of the user.
In one possible implementation, the method further includes:
and storing the credit granting result of the user, and constructing a user credit granting database.
In a third aspect, a training device for a wind control model is provided, which includes:
the determining module is used for determining target sample data according to the acquired original sample data; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user;
and the training module is used for training to obtain the wind control model based on the target sample data.
In a possible implementation manner, the determining module is specifically configured to:
carrying out data cleaning on original sample data; the data cleansing includes at least one of: converting original sample data containing null value fields into first preset fields; converting the field format of original sample data into a preset field format; converting first data in original sample data into a second preset field;
and performing feature extraction on the original sample data subjected to data cleaning to obtain target sample data.
In a possible implementation manner, the determining module is specifically configured to:
extracting original sample data with time characteristics and position characteristics in the original sample data after data cleaning;
determining a first weight of each original sample in original sample data with time characteristics and position characteristics according to a random forest algorithm;
determining a second weight of each original sample in the original sample data with the time characteristic and the position characteristic according to an elastic network algorithm;
and taking the original sample data with the first weight larger than a first preset value and the second weight larger than a second preset value as target sample data.
In one possible implementation, the apparatus further includes:
the determining module is used for inputting the target sample data into the wind control model and determining the credit score of the user;
the determining module is also used for determining a predicted credit granting result of the user according to the credit score;
and the adjusting module is used for adjusting the parameters of the wind control model according to the predicted credit granting result and the real credit granting result of the user to obtain the optimized wind control model.
In a fourth aspect, a user credit scoring apparatus, the apparatus comprising:
acquiring user behavior information;
inputting user behavior information into a wind control model according to any one of claims 1-4, determining a user credit score;
and determining the credit granting result of the user according to the credit score of the user.
In one possible implementation, the apparatus further includes a building module;
and the construction module is used for storing the credit granting result of the user and constructing a user credit granting database.
In a fifth aspect, an electronic device is provided, the device comprising: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements a method as in the first aspect or any of the possible implementations of the first aspect, or implements a method as in the second aspect or any of the possible implementations of the second aspect.
A sixth aspect provides a computer storage medium having stored thereon computer program instructions which, when executed by a processor, implement a method as in the first aspect or any of the possible implementations of the first aspect, or implement a method as in the second aspect or any of the possible implementations of the second aspect.
Based on the provided wind control model training method and the user credit evaluation method, target sample data is determined according to the obtained original sample data; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user; and training to obtain a wind control model based on target sample data, so that the credit evaluation accuracy of the user is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a credit scoring method according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of a method for training a wind control model according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating a credit evaluation method according to an embodiment of the invention;
FIG. 4 is a schematic structural diagram of a training apparatus for a wind-controlled model according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a credit evaluation device according to another embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
Features and exemplary embodiments of various aspects of the present invention will be described in detail below, and in order to make objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not to be construed as limiting the invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the present invention.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
At present, most internet companies with risk control in today's society have wide application value. In most cases, most of traditional wind control is established by related business personnel and safety experts according to previous experience, business and other conditions, the rule establishment period is long, and the rule establishment period cannot be completely suitable for specific industries. Therefore, related industries are continuously innovated by means of internet technology, and new changes appear in aspects of product research and development, user experience, business service and the like, so that internet finance is derived.
The credit scoring model, which may also be referred to as a credit scoring card, is a model that systematically analyzes features of a target customer and a related number field of existing customers by using a data mining technique and a statistical analysis method to discover customers who meet their own market goals and predict their future credit performance. Generally, a traditional grading card model is constructed based on data of own dimensionality of a bank, such as account information, deposit and withdrawal information, consumption behavior and the like of a user at a bank end.
As shown in fig. 1, the credit analysis model based on operator data implements personal comprehensive credit scoring by the method shown in fig. 1, and first performs data collection, then performs data cleaning on the collected data, and performs credit modeling based on the data after data cleaning to obtain a credit evaluation model. Therefore, the credit evaluation model is adopted, and the scoring result is used as a risk control factor of the financial institution. The credit analysis model realizes the user comprehensive credit scoring based on data of communication, short messages, flow and the like of users in operators. After the user comprehensive credit score is obtained, credit service can be provided based on the user comprehensive credit score.
Before modeling credit assessment, acquiring data of model construction, and executing S101: and (6) data acquisition.
The data acquisition comprises the acquisition of user consumption data, user short message service data and user call data. For example, carrier data may be categorized into "consumption, social, behavioral preference" and other types according to usage. For example, package data, traffic data, etc. reflecting consumption conditions, voice dialing data, short message data, etc. reflecting social contact, location data, etc. reflecting behavior preferences. By accessing the own user short message service, conversation service, consumption data and the like of an operator; for the acquired data: field analysis, such as defining field value meaning, statistical units, etc.; the modeling window of validation data, such as validation data, may be from months to years in length. And performing data proofreading and data cleaning integration, such as integrating data fields in different fields together to form a modeling data set and the like.
After data acquisition, certain preprocessing needs to be performed on the data, and S102 is executed: and (6) data cleaning.
And the data cleaning comprises data standardization, data cleaning, noise removal and data feature extraction. For example, preprocessing of data includes processing of missing and extreme values, and data deduplication. The data set used for modeling will inevitably contain missing values, such as filling the missing values in the data set with a specific value uniformly, or filling a plurality of values based on different data scenarios. Outliers in the data set are usually outside of reasonable limits, such as those that appear to be unfairly-20 years old or 300 years old. As above, outliers may be treated differently based on different scenarios, such as being deleted directly, or not treated, or replaced with other values.
In addition, special data processing such as feature extraction and standardization of data is completed. And selecting the characteristic fields with high contribution degree to the model effect in the data set by a specific method, processing and extracting. If the original data set contains 500 features, through feature extraction, 50 features are finally screened out to be used for establishing the model. Feature normalization, which may also be referred to as feature normalization, adjusts the value of each dimension of the dataset to within a certain range, such as uniformly adjusting the value to between 0-1 by data normalization for each of the 50 features screened in the above example, which helps to improve the accuracy of the model during modeling.
Based on the data obtained in S102, model construction is performed, and S103: and (4) modeling credit.
The model is constructed using conventional classification methods, such as linear regression. And (3) designing a complete user credit rating rule by combining a credit rating card thought and a mathematical model and combining expert experience through a mature algorithm in a classification algorithm, such as a linear regression algorithm, so as to finish the user credit rating.
Based on the credit assessment model obtained in S103, a user credit score can be determined. Providing credit service based on the user credit score, executing S104: and providing the credit service.
And based on the result output by the credit modeling link, a data service interface is packaged, diversified and convenient credit service is provided for the outside, and the specific requirements of the wind control links of a plurality of demanders on the data are met.
By performing user credit analysis on the existing credit analysis model, the following problems still exist:
the model cannot perform credit analysis on users in specific industries; traditional operator data personal credit analysis tends to be personal comprehensive credit rating, lacking close connection with application scenarios. For example, in the face of a more subdivided personal aquaculture fund credit user scenario, the existing credit model and the traditional analysis mode cannot perform credit analysis on the applicable scenario of the aquaculture user, for example, a large aquaculture user with less telecommunication service usage and expenditure is likely to be evaluated as a low credit user through the existing model.
Limited by the model data source, the accuracy of the credit analysis of the existing model is reduced: with the leap of communication technology and the impact of the internet, the traffic of telecommunication calls and short messages is gradually reduced, the original data analysis dimension cannot completely describe the consumption behavior of the user, and therefore the accuracy of the original credit analysis model is gradually reduced.
The core credit analysis algorithm needs to be further optimized: the original credit analysis model is only based on the linear correlation analysis of the service information of the user, and the credit score of the user is directly obtained by adding all judgment element variables according to weights, but the method has an unsatisfactory effect on the processing of the classification variables and the nonlinear correlation variables, and cannot cope with the interference caused by the internality among the variables and the construction of the autonomous learning capacity after the analysis dimensionality is greatly increased, and certain adjustment and optimization of the traditional modeling thought are required.
Therefore, the embodiment of the invention provides a wind control model training method and a user credit evaluation method, which can improve the credit accuracy of a user.
For convenience in understanding the embodiment of the present invention, a method for training a wind control model in the embodiment of the present invention is first described in detail.
Fig. 2 is a schematic flow chart of a training method of a wind control model according to an embodiment of the present invention.
As shown in fig. 2, the method for training a wind control model according to an embodiment of the present invention may include:
s201: determining target sample data according to the acquired original sample data; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user.
In one embodiment of the present invention, the original sample data is effective data that can effectively reflect individual financial behaviors and agricultural production activities. The original sample data comprises the following data sources:
and acquiring business data related to personal behavior activities, particularly reflecting that personal financial behaviors tend to agricultural production activities from the communication behavior data of the user. User location class, behavior class data.
In addition, related data of partial agricultural production equipment, such as agricultural machinery use condition data and the like, also need to be accessed.
The operator data capable of effectively reflecting personal financial behaviors mainly comprises four categories: basic information (age, gender, academic calendar, address, etc.), behavior preference (interest, hobbies, etc.), life stability (consumption, income, social circle, track, etc.), industry management status (industry, scale, production quality, income status).
The personal position data and the flow content data can be combined with other background data such as map service data and the like to greatly expand the effective information acquisition capacity of the data, and the characteristic variables of financial indexes such as personal repayment capacity, fraud tendency and the like are designed and judged by combining the personal position track, the land parcel attribute of a resident position point and the regional development condition and combining with expert experience.
The related data of the agricultural production equipment can reflect the activity level of regional or even individual agricultural production, and the widely collected data of the agricultural production equipment comprises irrigation water pump pumping water quantity data and harvester usage quantity data related to the planting industry, feed consumption monitoring data and fishpond oxygen supply equipment data related to the breeding industry and the like. After being added into the related industries such as the regional development information data of the aquaculture industry and the information of the aquaculture equipment corresponding to the individual, the credit wind control capability of the personal aquaculture special credit can be greatly enhanced, the credit cost is reduced, the whole credit line is increased, and the business institution and the individual farmer group are helped to realize win-win.
The target sample data comprises user historical behavior information, user behavior time information and user behavior position information. The historical behavior information of the user refers to fusion behavior information and agricultural production activity information in the communication behavior data of the user. The user behavior time information refers to time information of a behavior generated by the user. The user behavior location information refers to location information of a behavior generated by the user.
The target sample data refers to data obtained by performing data processing on original sample data.
The original sample data can be subjected to data processing in the following way, specifically:
carrying out data cleaning on original sample data; the data cleansing includes at least one of: converting original sample data containing null value fields into first preset fields; converting the field format of original sample data into a preset field format; converting first data in original sample data into a second preset field;
and performing feature extraction on the original sample data subjected to data cleaning to obtain target sample data.
In the embodiment of the present invention, the data processing on the original sample data includes the following data processing methods: data cleaning and standardization, data feature extraction and feature engineering.
Wherein the data cleaning and standardization comprises data cleaning of original sample data. In addition, in order to improve data quality and adjust variable metrics, some target sample data needs to be standardized. Specific data cleansing and normalization includes at least one of:
the NULL value field in the original sample data is captured, and may be loaded or replaced by a first preset field, for example, the NULL value field is "NULL", and the NULL value fields are all represented by "0000". And the distribution of different databases can be carried out according to the null value field.
And converting the field format in the original sample data into a preset field format. Data such as time, numerical values, characters and the like can be converted into a preset field format.
According to the service requirement, first data in original sample data, such as: invalid data and missing data are replaced by a second preset field.
The final purpose of data cleaning is to remove and correct various errors in the original data as much as possible, so that the effects of further data analysis and data model construction can be supported. The data measurement units of partial dimensions are different greatly, and the measurement ranges are different, and through the standardization process, the data can be scaled to fall into a small specific interval. The unit limit of the data is removed, the data is converted into a dimensionless pure numerical value, and the indexes of different units or orders of magnitude can be compared and weighted conveniently. For example, the pumping capacity data of a water pump and the accumulated use duration data of the harvester in the agricultural machinery data are uniformly scaled to a 0-1 interval, and after transformation, comparison and weighting operation can be carried out between two variables.
After data cleaning is carried out on original sample data, feature extraction is carried out on the basis of the original sample data after the data cleaning, wherein the problem that the scale of detailed data is huge still exists in the cleaned data, the feature extraction step is mainly applied to processing of operator data, and the whole feature extraction work comprises two parts of feature engineering and feature selection: feature engineering and feature selection.
The feature engineering needs to find out features which may be meaningful from original sample data, namely feature extraction is carried out, and the feature engineering is mainly applied to massive detailed data with time dimension, position dimension and the like, and the data exists in great amount in operator data, such as call detail data. Common methods for feature engineering include statistical methods and empirical methods, wherein the statistical methods are used for performing dimensionality reduction on data by using methods such as principal component analysis and kernel principal component analysis to achieve the purpose of feature extraction. The rule of thumb accomplishes the work of feature extraction by analyzing the actual meaning of the data and by customizing feature combinational logic. In the embodiment, the characteristics are generated in batches by multiplying the main body, the event and the time in three dimensions. Wherein the subject is an analysis object, i.e. a user. An event is a behavioral motive or behavioral preference of an analysis object. An event represents a time when a user has an action.
In addition, a part of the design with practical significance is designed according to business experience, and the design of the processing logic of the part of the feature depends on expert experience and industry background knowledge. In the implementation process, a certain number of characteristic variables of the user, which are related to the financial field and the agricultural production activities, are obtained by processing the behavior characteristics. For example: the social circles with the most closely connected users are depicted graphically. Reflecting multiple core behavior characteristics of the user from different sides according to the portrait description result, wherein the multiple core behavior characteristics include but are not limited to the proportion of the conversation frequency of the first six users with the highest conversation frequency in the last six months of the user to the total conversation frequency, the maximum conversation duration of the user between night and the first six people with the highest conversation frequency, the online duration, the position crossing frequency of the user in the last half year of the day, the overlapping amount of the first six people with the highest conversation frequency in each month in the last six months of the user, and the like; and counting the frequency of the user appearing at the agricultural land position, the longest retention time and the like.
The feature selection work needs to select a small part of core variables with the most obvious effect from a large amount of feature data, the small part of core variables are required to have strong prediction capability and interpretability, and meanwhile, the co-linearity influence among the variables is guaranteed to be as small as possible. Feature selection needs to be performed by means of expert experience knowledge and a machine learning feature selection method at the same time, and a large amount of statistical analysis needs to be performed in the selection process. The final set of core variables can only be determined after repeated comparisons and tradeoffs. For a huge number of basic variable sets except the core features, a machine learning and mathematical statistics method is adopted, and only the features which obviously have no effect and a small amount of adverse effects need to be removed. In the specific implementation process, the characteristic effect evaluation is carried out on the basic characteristic variable set by using algorithms with characteristic selection capability, such as random forests, elastic regression networks (ElasticNet) and the like, and the screening work of the basic characteristic variable set is carried out according to the comprehensive evaluation results of a plurality of algorithms. The random forest algorithm may be a boosting aggregation algorithm (bagging) for decision tree classes. The feature screening mechanism is to evaluate the importance of each feature by accumulating the feature weights corresponding to all split nodes in the whole decision tree group. The elastic regression network algorithm directly achieves the mechanism of scaling the secondary variable coefficients to near zero quantities through the lasso regression (L1) and ridge regression (L2) paradigms to function as feature selection.
Specifically, the method for extracting the features of the original sample data subjected to data cleaning to obtain target sample data includes:
extracting original sample data with time characteristics and position characteristics in the original sample data after data cleaning;
determining a first weight of each original sample in original sample data with time characteristics and position characteristics according to a random forest algorithm;
determining a second weight of each original sample in the original sample data with the time characteristic and the position characteristic according to an elastic network algorithm;
and taking the original sample data with the first weight larger than a first preset value and the second weight larger than a second preset value as target sample data.
As an example, the Feature selection work is performed by combining a random forest algorithm and an elastic network algorithm, wherein the Feature selection criterion of the random forest algorithm is that a first weight (Feature impedance value) is greater than 0.01, and the Feature selection criterion of the elastic network algorithm is that a second weight is greater than 0.005. And taking the original sample data meeting the condition that the first weight is more than 0.01 and the second weight is more than 0.005 in the original sample data as the target sample data.
And after the target sample data is determined, S202 is executed, and the sub-control model is trained.
S202: and training to obtain a wind control model based on the target sample data.
In the face of multi-dimensional data with complex conditions, to exert the mass data information extraction capability of a mathematical model algorithm, firstly, according to a feature screening result, selecting feature variables related to agricultural production and personal financial performance with high reliability and industry application experience from variables, wherein the embodiment includes but is not limited to: the communication frequency of the first six users with the highest communication frequency of the user in the last six months accounts for the proportion of the total communication frequency, the maximum communication time of the user at night and the first six people with the highest communication frequency, the on-line time, the cross-city domain frequency of the user in the last half year day position, the frequency of the user appearing in the agricultural land position and the like. For the remaining massive variables, a mathematical model is trained by using a mathematical model algorithm, and the final model construction strategy output result simultaneously contains a core field model and a supplementary mathematical model result, wherein in the embodiment, the fusion form of the two model results is as follows: and outputting results of different models in intervals, and summing according to the weight ratio of 7:3 to obtain a final credit scoring result.
The wind control model in the embodiment of the invention is constructed based on the XGboost algorithm. In the training process of the wind control mathematical model in this embodiment, the characteristic variables of the core are preferentially used, for the peripheral variables, the data admission quality requirement is lowered, the peripheral variables are allowed to moderately reduce the coverage rate and the accuracy, the lowest coverage rate of part of the peripheral variables can be reduced to 30% as appropriate, and because the number of the peripheral variables is large and the importance of the individual variables is small, data errors within 3% -5% of the total amount are allowed to occur in part of the variables as appropriate, and the measure can increase the adaptability of the model in the production environment.
Specifically, target sample data is input into a wind control model, and credit score of a user is determined;
determining a predicted credit granting result of the user according to the credit score;
and adjusting parameters of the wind control model according to the predicted credit granting result and the real credit granting result of the user to obtain the optimized wind control model.
And the credit score of the user is calculated based on the trained wind control model. And based on the credit score calculated by the wind control model, the predicted credit granting result of the user can be determined. For example, if the credit score of the user is 60 points, the result of credit granting is predicted as credit granting. And if the credit score of the user is 45, the credit granting result is predicted to be non-credit granting. And then, based on the real credit granting result and the predicted credit granting result of the user, the error of the wind control model is checked, and then the parameters of the wind control model are adjusted based on the error, so that a more accurate wind control model is obtained.
In addition, the wind control model obtained by the embodiment of the invention can be tested, the division ratio of the test set to the training set is 80% to 20%, a random sampling method is adopted in the division process, finally, the model effect evaluation index on the test set reaches over 0.65, the discrimination index of a client of the quality of a random sample reaches 0.25 (the larger the value is, the better the classification effect is), the upper level standard in the industry is reached, and the calling time of the wind control model is in millisecond level.
Based on the wind control model obtained in the embodiment, the embodiment of the invention can also be based on a wind control model firmware wind control model system, and the wind control model system needs to meet the functions of automatic data acquisition, data use authorization and authentication, rapid concurrent agricultural credit score calculation and data encryption transmission.
In an embodiment of the invention, a complete agricultural credit wind control system comprises: and (4) front-end real-name authentication, wherein the step can reject the personnel applying for identity obviously. And (4) the front-end credit blacklist hits the library, and the step can directly remove credit blacklist personnel. Front-end hard rule screening, which can reject some unqualified hard condition people, such as underage people, retired people, and the like. And (3) carrying out wind control approval on the mathematical model, wherein a final user credit granting conclusion is obtained according to a user credit scoring result given by the model, and an approval passing conclusion is given to an applicant larger than a threshold value.
According to the wind control model training method provided by the embodiment of the invention, target sample data is determined according to the acquired original sample data; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user; and training to obtain a wind control model based on target sample data, so that the credit evaluation accuracy of the user is improved.
Fig. 3 is a schematic flow chart of a user credit evaluation method according to an embodiment of the present invention.
As shown in fig. 3, a method for evaluating user credit provided by an embodiment of the present invention may include:
s301: and acquiring user behavior information.
The user behavior information is acquired through an operator network, and refers to information of the financial behavior of the user and related to the agricultural activity direction.
S302: and inputting the user behavior information into the wind control model, and determining the credit score of the user.
The user credit score refers to a credit score obtained according to the user behavior, and can also be a credit score obtained by performing credit evaluation on the user behavior. The wind control model may be the wind control model obtained in the embodiment corresponding to fig. 2.
S303: and determining the credit granting result of the user according to the credit score of the user.
And according to the credit score obtained by the wind control model, giving an approval passing conclusion to the user credit score larger than a preset threshold value.
In addition, aiming at the user credit score output by the wind control model, the embodiment of the invention adopts a diversified data exchange mode to construct data service.
Specifically, the credit granting result of the user is stored, and a user credit granting database is constructed.
Data exchange means include, but are not limited to: application Programming Interface (API), files and the like are used for meeting the output result of the wind control model, the wind control model can be docked with heterogeneous service systems or wind control decision systems of different demanders at the lowest cost and the highest efficiency, powerful wind control data support is rapidly provided, the deep combination of cross-boundary and diversified service scenes of the wind control model is realized, and finally the value realization of the wind control model facing to the targeted wind control scenes is realized.
Meanwhile, the external service system also constructs capabilities of service authority management and control, log audit, monitoring management and the like, and meets the requirements of data safety and management and control. When the external credit service is simultaneously served and a plurality of users, the system can continuously run in an efficient, stable, safe and easy-to-use state. In order to further improve the external value promotion of credit service, the service system also provides the flow variable dynamic management and control capability, the dynamic discrimination and the targeted control of the credit service usage amount, the usage frequency and the output result breadth depth can be carried out according to the requirement of different commercial scenes, the credit service provision of each unit is realized, and the corresponding value feedback can be produced.
According to the user credit evaluation method provided by the embodiment of the invention, user behavior information is obtained; inputting the user behavior information into a wind control model, and determining a user credit score; and the credit granting result of the user is determined according to the credit score of the user, so that the credit evaluation accuracy of the user is improved.
Fig. 4 is a schematic structural diagram of a training device for a wind control model according to an embodiment of the present invention.
As shown in fig. 4, a wind control training device provided in an embodiment of the present invention may include: a determination module 401 and a training module 402.
A determining module 401, configured to determine target sample data according to the acquired original sample data; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user;
and a training module 402, configured to train to obtain a wind control model based on the target sample data.
Optionally, in an embodiment, the determining module 401 is specifically configured to:
carrying out data cleaning on original sample data; the data cleansing includes at least one of: converting original sample data containing null value fields into first preset fields; converting the field format of original sample data into a preset field format; converting first data in original sample data into a second preset field;
and performing feature extraction on the original sample data subjected to data cleaning to obtain target sample data.
Optionally, in an embodiment, the determining module 401 is specifically configured to:
extracting original sample data with time characteristics and position characteristics in the original sample data after data cleaning; determining a first weight of each original sample in original sample data with time characteristics and position characteristics according to a random forest algorithm; determining a second weight of each original sample in the original sample data with the time characteristic and the position characteristic according to an elastic network algorithm; and taking the original sample data with the first weight larger than a first preset value and the second weight larger than a second preset value as target sample data.
Optionally, in an embodiment, the determining module 401 is configured to input the target sample data into the wind control model, and determine the credit score of the user.
The determining module 401 is further configured to determine a predicted credit granting result of the user according to the credit score.
And the adjusting module is used for adjusting the parameters of the wind control model according to the predicted credit granting result and the real credit granting result of the user to obtain the optimized wind control model.
The wind control model training device provided by the embodiment of the invention executes each step in the method shown in fig. 2, can achieve the technical effect of improving the credit evaluation accuracy of the user, and is not repeated herein for brevity.
The wind control model training device provided by the embodiment of the invention is used for determining target sample data according to the acquired original sample data through the determining module 401; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user; and the training module 402 is used for training to obtain a wind control model based on the target sample data, so that the credit evaluation accuracy of the user is improved.
Fig. 5 is a schematic structural diagram of a user credit scoring apparatus according to an embodiment of the present invention.
As shown in fig. 5, an apparatus for scoring a user credit according to an embodiment of the present invention may include: an acquisition module 501, a score determination module 502, and a result determination module 503.
An obtaining module 501, configured to obtain user behavior information;
the score determining module 502 is used for inputting the user behavior information into the wind control model and determining the user credit score;
and the result determining module 503 is configured to determine a credit granting result of the user according to the credit score of the user.
Optionally, in an embodiment, the apparatus further comprises a building module;
and the construction module is used for storing the credit granting result of the user and constructing a user credit granting database.
The user credit scoring device provided by the embodiment of the invention executes each step in the method shown in fig. 3, and can achieve the technical effect of improving the user credit evaluation accuracy, and for brevity, detailed descriptions are omitted here.
The user credit scoring device provided by the embodiment of the invention is used for acquiring user behavior information through the acquisition module 501; the score determining module 502 is used for inputting the user behavior information into the wind control model and determining the user credit score; and the result determining module 503 is configured to determine a credit granting result of the user according to the credit score of the user, so as to improve the credit evaluation accuracy of the user.
Fig. 6 is a schematic diagram illustrating a hardware structure of an electronic device according to an embodiment of the present invention.
The electronic device may comprise a processor 601 and a memory 602 in which computer program instructions are stored.
Specifically, the processor 601 may include a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or may be configured as one or more Integrated circuits implementing embodiments of the present invention.
The processor 601 may implement any one of the above-described methods for training a wind control model and for evaluating a user's credit by reading and executing computer program instructions stored in the memory 602.
In one example, the electronic device may also include a communication interface 603 and a bus 610. As shown in fig. 6, the processor 601, the memory 602, and the communication interface 603 are connected via a bus 610 to complete communication therebetween.
The communication interface 603 is mainly used for implementing communication between modules, apparatuses, units and/or devices in the embodiments of the present invention.
The bus 610 includes hardware, software, or both to couple the components of the electronic device to one another. By way of example, and not limitation, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hypertransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an infiniband interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a video electronics standards association local (VLB) bus, or other suitable bus or a combination of two or more of these. Bus 610 may include one or more buses, where appropriate. Although specific buses have been described and shown in the embodiments of the invention, any suitable buses or interconnects are contemplated by the invention.
The electronic device may execute the wind control model training method and the user credit evaluation method in the embodiment of the present invention, so as to implement the wind control model training method described in conjunction with fig. 2 and the user credit evaluation method described in fig. 3.
In addition, in combination with the wind control model training method and the user credit evaluation method in the above embodiments, embodiments of the present invention may provide a computer storage medium to implement. The computer storage medium having computer program instructions stored thereon; the computer program instructions, when executed by a processor, implement any of the above-described embodiments of a method for training a wind-controlled model and a method for assessing user credit.
It is to be understood that the invention is not limited to the specific arrangements and instrumentality described above and shown in the drawings. A detailed description of known methods is omitted herein for the sake of brevity. In the above embodiments, several specific steps are described and shown as examples. However, the method processes of the present invention are not limited to the specific steps described and illustrated, and those skilled in the art can make various changes, modifications and additions or change the order between the steps after comprehending the spirit of the present invention.
The functional blocks shown in the above-described structural block diagrams may be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an Application Specific Integrated Circuit (ASIC), suitable firmware, plug-in, function card, or the like. When implemented in software, the elements of the invention are the programs or code segments used to perform the required tasks. The program or code segments may be stored in a machine-readable medium or transmitted by a data signal carried in a carrier wave over a transmission medium or a communication link. A "machine-readable medium" may include any medium that can store or transfer information. Examples of a machine-readable medium include electronic circuits, semiconductor memory devices, ROM, flash memory, Erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, Radio Frequency (RF) links, and so forth. The code segments may be downloaded via computer networks such as the internet, intranet, etc.
It should also be noted that the exemplary embodiments mentioned in this patent describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above-described steps, that is, the steps may be performed in the order mentioned in the embodiments, may be performed in an order different from the order in the embodiments, or may be performed simultaneously.
As described above, only the specific embodiments of the present invention are provided, and it can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system, the module and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again. It should be understood that the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the present invention, and these modifications or substitutions should be covered within the scope of the present invention.
Claims (13)
1. A method for training a wind control model, the method comprising:
determining target sample data according to the acquired original sample data; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user;
and training to obtain a wind control model based on the target sample data.
2. The method according to claim 1, wherein determining target sample data from the acquired original sample data comprises:
performing data cleaning on the original sample data; the data cleansing includes at least one of: converting original sample data containing null value fields into first preset fields; converting the field format of original sample data into a preset field format; converting first data in original sample data into a second preset field;
and performing feature extraction on the original sample data subjected to data cleaning to obtain the target sample data.
3. The method according to claim 2, wherein the performing feature extraction on the original sample data after data cleaning to obtain the target sample data comprises:
extracting original sample data with time characteristics and position characteristics in the original sample data after data cleaning;
determining a first weight of each original sample in original sample data with time characteristics and position characteristics according to a random forest algorithm;
determining a second weight of each original sample in the original sample data with the time characteristic and the position characteristic according to an elastic network algorithm;
and taking the original sample data of which the first weight is greater than a first preset value and the second weight is greater than a second preset value as the target sample data.
4. The method according to any one of claims 1-3, further comprising:
inputting the target sample data into the wind control model, and determining a credit score of a user;
determining a predicted credit granting result of the user according to the credit score;
and adjusting parameters of the wind control model according to the predicted credit granting result and the real credit granting result of the user to obtain an optimized wind control model.
5. A method for assessing user credit, the method comprising:
acquiring user behavior information;
inputting the user behavior information into a wind control model according to any one of claims 1-4, determining a user credit score;
and determining the credit granting result of the user according to the credit rating of the user.
6. The method of claim 5, further comprising:
and storing the credit granting result of the user, and constructing a user credit granting database.
7. A wind control model training device, the device comprising:
the determining module is used for determining target sample data according to the acquired original sample data; the target sample data comprises user historical behavior information, user behavior time information and user behavior position information, wherein the user historical behavior information is information of agricultural production behaviors of the user;
and the training module is used for training to obtain a wind control model based on the target sample data.
8. The apparatus of claim 7, wherein the determining module is specifically configured to:
performing data cleaning on the original sample data; the data cleansing includes at least one of: converting original sample data containing null value fields into first preset fields; converting the field format of original sample data into a preset field format; converting first data in original sample data into a second preset field;
and performing feature extraction on the original sample data subjected to data cleaning to obtain the target sample data.
9. The apparatus of claim 8, wherein the determining module is specifically configured to:
extracting original sample data with time characteristics and position characteristics in the original sample data after data cleaning;
determining a first weight of each original sample in original sample data with time characteristics and position characteristics according to a random forest algorithm;
determining a second weight of each original sample in the original sample data with the time characteristic and the position characteristic according to an elastic network algorithm;
and taking the original sample data of which the first weight is greater than a first preset value and the second weight is greater than a second preset value as the target sample data.
10. The apparatus of any one of claims 7-9, further comprising:
the determining module is used for inputting the target sample data into the wind control model and determining the credit score of the user;
the determining module is further used for determining a predicted credit granting result of the user according to the credit score;
and the adjusting module is used for adjusting the parameters of the wind control model according to the predicted credit granting result and the real credit granting result of the user to obtain the optimized wind control model.
11. A user credit scoring apparatus, the apparatus comprising:
acquiring user behavior information;
inputting the user behavior information into a wind control model according to any one of claims 1-4, determining a user credit score;
and determining the credit granting result of the user according to the credit rating of the user.
12. An electronic device, characterized in that the device comprises: a processor and a memory storing computer program instructions; the processor, when executing the computer program instructions, implements a method for training a wind-controlled model according to any one of claims 1-4, or a method for assessing user credit according to any one of claims 5-6.
13. A computer storage medium having computer program instructions stored thereon, which when executed by a processor implement the method of training a wind control model according to any one of claims 1-4 or the method of assessing user credit according to any one of claims 5-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911383610.3A CN113052422A (en) | 2019-12-28 | 2019-12-28 | Wind control model training method and user credit evaluation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911383610.3A CN113052422A (en) | 2019-12-28 | 2019-12-28 | Wind control model training method and user credit evaluation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113052422A true CN113052422A (en) | 2021-06-29 |
Family
ID=76507174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911383610.3A Pending CN113052422A (en) | 2019-12-28 | 2019-12-28 | Wind control model training method and user credit evaluation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113052422A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113313328A (en) * | 2021-06-30 | 2021-08-27 | 广州酷狗计算机科技有限公司 | Trust parameter determination method and device, computer equipment and storage medium |
| CN113536672A (en) * | 2021-07-09 | 2021-10-22 | 浙江网商银行股份有限公司 | Target object processing method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150235321A1 (en) * | 2014-02-18 | 2015-08-20 | Mastercard International Incorporated | Insurance risk modeling method and apparatus |
| CN106686265A (en) * | 2015-11-10 | 2017-05-17 | 中国移动通信集团公司 | A service providing method and device based on communication records |
| CN108564286A (en) * | 2018-04-19 | 2018-09-21 | 天合泽泰（厦门）征信服务有限公司 | A kind of artificial intelligence finance air control credit assessment method and system based on big data reference |
| CN109741177A (en) * | 2019-01-09 | 2019-05-10 | 阳光财产保险股份有限公司 | Appraisal procedure, device and the intelligent terminal of user credit |
-
2019
- 2019-12-28 CN CN201911383610.3A patent/CN113052422A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150235321A1 (en) * | 2014-02-18 | 2015-08-20 | Mastercard International Incorporated | Insurance risk modeling method and apparatus |
| CN106686265A (en) * | 2015-11-10 | 2017-05-17 | 中国移动通信集团公司 | A service providing method and device based on communication records |
| CN108564286A (en) * | 2018-04-19 | 2018-09-21 | 天合泽泰（厦门）征信服务有限公司 | A kind of artificial intelligence finance air control credit assessment method and system based on big data reference |
| CN109741177A (en) * | 2019-01-09 | 2019-05-10 | 阳光财产保险股份有限公司 | Appraisal procedure, device and the intelligent terminal of user credit |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113313328A (en) * | 2021-06-30 | 2021-08-27 | 广州酷狗计算机科技有限公司 | Trust parameter determination method and device, computer equipment and storage medium |
| CN113536672A (en) * | 2021-07-09 | 2021-10-22 | 浙江网商银行股份有限公司 | Target object processing method and device |
| CN113536672B (en) * | 2021-07-09 | 2024-05-31 | 浙江网商银行股份有限公司 | Target object processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110400215B (en) | Method and system for constructing enterprise family-oriented small micro enterprise credit assessment model | |
| CN107248082B (en) | Card maintenance identification method and device | |
| WO2017133456A1 (en) | Method and device for determining risk evaluation parameter | |
| CN107403326A (en) | A kind of Insurance Fraud recognition methods and device based on teledata | |
| CN110837963A (en) | Risk control platform construction method based on data, model and strategy | |
| CN111915156B (en) | Service pushing method based on user value, electronic equipment and storage medium | |
| CN106408325A (en) | User consumption behavior prediction analysis method based on user payment information and system | |
| CN112598225A (en) | Evaluation index determination method and apparatus, storage medium, and electronic apparatus | |
| CN112613971A (en) | User credit risk assessment method and device, computer equipment and storage medium | |
| CN110163378A (en) | Characteristic processing method, apparatus, computer readable storage medium and computer equipment | |
| CN113052422A (en) | Wind control model training method and user credit evaluation method | |
| CN114154672A (en) | Data mining method for customer churn prediction | |
| CN115018562A (en) | User pre-churn prediction method, device and system | |
| CN111738824A (en) | Method, device and system for screening financial data processing modes | |
| CN118333684A (en) | Forward incentive convenience service system based on user data asset value | |
| CN113240259B (en) | Rule policy group generation method and system and electronic equipment | |
| CN114782123A (en) | Credit assessment method and system | |
| CN113837512B (en) | Abnormal user identification method and device | |
| CN113450158A (en) | Bank activity information pushing method and device | |
| CN111951099A (en) | Credit card issuing model and application method thereof | |
| CN116843446A (en) | Method for evaluating and attributing infraction risk of small amount credit based on interpretable machine learning | |
| KR102336462B1 (en) | Apparatus and method of credit rating | |
| CN114581209A (en) | Method, device and equipment for training financial analysis model and storage medium | |
| CN110570301B (en) | Risk identification method, device, equipment and medium | |
| CN110766544A (en) | Credit risk detection method and device, storage medium and electronic device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210629 |
|
| RJ01 | Rejection of invention patent application after publication |