CN112950347A - Resource data processing optimization method and device, storage medium and terminal - Google Patents
Resource data processing optimization method and device, storage medium and terminal Download PDFInfo
- Publication number
- CN112950347A CN112950347A CN202110137632.2A CN202110137632A CN112950347A CN 112950347 A CN112950347 A CN 112950347A CN 202110137632 A CN202110137632 A CN 202110137632A CN 112950347 A CN112950347 A CN 112950347A
- Authority
- CN
- China
- Prior art keywords
- characteristic
- resource data
- feature
- data
- resource
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- 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
- G06F16/24—Querying
- G06F16/245—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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"
Abstract
The invention discloses an optimization method and device for resource data processing, a storage medium and a terminal, relates to the technical field of data processing, and mainly aims to solve the problem that the asset value of the existing poor asset cannot meet the accurate requirement of asset value valuation. The method mainly comprises the following steps: acquiring characteristic resource data of the bad assets to be processed; performing prediction processing on the characteristic resource data according to a trained resource prediction model, wherein the resource prediction model is obtained by completing model training after characteristic marking is performed on the characteristic resource data in a training data set based on a characteristic level; and determining repayment intention information of the characteristic resource data obtained through prediction processing as an influence parameter corresponding to the asset value evaluation of the characteristic resource data. The method is mainly used for optimizing the resource data processing.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to an optimization method and apparatus for resource data processing, a storage medium, and a terminal.
Background
In the field of financial credit, borrowers cannot return interest according to term and full amount, and assets which cannot be dealt with by credit enterprises are poor assets, such as mortgage houses and the like which cannot be dealt with. In order to reduce the economic loss caused by the poor assets, credit enterprises utilize a big data method to estimate the asset value of the poor asset data so as to accurately customize the collection promotion scheme.
At present, the asset value of the bad asset data is predicted and estimated only by using a deep learning algorithm, the diversified matching requirements of the collection accelerating scheme cannot be accurately met, and the individual prediction of unknown people cannot be realized, so that the asset value estimation accuracy is poor, and therefore, an optimization method for resource data processing is urgently needed to solve the problems.
Disclosure of Invention
In view of this, the present invention provides an optimization method and apparatus for resource data processing, a storage medium, and a terminal, and mainly aims to solve the problem that the asset value of the existing poor asset cannot meet the requirement for accurate asset value estimation.
According to an aspect of the present invention, there is provided a method for optimizing resource data processing, including:
acquiring characteristic resource data of the bad assets to be processed;
performing prediction processing on the characteristic resource data according to a trained resource prediction model, wherein the resource prediction model is obtained by completing model training after characteristic marking is performed on the characteristic resource data in a training data set based on a characteristic level;
and determining repayment intention information of the characteristic resource data obtained through prediction processing as an influence parameter corresponding to the asset value evaluation of the characteristic resource data.
According to another aspect of the present invention, there is provided an optimization apparatus for resource data processing, including:
the acquisition module is used for acquiring the characteristic resource data of the bad assets to be processed;
the processing module is used for carrying out prediction processing on the characteristic resource data according to a trained resource prediction model, and the resource prediction model is obtained by completing model training after carrying out feature marking on the characteristic resource data in the training data set based on the feature level;
and the determining module is used for determining repayment willingness information of the characteristic resource data obtained through prediction processing as an influence parameter corresponding to the asset value evaluation of the characteristic resource data.
According to another aspect of the present invention, there is provided a storage medium having at least one executable instruction stored therein, where the executable instruction causes a processor to perform operations corresponding to the optimization method for processing resource data as described above.
According to still another aspect of the present invention, there is provided a terminal including: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;
the memory is used for storing at least one executable instruction, and the executable instruction enables the processor to execute the operation corresponding to the optimization method for processing the resource data.
By the technical scheme, the technical scheme provided by the embodiment of the invention at least has the following advantages:
the invention provides an optimization method and device for resource data processing, a storage medium and a terminal. Compared with the prior art, the method and the device have the advantages that the characteristic resource data of the to-be-processed bad assets are obtained; performing prediction processing on the characteristic resource data according to a trained resource prediction model, wherein the resource prediction model is obtained by completing model training after characteristic marking is performed on the characteristic resource data in a training data set based on a characteristic level; and determining repayment intention information of the characteristic resource data obtained by prediction processing as an influence parameter corresponding to the asset value valuation of the characteristic resource data, increasing a calculation parameter of the asset value valuation, accurately meeting diversified collection urging scheme matching requirements, and realizing the prediction requirement of the individualized repayment intention information of unknown people by utilizing a deep learning algorithm, thereby improving the accuracy of the asset value valuation.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flow chart of an optimization method for processing resource data according to an embodiment of the present invention;
FIG. 2 is a block diagram illustrating an apparatus for optimizing resource data processing according to an embodiment of the present invention;
fig. 3 shows a schematic structural diagram of a terminal according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
An embodiment of the present invention provides an optimization method for resource data processing, as shown in fig. 1, the method includes:
101. and acquiring characteristic resource data of the bad assets to be processed.
In the embodiment of the invention, the characteristic resource data is used for characterizing information which is generated by a group or an individual borrower in credit transaction and is related to repayment capacity, such as user basic data, asset data, borrowing data and the like of white-collar people. In order to acquire the characteristic resource data of the bad assets in time and determine the confidentiality of the asset information of the repayment person, the current execution end stores the acquired data in a preset database when processing the resource data. In addition, since different users can perform credit transactions from at least one credit company, in order to determine the payment willingness based on the processing situation of the payers on different credit platforms, when the characteristic resource data of the bad assets to be processed is acquired, the characteristic resource data opened in all the credit companies with the association is acquired, for example, the characteristic resource data generated by acquiring a certain flower and a certain loan used by a crowd a, namely, the asset data, the loan data and the like generated on the certain flower and the loan for the payment crowd a is judged.
It should be noted that the undesirable assets are assets which cannot be returned by the borrower according to term and full amount and cannot be dealt with by the credit enterprise, for example, the assets include mortgage houses and the like which cannot be dealt with.
102. And performing prediction processing on the characteristic resource data according to the trained resource prediction model.
In the embodiment of the invention, in order to realize the data processing capacity of the characteristic resource data and obtain an accurate repayment people's intention prediction result, the characteristic resource data is processed through a trained resource prediction model, and the resource prediction model is obtained by completing model training after characteristic marking is carried out on the characteristic resource data in a training data set based on a characteristic level. The resource prediction model is a mathematical model constructed based on a machine learning algorithm, including but not limited to a neural network model, a support vector machine model and the like, the constructed resource prediction model is trained in advance through a training data set, and when the set training precision and accuracy are reached, the currently trained resource prediction model is determined to complete training so as to be used for prediction processing.
It should be noted that, in the embodiment of the present invention, the feature resource data in the training data set is data that is subjected to feature labeling after being divided according to feature levels, and is mapped to different payment willingness information according to the feature resource data of different feature labels, so that a prediction result of the willingness information of the payer is obtained after the resource prediction model is trained by using the feature resource data. In the embodiment of the invention, the resource prediction model is preferably an xgboost model, so that model training is performed.
103. And determining repayment intention information of the characteristic resource data obtained through prediction processing as an influence parameter corresponding to the asset value evaluation of the characteristic resource data.
In the embodiment of the invention, after the characteristic resource data is subjected to prediction processing, the repayment intention information of the crowd corresponding to the characteristic resource data is obtained, so that the asset value estimation is determined based on the repayment intention information, and the repayment intention information obtained by the prediction processing is determined as the influence parameter of the asset value estimation, so that the asset value estimation is accurately recalculated according to the combination of other influence parameters. The asset value valuation is used for evaluating the value of the undesirable assets, so that a matched collection urging scheme is selected according to different asset value valuations, and the processing accuracy of the asset value valuation serving as resource data is improved. In addition, the influence parameters are parameters for calculating asset value valuation, for the calculation of the asset value valuation, manual estimation can be performed by combining repayment willingness information with other existing influence parameters according to the existing asset valuation calculation mode, prediction estimation can be performed according to a deep learning algorithm, and the like, and the embodiment of the invention is not particularly limited.
In another embodiment, in order to improve the training accuracy of building the resource prediction model and thus improve the resource data processing efficiency, before acquiring the characteristic resource data of the undesirable asset to be processed, the method further includes: acquiring characteristic resource data in a training data set; performing characteristic hierarchy division on the characteristic resource data according to characteristic hierarchies; and performing characteristic marking on the divided characteristic resource data to obtain characteristic resource data marked by different characteristic levels.
Specifically, the characteristic resource data is further limited to at least one of user basic data, asset data, financial attribute data and borrowing data, the user basic data is data of the age, sex, marital status, academic history, work and the like of the user, the asset data is data of whether a house exists, whether a car exists, the number of cars in the house and the like, the financial attribute data is attribute content used for representing the participation of the user in financial transactions and comprises credit conditions such as public deposit, wage, financing, bank card collection, credit card, internet loan and the like and repayment conditions, and the borrowing data is debit amount, debit interest, returned amount, unretained amount, overdue time and the like. The training data set is pre-stored with a mass of feature resource data of users, and in order to improve the processing efficiency of the feature resource data and reduce the processing resource time, the feature resource data are divided into feature levels according to the feature levels. The feature hierarchy is used for distinguishing a first level of feature classification in the feature resource data, for example, the feature classification is classified according to age features, so that the classification of 5 age hierarchies can be obtained, wherein the classification is greater than or equal to 18 years and less than 22 years (less than 18 years without borrowing capability), greater than or equal to 22 years and less than 27 years, greater than or equal to 27 years and equal to 50 years, greater than or equal to 50 years and less than 65 years, and then the feature resource data of different age hierarchies are subjected to feature marking. In the embodiment of the present invention, the feature label is a method for labeling and identifying a feature level, and the identifier may include a score, a color, a symbol, and the like.
It should be noted that, since the feature resource data may be of one borrower or may include a plurality of borrowers, the feature resource data is subjected to feature hierarchy division, and the feature resource data of each user is subjected to one feature hierarchy division, for example, the user basic data, the asset data, the financial attribute data, and the borrowing data of the borrower a are subjected to feature hierarchy division, and the user basic data, the asset data, the financial attribute data, and the borrowing data of the borrower b are subjected to feature hierarchy division, and are sequentially advanced to complete the hierarchy division of all the borrowers.
In yet another embodiment, for further definition and illustration, the feature hierarchy includes at least an age hierarchy, an arrears amount hierarchy, and the feature hierarchy partitioning of the feature resource data by feature hierarchy includes: determining age information from the user basic data, and determining that the characteristic resource data is in a first characteristic level in the age levels according to the age information; determining repayment information from the borrowing data, and determining that the characteristic resource data is in a second characteristic level in the arrearage amount level relative to the first characteristic level according to the repayment information; determining model attribute information that matches the second feature hierarchy based on the asset data, the financial attribute data.
For the embodiment of the invention, in order to improve the accuracy of hierarchical division of the feature resource data and thus the training accuracy of the resource prediction model, the feature hierarchy at least comprises an age hierarchy and a debt amount hierarchy, therefore, the age information is determined by using the user basic data in the feature resource data, the first feature hierarchy is divided, so that the basic life state of the user is obtained according to different age hierarchies, for example, the academic work of 18 years or more and less than 22 years old is not stable, the academic work of 22 years or more and less than 27 years old is stable, the people of 27 years or more and 50 years old enter a marital child education stage, the educational factors of 50 years or more and less than 65 years old are reduced, and the people of 65 years old or more exceed the loan application age. And then, determining repayment information by using borrowing data in the characteristic resource data, for example, whether the repayment is carried out on time, whether the debt is overdue and the like, and determining that the characteristic resource data is in a second characteristic level in the arrearage amount level relative to the first characteristic level according to the repayment information. Specifically, the second characteristic hierarchy is a low level, a medium level and a high level which represent arrearage levels, and different age hierarchies each include 3 arrearage levels, that is, after the first characteristic hierarchy which is an age hierarchy is determined, the low level or the high level in the arrearage level of the age hierarchy, for example, between 22 and 27 years can be further determined.
It should be noted that, in order to improve the training optimization effect on the resource prediction model, the corresponding prediction models at different feature levels have different effects, and therefore, the model attribute information matched with the second feature level is determined based on the asset data and the financial attribute data in the feature resource data. The model attribute information represents weight configuration conditions of different algorithm models, the model attribute information corresponding to different age levels and different arrearage amount levels is different, for example, a first characteristic level of a 25-year borrower is a substantially stable academic work, a second characteristic level is a middle level, and the matched model attribute information is that the active repayment intention of each platform, the overdue condition of each platform and the model weight value configuration of the working condition are larger, so that the active repayment intention of each platform, the overdue condition of each platform, the working condition and the like in the recent interest of each platform in model training are reflected, and the embodiment of the invention is not specifically limited.
In another embodiment, for further limitation and explanation, the performing feature labeling on the divided feature resource data to obtain feature resource data labeled at different feature levels includes: and carrying out feature marking on the first feature level, the second feature level and the model attribute information of the feature resource data according to preset tag content, wherein the preset tag content is used for limiting repayment willingness information matched with the first feature level, the second feature level and the model attribute information.
In the embodiment of the invention, in order to improve the determination of the repayment wish information after the characteristic hierarchy of the characteristic resource data information is divided, the first characteristic hierarchy, the second characteristic hierarchy and the model attribute information of the characteristic resource data are subjected to characteristic marking according to the preset label content. The preset label content is used for limiting the repayment willingness information matched with the first characteristic level, the second characteristic level and the model attribute information, so that the matched repayment willingness information is determined for the characteristic resource data corresponding to the first characteristic level, the second characteristic level and the model attribute information. Specifically, the preset label content is a preset label value between 0 and 1, and is characterized by the strong degree of the payment willingness, for example, the characteristic marking of a 23-year-old low-level borrower is utilized to be 0.8, and the payment willingness is strong.
In yet another embodiment, in order to satisfy the prediction of the payment willingness information so as to improve the accuracy of the asset value estimation of the bad assets processed according to the resource data, the method further comprises the following steps: performing model construction pretreatment on the feature resource data subjected to feature marking, wherein the model construction pretreatment at least comprises one of numerical treatment, feature normalization treatment, abnormal value treatment and feature screening treatment; constructing a resource prediction model through a determined deep learning algorithm, and training the resource prediction model based on the model attribute information and feature resource data which are preprocessed by model construction; and when the model training parameters match the preset model requirements, determining the resource prediction model which completes training and has the prediction processing result of repayment willingness information.
Specifically, in order to improve the training efficiency of building the resource prediction model, model building preprocessing, that is, at least including digitization processing, feature normalization processing, abnormal value processing, feature screening processing, and the like, is performed on the feature resource data on which the feature labels are completed in advance. The digitization processing is to convert data such as texts into numerical data so as to be used as input of model training for training operation, and after the digitization, the numerical value ranges of all the characteristics are inconsistent, so that the sizes of the different types of characteristic values are adjusted to be consistent through the characteristic normalization processing, and the characteristic normalization processing can be performed through a linear scale transformation method, a 0-mean normalization method and the like. In the process, in order to avoid the influence of abnormal data on the model training efficiency, abnormal value processing can be performed, namely, the numerical values in the abnormal range are screened out, and data with stable numerical value distribution are reserved. In addition, in order to reduce the running time of the learning algorithm and increase the interpretability of the model, feature screening may also be performed, including methods such as Principal Component Analysis (PCA), recursive feature elimination, and the like.
In addition, a resource prediction model can be constructed through a selected deep learning algorithm, wherein the deep learning algorithm can comprise a Convolutional Neural Network (CNN), a cyclic neural network (RNN) and the like, then weight values in the constructed resource prediction model are determined by utilizing model attribute information, and the resource prediction model is trained by utilizing feature resource data subjected to model construction preprocessing. And determining whether to stop the iterative process of training or not by presetting model requirements, wherein the preset model requirements comprise preset model precision and model errors, so as to finish model training.
In yet another embodiment, for further limitation and explanation, the determining, as the impact parameters corresponding to the asset value valuation of the feature resource data, the repayment willingness information of the feature resource data obtained by the prediction processing includes: analyzing repayment intention factors in the repayment intention information, wherein the repayment intention factors at least comprise one of behavior factors, economic factors and time factors; and determining the asset value valuation matched with the repayment willingness factor based on the preset asset value valuation corresponding relation so as to determine a repayment urging scheme according to the asset value valuation.
In the embodiment of the invention, in order to accurately determine the asset value evaluation by using the obtained repayment intention information and analyze the repayment intention factors in the repayment intention information, the repayment intention information can be strong, general, poor, extremely poor and the like, so the repayment intention factors are analyzed by combining user basic data, asset data, financial attribute data and borrowing data in the characteristic resource data, for example, if the repayment intention is strong, and 3 suites of rooms are obtained according to the asset data in the characteristic resource data, the bank income in the financial attribute data is greater than the expenditure, and the repayment according to date in the borrowing data can be analyzed to be an economic factor. And after the repayment intention factors are determined, determining matched asset value valuation by utilizing a preset asset value valuation corresponding relation, wherein the preset asset value valuation corresponding relation can comprise a calculation method of the asset value valuation and can also be matched and mapped with the repayment intention factors, so that the asset value valuation is determined based on the asset value valuation corresponding relation. Further, when determining the repayment urging scheme by using the asset value estimate, the repayment urging scheme is a specific urging manner and urging time for urging the repayment of the repayment person, for example, if the asset value estimate is large, the matched repayment urging scheme may be twice-a-month call urging, and the embodiment of the present invention is not particularly limited.
The embodiment of the invention provides an optimization method for resource data processing. Compared with the prior art, the method and the device have the advantages that the characteristic resource data of the to-be-processed bad assets are obtained; performing prediction processing on the characteristic resource data according to a trained resource prediction model, wherein the resource prediction model is obtained by completing model training after characteristic marking is performed on the characteristic resource data in a training data set based on a characteristic level; and determining repayment intention information of the characteristic resource data obtained by prediction processing as an influence parameter corresponding to the asset value valuation of the characteristic resource data, accurately meeting diversified collection urging scheme matching requirements, and realizing the prediction requirement of individualized repayment intention information of unknown people by using a deep learning algorithm, thereby improving the asset value valuation accuracy.
Further, as an implementation of the method shown in fig. 1, an embodiment of the present invention provides an optimization apparatus for resource data processing, as shown in fig. 2, the apparatus includes:
an obtaining module 21, configured to obtain feature resource data of a to-be-processed bad asset;
the processing module 22 is configured to perform prediction processing on the feature resource data according to a trained resource prediction model, where the resource prediction model is obtained by performing feature labeling on feature resource data in a training data set based on a feature level and then completing model training;
and the determining module 23 is configured to determine payment willingness information of the feature resource data obtained through the prediction processing as an influence parameter corresponding to the asset value estimation of the feature resource data.
Further, the apparatus further comprises: a dividing module, a marking module,
the acquisition module is further used for acquiring characteristic resource data in the training data set;
the dividing module is used for dividing the feature resource data according to feature levels;
the marking module is used for carrying out feature marking on the divided feature resource data to obtain feature resource data marked by different feature levels.
Further, the characteristic resource data includes at least one of user basic data, asset data, financial attribute data and borrowing data.
Further, the feature levels at least comprise an age level and an arrearage level, and the dividing module comprises:
a first determining unit, configured to determine age information from the user basic data, and determine, according to the age information, that the feature resource data is in a first feature level of the age levels;
a second determining unit, configured to determine repayment information from the borrowing data, and determine, according to the repayment information, that the feature resource data is at a second feature level in the arrearage amount level relative to the first feature level;
and a third determining unit, configured to determine model attribute information that matches the second feature hierarchy based on the asset data and the financial attribute data.
Further, the marking module is specifically configured to perform feature marking on the first feature level, the second feature level, and the model attribute information of the feature resource data according to preset tag content, where the preset tag content is used to limit repayment willingness information matched with the first feature level, the second feature level, and the model attribute information.
Further, the apparatus further comprises: the building block is constructed by the method of the building block,
the processing module is further configured to perform model construction preprocessing on the feature resource data subjected to the feature labeling, where the model construction preprocessing at least includes one of digitization processing, feature normalization processing, outlier processing, and feature screening processing;
the building module is used for building a resource prediction model through a determined deep learning algorithm and training the resource prediction model based on the model attribute information and feature resource data which are preprocessed by model building;
and the determining module is used for determining the resource prediction model which is trained and has a prediction processing result of repayment willingness information when the model training parameters are matched with the preset model requirements.
Further, the determining module includes: an analysis module for analyzing the received data,
the analysis module is used for analyzing repayment intention factors in the repayment intention information, wherein the repayment intention factors at least comprise one of behavior factors, economic factors and time factors;
and the determining module is used for determining the asset value valuation matched with the repayment willingness factor based on the preset asset value valuation corresponding relation so as to determine a repayment urging scheme according to the asset value valuation.
The embodiment of the invention provides an optimization device for resource data processing. Compared with the prior art, the method and the device have the advantages that the characteristic resource data of the to-be-processed bad assets are obtained; performing prediction processing on the characteristic resource data according to a trained resource prediction model, wherein the resource prediction model is obtained by completing model training after characteristic marking is performed on the characteristic resource data in a training data set based on a characteristic level; and determining repayment intention information of the characteristic resource data obtained by prediction processing as an influence parameter corresponding to the asset value valuation of the characteristic resource data, accurately meeting diversified collection urging scheme matching requirements, and realizing the prediction requirement of individualized repayment intention information of unknown people by using a deep learning algorithm, thereby improving the asset value valuation accuracy.
According to an embodiment of the present invention, a storage medium is provided, where at least one executable instruction is stored, and the computer executable instruction can execute the optimization method of resource data processing in any of the above method embodiments.
Fig. 3 is a schematic structural diagram of a terminal according to an embodiment of the present invention, and the specific embodiment of the present invention does not limit the specific implementation of the terminal.
As shown in fig. 3, the terminal may include: a processor (processor)302, a communication Interface 304, a memory 306, and a communication bus 308.
Wherein: the processor 302, communication interface 304, and memory 306 communicate with each other via a communication bus 308.
A communication interface 304 for communicating with network elements of other devices, such as clients or other servers.
The processor 302 is configured to execute the program 310, and may specifically perform relevant steps in the above-described optimization method for processing resource data.
In particular, program 310 may include program code comprising computer operating instructions.
The processor 302 may be a central processing unit CPU, or an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement an embodiment of the present invention. The terminal comprises one or more processors, which can be the same type of processor, such as one or more CPUs; or may be different types of processors such as one or more CPUs and one or more ASICs.
And a memory 306 for storing a program 310. Memory 306 may comprise high-speed RAM memory and may also include non-volatile memory (non-volatile memory), such as at least one disk memory.
The program 310 may specifically be configured to cause the processor 302 to perform the following operations:
acquiring characteristic resource data of the bad assets to be processed;
performing prediction processing on the characteristic resource data according to a trained resource prediction model, wherein the resource prediction model is obtained by completing model training after characteristic marking is performed on the characteristic resource data in a training data set based on a characteristic level;
and determining repayment intention information of the characteristic resource data obtained through prediction processing as an influence parameter corresponding to the asset value evaluation of the characteristic resource data.
It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method for optimizing resource data processing, comprising:
acquiring characteristic resource data of the bad assets to be processed;
performing prediction processing on the characteristic resource data according to a trained resource prediction model, wherein the resource prediction model is obtained by completing model training after characteristic marking is performed on the characteristic resource data in a training data set based on a characteristic level;
and determining repayment intention information of the characteristic resource data obtained through prediction processing as an influence parameter corresponding to the asset value evaluation of the characteristic resource data.
2. The method of claim 1, wherein prior to obtaining the characteristic resource data of the undesirable asset to be processed, the method further comprises:
acquiring characteristic resource data in a training data set;
performing characteristic hierarchy division on the characteristic resource data according to characteristic hierarchies;
and performing characteristic marking on the divided characteristic resource data to obtain characteristic resource data marked by different characteristic levels.
3. The method according to claim 1 or 2, wherein the characteristic resource data comprises at least one of user base data, asset data, financial attribute data, and debit data.
4. The method of claim 3, wherein the feature hierarchy includes at least an age hierarchy, an arrears amount hierarchy, and wherein the feature hierarchy partitioning of the feature resource data by feature hierarchy includes:
determining age information from the user basic data, and determining that the characteristic resource data is in a first characteristic level in the age levels according to the age information;
determining repayment information from the borrowing data, and determining that the characteristic resource data is in a second characteristic level in the arrearage amount level relative to the first characteristic level according to the repayment information;
determining model attribute information that matches the second feature hierarchy based on the asset data, the financial attribute data.
5. The method according to claim 4, wherein the performing feature labeling on the divided feature resource data to obtain feature resource data labeled at different feature levels comprises:
and carrying out feature marking on the first feature level, the second feature level and the model attribute information of the feature resource data according to preset tag content, wherein the preset tag content is used for limiting repayment willingness information matched with the first feature level, the second feature level and the model attribute information.
6. The method of claim 4, further comprising:
performing model construction pretreatment on the feature resource data subjected to feature marking, wherein the model construction pretreatment at least comprises one of numerical treatment, feature normalization treatment, abnormal value treatment and feature screening treatment;
constructing a resource prediction model through a determined deep learning algorithm, and training the resource prediction model based on the model attribute information and feature resource data which are preprocessed by model construction;
and when the model training parameters match the preset model requirements, determining the resource prediction model which completes training and has the prediction processing result of repayment willingness information.
7. The method according to any one of claims 1 to 6, wherein the determining of the repayment willingness information of the feature resource data obtained by the prediction processing as the influence parameter corresponding to the asset value estimation of the feature resource data comprises:
analyzing repayment intention factors in the repayment intention information, wherein the repayment intention factors at least comprise one of behavior factors, economic factors and time factors;
and determining the asset value valuation matched with the repayment willingness factor based on the preset asset value valuation corresponding relation so as to determine a repayment urging scheme according to the asset value valuation.
8. An apparatus for optimizing processing of resource data, comprising:
the acquisition module is used for acquiring the characteristic resource data of the bad assets to be processed;
the processing module is used for carrying out prediction processing on the characteristic resource data according to a trained resource prediction model, and the resource prediction model is obtained by completing model training after carrying out feature marking on the characteristic resource data in the training data set based on the feature level;
and the determining module is used for determining repayment willingness information of the characteristic resource data obtained through prediction processing as an influence parameter corresponding to the asset value evaluation of the characteristic resource data.
9. A storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the optimization method for resource data processing according to any one of claims 1 to 7.
10. A terminal, comprising: the system comprises a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface complete mutual communication through the communication bus;
the memory is used for storing at least one executable instruction, and the executable instruction causes the processor to execute the operation corresponding to the optimization method of the resource data processing according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110137632.2A CN112950347B (en) | 2021-02-01 | 2021-02-01 | Resource data processing optimization method and device, storage medium and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110137632.2A CN112950347B (en) | 2021-02-01 | 2021-02-01 | Resource data processing optimization method and device, storage medium and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112950347A true CN112950347A (en) | 2021-06-11 |
| CN112950347B CN112950347B (en) | 2023-01-13 |
Family
ID=76240946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110137632.2A Active CN112950347B (en) | 2021-02-01 | 2021-02-01 | Resource data processing optimization method and device, storage medium and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112950347B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114118083A (en) * | 2021-12-02 | 2022-03-01 | 帕特思科技咨询(杭州)有限公司 | Industrial resource information matching optimization method |
| CN115660811A (en) * | 2022-11-07 | 2023-01-31 | 杭州度言软件有限公司 | Asset management method for improving recovery rate of consumption financial overdue assets |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020013752A1 (en) * | 1999-12-30 | 2002-01-31 | Johnson Christopher D. | Rapid valuation of portfolios of assets such as financial instruments |
| CN108416669A (en) * | 2018-03-13 | 2018-08-17 | 腾讯科技（深圳）有限公司 | User behavior data processing method, device, electronic equipment and computer-readable medium |
| CN108985919A (en) * | 2018-06-11 | 2018-12-11 | 北京玖富普惠信息技术有限公司 | A kind of class information statistical method, electronic equipment and readable storage medium storing program for executing |
| CN110060144A (en) * | 2019-03-18 | 2019-07-26 | 平安科技（深圳）有限公司 | Amount model training method, amount appraisal procedure, device, equipment and medium |
| CN110070430A (en) * | 2019-03-12 | 2019-07-30 | 平安科技（深圳）有限公司 | Assess method and device, the storage medium, electronic equipment of refund risk |
| CN110610412A (en) * | 2019-09-02 | 2019-12-24 | 深圳中兴飞贷金融科技有限公司 | Credit risk assessment method and device, storage medium and electronic equipment |
| CN111145009A (en) * | 2019-12-12 | 2020-05-12 | 北京淇瑀信息科技有限公司 | Method and device for evaluating risk after user loan and electronic equipment |
| CN111429245A (en) * | 2020-03-31 | 2020-07-17 | 王春年 | Method and device for assessing value of poor asset creditor |
| CN111986026A (en) * | 2020-09-07 | 2020-11-24 | 中国银行股份有限公司 | Personal credit business risk assessment method and device based on analytic hierarchy process |
-
2021
- 2021-02-01 CN CN202110137632.2A patent/CN112950347B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020013752A1 (en) * | 1999-12-30 | 2002-01-31 | Johnson Christopher D. | Rapid valuation of portfolios of assets such as financial instruments |
| CN108416669A (en) * | 2018-03-13 | 2018-08-17 | 腾讯科技（深圳）有限公司 | User behavior data processing method, device, electronic equipment and computer-readable medium |
| CN108985919A (en) * | 2018-06-11 | 2018-12-11 | 北京玖富普惠信息技术有限公司 | A kind of class information statistical method, electronic equipment and readable storage medium storing program for executing |
| CN110070430A (en) * | 2019-03-12 | 2019-07-30 | 平安科技（深圳）有限公司 | Assess method and device, the storage medium, electronic equipment of refund risk |
| CN110060144A (en) * | 2019-03-18 | 2019-07-26 | 平安科技（深圳）有限公司 | Amount model training method, amount appraisal procedure, device, equipment and medium |
| CN110610412A (en) * | 2019-09-02 | 2019-12-24 | 深圳中兴飞贷金融科技有限公司 | Credit risk assessment method and device, storage medium and electronic equipment |
| CN111145009A (en) * | 2019-12-12 | 2020-05-12 | 北京淇瑀信息科技有限公司 | Method and device for evaluating risk after user loan and electronic equipment |
| CN111429245A (en) * | 2020-03-31 | 2020-07-17 | 王春年 | Method and device for assessing value of poor asset creditor |
| CN111986026A (en) * | 2020-09-07 | 2020-11-24 | 中国银行股份有限公司 | Personal credit business risk assessment method and device based on analytic hierarchy process |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114118083A (en) * | 2021-12-02 | 2022-03-01 | 帕特思科技咨询(杭州)有限公司 | Industrial resource information matching optimization method |
| CN115660811A (en) * | 2022-11-07 | 2023-01-31 | 杭州度言软件有限公司 | Asset management method for improving recovery rate of consumption financial overdue assets |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112950347B (en) | 2023-01-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111340616B (en) | Method, device, equipment and medium for approving online loan | |
| CN109739844B (en) | Data classification method based on attenuation weight | |
| CN110704730A (en) | Product data pushing method and system based on big data and computer equipment | |
| CN107704512A (en) | Financial product based on social data recommends method, electronic installation and medium | |
| CN114265967B (en) | Sensitive data security level marking method and device | |
| CN112950347B (en) | Resource data processing optimization method and device, storage medium and terminal | |
| CN111523678A (en) | Service processing method, device, equipment and storage medium | |
| CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
| CN111210332A (en) | Method and device for generating post-loan management strategy and electronic equipment | |
| CN111091408A (en) | User identification model creating method and device and identification method and device | |
| CN112487284A (en) | Bank customer portrait generation method, equipment, storage medium and device | |
| CN108647714A (en) | Acquisition methods, terminal device and the medium of negative label weight | |
| CN110597984B (en) | Method and device for determining abnormal behavior user information, storage medium and terminal | |
| CN112836750A (en) | System resource allocation method, device and equipment | |
| CN116821759A (en) | Identification prediction method and device for category labels, processor and electronic equipment | |
| CN116629423A (en) | User behavior prediction method, device, equipment and storage medium | |
| CN112529628B (en) | Client label generation method and device, computer equipment and storage medium | |
| CN113779116A (en) | Object sorting method, related equipment and medium | |
| CN114048330A (en) | Risk conduction probability knowledge graph generation method, device, equipment and storage medium | |
| CN113052512A (en) | Risk prediction method and device and electronic equipment | |
| CN110570301A (en) | Risk identification method, device, equipment and medium | |
| CN116484230B (en) | Method for identifying abnormal business data and training method of AI digital person | |
| Oelerich et al. | Evaluation of rating systems | |
| CN112396513B (en) | Data processing method and device | |
| CN113591932A (en) | User abnormal behavior processing method and device based on support vector machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |