CN113837506A - Logistics data-based wind control modeling method and device and computer equipment - Google Patents
Logistics data-based wind control modeling method and device and computer equipment Download PDFInfo
- Publication number
- CN113837506A CN113837506A CN202010510371.XA CN202010510371A CN113837506A CN 113837506 A CN113837506 A CN 113837506A CN 202010510371 A CN202010510371 A CN 202010510371A CN 113837506 A CN113837506 A CN 113837506A
- Authority
- CN
- China
- Prior art keywords
- variables
- derivative
- logistics
- index
- financial
- 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
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
- 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
-
- 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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
- G06Q10/0838—Historical data
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a method and a device for wind control modeling based on logistics data, computer equipment and a storage medium. The method comprises the following steps: acquiring logistics data, screening the logistics data, and determining logistics variables with financial characteristics; performing index derivation on the logistics variables based on credit index derivation logic to obtain derivative logistics variables; verifying the financial characteristics of the derivative flow variables and determining the financial characteristics of the derivative flow variables; and selecting a derivative flow variable according to the financial characteristic degree for modeling to obtain a wind control model. By adopting the method, the generalization capability and the accuracy of the model can be improved.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for wind-controlled modeling based on logistics data, a computer device, and a storage medium.
Background
With the development of various industries, the importance of risk control work in various industries is self evident, especially in the financial industry. Generally, according to the needs of the business and in order to avoid unnecessary loss, a certain risk judgment needs to be performed on the customer group. Traditionally, the judgment of the passenger group risk is realized by utilizing a wind control model obtained by wind control modeling based on credit data of a user.
However, due to data security supervision and the limited availability of users for actual use of financial products, credit data is lacking, thereby reducing the generalization ability and accuracy of the wind control model.
Disclosure of Invention
In view of the above, it is necessary to provide a method, an apparatus, a computer device and a storage medium for modeling a wind control based on logistics data, which can improve generalization ability and accuracy of a wind control model.
A wind control modeling method based on logistics data is characterized by comprising the following steps:
acquiring logistics data, screening the logistics data, and determining logistics variables with financial characteristics;
performing index derivation on the logistics variables based on credit index derivation logic to obtain derivative logistics variables;
verifying the financial characteristics of the derivative flow variable and determining the financial characteristic degree of the derivative flow variable;
and selecting the derivative flow variable according to the financial characteristic degree for modeling to obtain a wind control model.
In one embodiment, the index derivation of the logistics variables based on credit index derivation logic to derive derivative logistics variables comprises:
performing primary index derivation on the logistics variables to obtain primary derivation indexes;
carrying out depth index derivation on the logistics variables according to a preset time period interval to obtain a depth derivation index;
and combining the primary derivation index and the depth derivation index to obtain a derivative flow variable.
In one embodiment, the deriving the primary indicator of the stream variable to obtain a primary derived indicator includes:
and carrying out any one or more of summation, average calculation, maximum calculation, minimum calculation, frequency statistics, variance calculation and variation coefficient calculation on the logistics variables to obtain a primary derivative index.
In one embodiment, the depth index derivation performed on the logistics variables according to a preset time period interval to obtain a depth derivation index includes:
dividing and counting the logistics variables according to a preset time period interval to obtain a plurality of period derivative variables;
and performing index derivation based on each period derivative variable to obtain a depth derivative index.
In one embodiment, the deriving an index based on each of the cycle derived variables to obtain a depth derived index includes:
and carrying out any one or more of comparison, frequency statistics, extreme value calculation, mean value calculation, dispersion calculation, summation and proportion calculation on the basis of the period derivative variables to obtain a depth derivative index.
In one embodiment, the verifying the financial characteristics of the derivative flow variable and determining the financial characteristics of the derivative flow variable comprises:
calculating the information value of each derivative flow variable;
and determining the financial characteristic degree of the derivative flow variable according to the information value.
In one embodiment, the selecting the derivative flow variable according to the financial characteristic degree for modeling to obtain a wind control model includes:
screening out derivative flow variables with unqualified financial characteristics according to the financial characteristic degree to obtain residual flow variables;
selecting a derivative flow variable from the residual flow variables as a model entering variable according to a characteristic degree threshold value;
and performing logistic regression modeling based on the model entering variables to obtain a wind control model.
A wind-controlled modeling apparatus based on logistics data, the apparatus comprising:
the screening module is used for acquiring logistics data, screening the logistics data and determining logistics variables with financial characteristics;
the derivative module is used for performing index derivation on the logistics variables based on credit index derivation logic to obtain derivative logistics variables;
the verification module is used for verifying the financial characteristics of the derivative flow variables and determining the financial characteristic degrees of the derivative flow variables;
and the modeling module is used for selecting the derivative flow variable according to the financial characteristic degree to perform modeling so as to obtain a wind control model.
Computer equipment, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above-mentioned logistics data-based wind control modeling methods when executing the computer program.
A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for logistics data based wind control modeling according to any of the above.
According to the method, the device, the computer equipment and the storage medium for wind control modeling based on the logistics data, the acquired logistics data are screened to obtain the logistics variables with financial characteristics preliminarily, the logistics variables are derived based on the credit index derivation logic, the derived logistics variables are verified for the financial characteristics, the financial characteristic degree is determined, and finally, the variables are selected according to the financial characteristic degree to perform wind control model modeling. The method selects modeling variables through processes of screening logistics data, deriving and verifying financial characteristics based on credit indexes and the like, fully applies the logistics data to the field of financial wind control, and adds new risk identification dimensionality to a wind control model, so that the generalization capability and the accuracy of the model are improved.
Drawings
FIG. 1 is a diagram of an application environment of a method for modeling wind control based on logistics data according to an embodiment;
FIG. 2 is a schematic flow chart of a method for modeling wind control based on logistics data according to one embodiment;
FIG. 3 is a flowchart illustrating the steps of index-deriving a derivative flow variable based on credit index derivation logic in one embodiment;
FIG. 4 is a block diagram of a structure of a wind control modeling device based on logistics data according to an embodiment;
FIG. 5 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The logistics data-based wind control modeling method can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. After the terminal 102 acquires the logistics data, the terminal 102 can separately implement the logistics data-based wind control modeling method. The terminal 102 may also send the logistics data to the server 104 for communication, and the server 104 implements a method for wind control modeling based on the logistics data. Taking the server 104 as an example, specifically, the server 104 obtains logistics data, screens the logistics data, and determines a logistics variable with financial characteristics; the server 104 performs index derivation on the logistics variables based on credit index derivation logic to obtain derivative logistics variables; the server 104 verifies the financial characteristics of the derivative flow variables and determines the financial characteristic degrees of the derivative flow variables; the server 104 selects the derivative flow variables to model according to the financial characteristic degree, and a wind control model is obtained. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, a method for modeling wind control based on logistics data is provided, which is illustrated by taking the method as an example applied to the server in fig. 1, and includes the following steps:
step S202, acquiring logistics data, screening the logistics data, and determining logistics variables with financial characteristics.
The logistics data refers to data generated by the user performing logistics activities, such as freight rate, weight, consignment name, address, sender, and the like. The logistics variables with financial characteristics refer to logistics data which can be used in financial products for risk control.
Specifically, after the server acquires the logistics data, the logistics data is screened first. And removing the logistics data obviously without financial characteristics, wherein the rest logistics data are logistics variables with financial characteristics. The data apparently having no financial characteristics includes internal fields and the like, such as data of basic information of logistics employees.
And step S204, performing index derivation on the logistics variables based on the credit index derivation logic to obtain derivative logistics variables.
The credit index derivation logic refers to a derivation rule for deriving a credit index in the financial field, and the index derivation can be understood as an index calculated twice based on a basic index.
Specifically, after the server obtains the logistics variables with financial characteristics through screening, the credit index derivation logic which is configured locally in advance is obtained. And performing index derivation on the logistics variables according to a derivation rule recorded in the credit index derivation logic to obtain derivative logistics variables. Wherein the derivative flow variable may comprise the flow variable being derived. I.e., the derivative flow variables include the original flow variables and the flow variables derived based on the flow variables.
And step S206, verifying the financial characteristics of the derivative flow variables and determining the financial characteristic degrees of the derivative flow variables.
The financial characteristic degree is used for representing the degree of the derivative flow variable having the financial characteristic, and the higher the financial characteristic degree is, the more remarkable the financial characteristic of the derivative flow variable is.
Specifically, after the server derives the derivative flow variable through the index, the derivative flow variable is verified for the financial characteristics. The financial characteristics of the derivative flow variables are obtained by calculating the Information Value (IV) of each derivative flow variable.
And S208, selecting a derivative flow variable according to the financial characteristic degree for modeling to obtain a wind control model.
Specifically, after the server determines the financial characteristic degree of each derivative flow variable through verification, the derivative flow variables completely without the financial characteristic are screened out through the financial characteristic degree, and the residual flow variables are obtained. And then, returning the financial characteristic degree corresponding to the residual flow variable to the terminal. And selecting the corresponding financial characteristic degree by the terminal user according to the actual requirement of the wind control model. And after the server receives the financial feature degree selected by the user and returned by the terminal, screening the corresponding residual flow variable from the residual flow variables according to the financial feature degree selected by the user to be used as a modulus entering variable. And finally, performing logistic regression modeling based on the screened model entering variables to obtain a wind control model.
The financial characteristics include fail, pass, good, excellent. For example, assume that the financial characteristics of the derivative flow variable derived this time include only three degrees of fail, pass, and good. The server firstly screens out the derivative variable with unqualified financial characteristic degree to obtain the residual variable with qualified financial characteristic degree and good financial characteristic degree. And then the server returns the qualified financial feature degree and the good financial feature degree to the terminal, and the terminal user selects the qualified financial feature degree and the good financial feature degree according to actual requirements. When the server receives the end-user selected good, the server selects a good derivative flow variable from the residual flow variables as the in-mode variable. When the user of the terminal selects qualified, the server can select only the qualified derivative flow variables or can select both the qualified and the good derivative flow variables because the good is better than the qualified. It should be understood that the server may also return the derivative flow variable corresponding to the financial characteristic degree to the terminal, so as to facilitate the terminal user to view the specific derivative flow index.
According to the wind control modeling method based on the logistics data, the logistics variables with financial characteristics are obtained preliminarily by screening the acquired logistics data, the logistics variables are derived based on credit index derivation logic, the derived flow variables are verified for the financial characteristics, the financial characteristic degree is determined, and finally, the variables are selected according to the financial characteristic degree to perform modeling of the wind control model. The method selects modeling variables through processes of screening logistics data, deriving and verifying financial characteristics based on credit indexes and the like, fully applies the logistics data to the field of financial wind control, and adds new risk identification dimensionality to a wind control model, so that the generalization capability and the accuracy of the model are improved.
In one embodiment, as shown in fig. 3, step S204 includes:
and step S302, performing primary index derivation on the logistics variables to obtain primary derived indexes.
The primary index derivation refers to performing conventional description statistical derivation on all logistics variables, and includes any one or more of summation, average calculation, maximum calculation, minimum calculation, frequency statistics, variance calculation and coefficient of variation calculation.
Specifically, when the server performs index derivation on the logistics variables based on the credit index derivation logic, the server performs summation, average calculation, maximum calculation, minimum calculation, frequency statistics, variance calculation and variation coefficient calculation on the logistics variables, and the values obtained by the calculation are derived indexes to obtain primary derived indexes. It should be understood that index derivation may be the simultaneous calculation of any one or more of the variables of the stream. For example, assuming that the logistics variables include freight rate and weight, the freight rate and the weight may be summed, averaged, maximum, minimum, statistics, variance, and coefficient of variation, respectively. The obtained freight sum, maximum freight value, minimum freight value, mean freight value, freight times, variance of freight and the like are primary derived indexes of freight. And the obtained weight sum, weight maximum value, weight minimum value, weight average value, weight variance and the like are primary derivative indexes corresponding to the weight.
And step S304, carrying out depth index derivation on the logistics variables according to a preset time period interval to obtain a depth derivation index.
The preset time period interval refers to a set time interval, such as day, month, year, etc. In this embodiment, the predetermined time period interval is preferably a month.
Specifically, the logistics variables are divided according to a preset time period interval, and index derivation is performed by taking the divided variables as units to obtain a depth derivation index.
In one embodiment, step S304 includes: dividing and counting the logistics variables according to a preset time period interval to obtain a plurality of period derivative variables; and performing index derivation based on the derived data of each period to obtain a depth derived index.
Taking the preferred month of the embodiment as the preset time period interval as an example, dividing all the logistics variables into statistics by month to obtain a plurality of cycle derivative variables, wherein one cycle derivative variable includes a variable obtained by statistics from the logistics variables of one month.
Specifically, the logistics variables take the freight as an example, the server divides and counts the freight according to months, counts the freight in january, counts the freight in february, … …, and counts the freight in december to obtain twelve cycle derivative variables, respectively, and each cycle derivative variable includes a variable obtained by counting the freight. And then, performing index derivation on the derived variables of each period, wherein the index derivation comprises any one or more of comparison, frequency statistics, extreme value calculation, mean value calculation, dispersion calculation, summation and proportion calculation, so as to obtain the depth derived index.
The comparison of the derived variables for each cycle can be from the last m months, the last mon >0 to the current number of months. Last m months, last mon ═ 0 to current month parts. Maximum of mon increase every two months for the last m months. Maximum reduction of mon every two months for the last m months. The number of months in the last m months is increased in the latter month compared to the former month. In the last m months, the latter month is a reduced number of months from the former month. The number of months equal to mon in the nearest m months. The number of months equal to mon in the nearest m months. The month in which the last m months mon takes the maximum value is a number from the current month. And if in the last m months, for any month i, there is mon [ i ] > mon [ i +1], i.e. strictly increasing is kept, and mon >0, then label (label) ═ 1Else label (label) ═ 0. If in the last m months, for any month i, mon [ i ] < mon [ i +1], i.e. strictly decreasing, and mon >0, then label ═ 1Else label ═ 0.
The count may be the number of months of the last m months, mon > 0. Month fraction of mon ═ 0 for the last m months. And in the last m months, whether the month number of mon >0 is > 1.
The extremum calculation may be the most recent m months, mon max. Last m months, mon minimum.
The mean calculation may be the most recent m months, mon mean. The mean of mon constructed over the last m months.
The dispersion calculation may be the mean square error of mon, the last m months. Last m months, the coefficient of variation of mon. (mean of mon last 1 month) - (mon last m months). (mon last 1 month) - (minimum of mon last m months). (mon last 1 month) - (maximum mon last m months). ((last 1 month mon) - (maximum of last m months mon))/(maximum of last m months mon)). Mon was very poor for the last m months.
The summation calculation may be the mon sum of the last m months, and the mon sum of the last (2, m +1) months.
The occupancy calculation may be the mean of mon/(mon last m months). Mon/(minimum of mon last m months). Mon in the last 1 month/maximum in mon in the last m months. (mon last 1 month-mon mean last m months)/mon mean. (mon last 1 month-mon minimum last m months)/mon minimum. Mean of last m months/((m, 2m) month mon mean). Mean of last m months- ((mon mean of m,2m) months). (mon max for the last m months)/(mon max for the last (m,2m) months). (mon min for the last m months)/(mon min for the last (m,2m) months). Where mon represents the monthly statistics, the logistics variables are illustrated as freight rates, and mon is the monthly statistics of freight rates.
And S306, combining the primary derivative indexes and the depth derivative indexes to obtain derivative flow variables.
Specifically, after the primary derivation index and the depth derivation index are obtained, the primary derivation index and the depth derivation index are combined to obtain the derivative flow variable. For example, 50 primary derivative indices and 3000 deep derivative indices, 3050 derivative flow variables were obtained from the combination.
In the embodiment, the logistics variables are subjected to primary index derivation and depth index derivation, so that indexes with financial characteristics at a deeper level can be derived by fully utilizing logistics data, and sufficient model entering variables are provided for wind control modeling.
In one embodiment, step S206 includes: calculating the information value of each derivative flow variable; and determining the financial characteristic degree of the derivative flow variable according to the information value.
The Information Value (IV) is mainly used for encoding and predicting capability evaluation of the variable, and the financial characteristic degree of the derivative flow variable is determined through the Information Value, that is, the capability of encoding and predicting the derivative flow variable in the wind control model is determined.
Specifically, when the server verifies the financial characteristics of each derivative flow variable, the information value of each derivative flow variable is calculated. The information value of the derivative flow variable can be calculated in any conventional manner, for example, by WOE (weight of evidence). The server then determines the financial characterization of the derivative flow variables based on the information value. And when the information value of the derivative flow variable is less than 0.02, determining that the derivative flow variable is not suitable for wind control modeling, and the derivative flow variable does not have financial characteristics, and the financial characteristics are unqualified. And when the information value of the derivative flow variable is greater than or equal to 0.02 and less than 0.1, determining the financial characteristic degree of the derivative flow variable as qualified. When the information value of the derivative flow variable is equal to or more than 0.1 and less than 0.5, the financial character degree of the derivative flow variable is determined to be good. When the information value of the derivative flow variable is 0.5 or more, the financial character degree of the derivative flow variable is determined to be excellent.
In the embodiment, the financial characteristic degree of the derivative flow variable is evaluated through the information value, and reference can be given to the subsequent selection of the model entering variable.
In one embodiment, step S208 includes: screening out derivative flow variables with unqualified financial characteristics according to the financial characteristic degree to obtain residual flow variables; selecting a derivative flow variable from the residual flow variables according to the characteristic degree threshold value as a model entering variable; and performing logistic regression modeling based on the model entering variables to obtain the wind control model.
Wherein, the feature threshold is configured in advance and used for restricting the financial feature of the module entering variable, including any one or more of qualification, good and excellence.
Specifically, after the server determines the financial characteristics of each derivative flow variable, the derivative flow variables with poor financial characteristics are first screened out. Namely, the server deletes the derivative flow variables with the information value less than 0.02 to obtain residual flow variables. Then, a preset feature threshold is obtained. And the server selects derivative flow variables with the golden fusion characteristic degrees equal to the characteristic degree threshold value from the residual flow variables, for example, the characteristic degree threshold value is qualified, and the server selects the qualified derivative flow variables from the residual flow variables as the model entering variables. Or selecting derivative flow variables with the golden characteristic degree greater than or equal to the characteristic degree threshold from the residual flow variables, for example, the characteristic degree threshold is qualified, and the server simultaneously selects qualified, good and excellent derivative flow variables from the residual flow variables as the model entry variables. And finally, the server performs logistic regression modeling based on the selected model entering variables to obtain the wind control model.
In addition, after the server obtains the wind control model, the KS value of the wind control model can be further calculated, and the distinguishing capability of the wind control model obtained through modeling based on the selected model entering variable is evaluated through the KS value, so that the accuracy of the wind control model is determined. And when the KS value of the framed wind control model does not meet the threshold requirement, a new model entering variable can be selected again according to the financial feature degree to perform modeling again. For example, if the KS value of the wind control model obtained by modeling the originally selected derivative flow variables with qualified financial characteristics does not meet the threshold requirement, the derivative flow variables with good financial characteristics can be reselected for modeling again. And when the originally selected model entering variables simultaneously comprise qualified and good derivative flow variables, the qualified derivative flow variables can be deleted, and the model can be built again by using the remaining good derivative flow variables.
In the embodiment, the proper model entering variable of the wind control model is selected through the financial characteristic degree, so that the accuracy of selecting the model entering variable is ensured, and the accuracy of the model is improved.
In one embodiment, when derivative flow variables are derived from indices, the derivative flow variables may be stored according to different levels of indices.
Specifically, derivative flow variables are stored by four classes of level indicators, including field-based classification, attribute-based classification, statistical-sense-based classification, and code-based implementation of the indicator classification, respectively. The field classification means that corresponding derivative flow indexes, such as address, cost, time and the like, can be queried by directly using the field, that is, all derivative flow variables related to the address can be queried through the address field. The attribute-based classification is an attribute according to the derivative flow variable itself, and includes a numerical type, a category type and a time type. That is, all the derivative flow variables of the numerical type can be queried by the numerical type. The classification based on statistical significance is based on the derivation logic, including times, extrema, summation, mean, maximum, minimum, dispersion, and ratio. For example, derivative flow variables that are extrema may be queried by extremum tags. The index classification is realized based on the code, that is, the derivative flow variable is queried through a code tag added to the derivative flow variable, and the code tag can be customized, for example, the code tag can include dstc, avg, cnt, max _ min, mean _ var, min, num, cav, and the like.
In this embodiment, the derivative flow variables are stored by using indexes of different levels, which facilitates the query of the derivative flow variables.
It should be understood that although the various steps in the flow diagrams of fig. 2-3 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-3 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.
In one embodiment, as shown in fig. 4, there is provided a wind control modeling apparatus based on logistics data, including: a screening module 402, a derivation module 404, a verification module 406, and a modeling module 408, wherein:
and the screening module 402 is configured to obtain logistics data, screen the logistics data, and determine a logistics variable with financial characteristics.
And a derivation module 404, configured to perform index derivation on the logistics variable based on the credit index derivation logic to obtain a derivative logistics variable.
And the verification module 406 is used for verifying the financial characteristics of the derivative flow variables and determining the financial characteristic degrees of the derivative flow variables.
And the modeling module 408 is used for selecting a derivative flow variable according to the financial characteristic degree to perform modeling to obtain a wind control model.
In one embodiment, the derivation module 404 is further configured to perform primary index derivation on the logistics variable to obtain a primary derivation index; carrying out depth index derivation on the logistics variables according to a preset time period interval to obtain a depth derivation index; and combining the primary derivation index and the depth derivation index to obtain the derivative flow variable.
In one embodiment, the derivation module 404 is further configured to perform any one or more of summation, average calculation, maximum calculation, minimum calculation, order statistics, variance calculation, and coefficient of variation calculation on the logistics variables to obtain the primary derivation index.
In one embodiment, the derivation module 404 is further configured to perform division statistics on the logistics variables according to a preset time period interval to obtain a plurality of period derived variables; and performing index derivation based on the derivative variables of each period to obtain a depth derivative index.
In one embodiment, the deriving module 404 is further configured to obtain the depth derivative indicator based on any one or more of comparison, order statistics, extremum calculation, mean calculation, dispersion calculation, summation, and proportion calculation of each of the periodic derivative variables.
In one embodiment, the verification module 406 is further configured to calculate the information value of each derivative flow variable; and determining the financial characteristic degree of the derivative flow variable according to the information value.
In one embodiment, the modeling module 408 is further configured to screen out the derivative flow variables with unqualified financial characteristics according to the financial characteristics to obtain residual flow variables; selecting a derivative flow variable from the residual flow variables according to the characteristic degree threshold value as a model entering variable; and performing logistic regression modeling based on the model entering variables to obtain the wind control model.
For specific definition of the logistics data-based wind control modeling device, reference may be made to the above definition of the logistics data-based wind control modeling method, and details are not repeated here. All or part of each module in the logistics data based wind control modeling device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, the internal structure of which may be as shown in fig. 5. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing the logistics data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of wind-controlled modeling based on logistics data.
Those skilled in the art will appreciate that the architecture shown in fig. 5 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:
acquiring logistics data, screening the logistics data, and determining logistics variables with financial characteristics;
performing index derivation on the logistics variables based on credit index derivation logic to obtain derivative logistics variables;
verifying the financial characteristics of the derivative flow variables and determining the financial characteristics of the derivative flow variables;
and selecting a derivative flow variable according to the financial characteristic degree for modeling to obtain a wind control model.
In one embodiment, the processor, when executing the computer program, further performs the steps of: carrying out primary index derivation on the logistics variables to obtain primary derivation indexes; carrying out depth index derivation on the logistics variables according to a preset time period interval to obtain a depth derivation index; and combining the primary derivation index and the depth derivation index to obtain the derivative flow variable.
In one embodiment, the processor, when executing the computer program, further performs the steps of: and carrying out any one or more of summation, average value calculation, maximum value calculation, minimum value calculation, frequency statistics, variance calculation and variation coefficient calculation on the logistics variables to obtain a primary derivative index.
In one embodiment, the processor, when executing the computer program, further performs the steps of: dividing and counting the logistics variables according to a preset time period interval to obtain a plurality of period derivative variables; and performing index derivation based on the derivative variables of each period to obtain a depth derivative index.
In one embodiment, the processor, when executing the computer program, further performs the steps of: and carrying out any one or more of comparison, frequency statistics, extreme value calculation, mean value calculation, dispersion calculation, summation and proportion calculation on the basis of the period derivative variables to obtain a depth derivative index.
In one embodiment, the processor, when executing the computer program, further performs the steps of: calculating the information value of each derivative flow variable; and determining the financial characteristic degree of the derivative flow variable according to the information value.
In one embodiment, the processor, when executing the computer program, further performs the steps of: screening out derivative flow variables with unqualified financial characteristics according to the financial characteristic degree to obtain residual flow variables; selecting a derivative flow variable from the residual flow variables according to the characteristic degree threshold value as a model entering variable; and performing logistic regression modeling based on the model entering variables to obtain the wind control model.
In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:
acquiring logistics data, screening the logistics data, and determining logistics variables with financial characteristics;
performing index derivation on the logistics variables based on credit index derivation logic to obtain derivative logistics variables;
verifying the financial characteristics of the derivative flow variables and determining the financial characteristics of the derivative flow variables;
and selecting a derivative flow variable according to the financial characteristic degree for modeling to obtain a wind control model.
In one embodiment, the computer program when executed by the processor further performs the steps of: carrying out primary index derivation on the logistics variables to obtain primary derivation indexes; carrying out depth index derivation on the logistics variables according to a preset time period interval to obtain a depth derivation index; and combining the primary derivation index and the depth derivation index to obtain the derivative flow variable.
In one embodiment, the computer program when executed by the processor further performs the steps of: and carrying out any one or more of summation, average value calculation, maximum value calculation, minimum value calculation, frequency statistics, variance calculation and variation coefficient calculation on the logistics variables to obtain a primary derivative index.
In one embodiment, the computer program when executed by the processor further performs the steps of: dividing and counting the logistics variables according to a preset time period interval to obtain a plurality of period derivative variables; and performing index derivation based on the derivative variables of each period to obtain a depth derivative index.
In one embodiment, the computer program when executed by the processor further performs the steps of: and carrying out any one or more of comparison, frequency statistics, extreme value calculation, mean value calculation, dispersion calculation, summation and proportion calculation on the basis of the period derivative variables to obtain a depth derivative index.
In one embodiment, the computer program when executed by the processor further performs the steps of: calculating the information value of each derivative flow variable; and determining the financial characteristic degree of the derivative flow variable according to the information value.
In one embodiment, the computer program when executed by the processor further performs the steps of: screening out derivative flow variables with unqualified financial characteristics according to the financial characteristic degree to obtain residual flow variables; selecting a derivative flow variable from the residual flow variables according to the characteristic degree threshold value as a model entering variable; and performing logistic regression modeling based on the model entering variables to obtain the wind control model.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (10)
1. A wind control modeling method based on logistics data is characterized by comprising the following steps:
acquiring logistics data, screening the logistics data, and determining logistics variables with financial characteristics;
performing index derivation on the logistics variables based on credit index derivation logic to obtain derivative logistics variables;
verifying the financial characteristics of the derivative flow variable and determining the financial characteristic degree of the derivative flow variable;
and selecting the derivative flow variable according to the financial characteristic degree for modeling to obtain a wind control model.
2. The method of claim 1, wherein the index-derived the logistic variable based on credit index derivation logic to derive a derivative logistic variable comprises:
performing primary index derivation on the logistics variables to obtain primary derivation indexes;
carrying out depth index derivation on the logistics variables according to a preset time period interval to obtain a depth derivation index;
and combining the primary derivation index and the depth derivation index to obtain a derivative flow variable.
3. The method of claim 2, wherein the deriving the primary indicator of the stream variable to obtain a primary derived indicator comprises:
and carrying out any one or more of summation, average calculation, maximum calculation, minimum calculation, frequency statistics, variance calculation and variation coefficient calculation on the logistics variables to obtain a primary derivative index.
4. The method of claim 2, wherein the depth-index deriving the logistics variables at predetermined time period intervals to obtain a depth-derived index comprises:
dividing and counting the logistics variables according to a preset time period interval to obtain a plurality of period derivative variables;
and performing index derivation based on each period derivative variable to obtain a depth derivative index.
5. The method of claim 4, wherein said deriving an index based on each of said cycle derived variables to obtain a depth derived index comprises:
and carrying out any one or more of comparison, frequency statistics, extreme value calculation, mean value calculation, dispersion calculation, summation and proportion calculation on the basis of the period derivative variables to obtain a depth derivative index.
6. The method of claim 1, wherein said validating a financial characteristic of said derivative flow variable and determining a financial characteristic measure of said derivative flow variable comprises:
calculating the information value of each derivative flow variable;
and determining the financial characteristic degree of the derivative flow variable according to the information value.
7. The method of claim 1, wherein selecting the derivative flow variables for modeling according to the financial characteristics to obtain a wind control model comprises:
screening out derivative flow variables with unqualified financial characteristics according to the financial characteristic degree to obtain residual flow variables;
selecting a derivative flow variable from the residual flow variables as a model entering variable according to a characteristic degree threshold value;
and performing logistic regression modeling based on the model entering variables to obtain a wind control model.
8. A wind-controlled modeling apparatus based on logistics data, the apparatus comprising:
the screening module is used for acquiring logistics data, screening the logistics data and determining logistics variables with financial characteristics;
the derivative module is used for performing index derivation on the logistics variables based on credit index derivation logic to obtain derivative logistics variables;
the verification module is used for verifying the financial characteristics of the derivative flow variables and determining the financial characteristic degrees of the derivative flow variables;
and the modeling module is used for selecting the derivative flow variable according to the financial characteristic degree to perform modeling so as to obtain a wind control model.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010510371.XA CN113837506A (en) | 2020-06-08 | 2020-06-08 | Logistics data-based wind control modeling method and device and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010510371.XA CN113837506A (en) | 2020-06-08 | 2020-06-08 | Logistics data-based wind control modeling method and device and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113837506A true CN113837506A (en) | 2021-12-24 |
Family
ID=78963367
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010510371.XA Pending CN113837506A (en) | 2020-06-08 | 2020-06-08 | Logistics data-based wind control modeling method and device and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113837506A (en) |
-
2020
- 2020-06-08 CN CN202010510371.XA patent/CN113837506A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11757938B2 (en) | Method, apparatus, and computer-readable medium for data protection simulation and optimization in a computer network | |
| US10496815B1 (en) | System, method, and computer program for classifying monitored assets based on user labels and for detecting potential misuse of monitored assets based on the classifications | |
| US10178108B1 (en) | System, method, and computer program for automatically classifying user accounts in a computer network based on account behavior | |
| CN110223133B (en) | Product display method, device, computer equipment and storage medium | |
| US20170068723A1 (en) | Organization categorization system and method | |
| CN111597348A (en) | User image drawing method, device, computer equipment and storage medium | |
| CN110222733B (en) | High-precision multi-order neural network classification method and system | |
| CN112633962A (en) | Service recommendation method and device, computer equipment and storage medium | |
| CN107622326A (en) | User's classification, available resources Forecasting Methodology, device and equipment | |
| JP2016206878A (en) | Salesperson raising support system and salesperson raising support method | |
| CN112559900A (en) | Product recommendation method and device, computer equipment and storage medium | |
| US20150120731A1 (en) | Preference based clustering | |
| CN115545103A (en) | Abnormal data identification method, label identification method and abnormal data identification device | |
| CN114140152A (en) | Cloud platform customer management system and method | |
| CN112990583A (en) | Method and equipment for determining mold entering characteristics of data prediction model | |
| CN112215655A (en) | Client portrait label management method and system | |
| CN116501979A (en) | Information recommendation method, information recommendation device, computer equipment and computer readable storage medium | |
| CN113837506A (en) | Logistics data-based wind control modeling method and device and computer equipment | |
| CN115759742A (en) | Enterprise risk assessment method and device, computer equipment and storage medium | |
| CN114170000A (en) | Credit card user risk category identification method, device, computer equipment and medium | |
| CN113127563A (en) | Intelligent retail management method and system based on block chain | |
| CN113779116A (en) | Object sorting method, related equipment and medium | |
| WO2017062026A1 (en) | Generating cohorts using automated weighting and multi-level ranking | |
| US20150269502A1 (en) | Method and apparatus for finding process deviations and recommending repairs in a business process | |
| CN113094615A (en) | Message pushing method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |