CN113362115A - Transaction resource analysis method, device, equipment and medium based on machine learning - Google Patents

Transaction resource analysis method, device, equipment and medium based on machine learning Download PDF

Info

Publication number
CN113362115A
CN113362115A CN202110732054.7A CN202110732054A CN113362115A CN 113362115 A CN113362115 A CN 113362115A CN 202110732054 A CN202110732054 A CN 202110732054A CN 113362115 A CN113362115 A CN 113362115A
Authority
CN
China
Prior art keywords
transaction
market
trading
service
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.)
Pending
Application number
CN202110732054.7A
Other languages
Chinese (zh)
Inventor
江川
罗力力
白育龙
程茜
罗水权
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Ping An Asset Management Co Ltd
Original Assignee
Ping An Asset Management Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Ping An Asset Management Co Ltd filed Critical Ping An Asset Management Co Ltd
Priority to CN202110732054.7A priority Critical patent/CN113362115A/en
Publication of CN113362115A publication Critical patent/CN113362115A/en
Pending legal-status Critical Current

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
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • 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
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
    • 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
    • G06Q40/06—Asset management; Financial planning or analysis

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Data Mining & Analysis (AREA)
  • Operations Research (AREA)
  • Software Systems (AREA)
  • Technology Law (AREA)
  • Tourism & Hospitality (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Quality & Reliability (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

The application relates to a transaction resource analysis method, a device, equipment and a medium based on machine learning. The method comprises the following steps: and receiving a resource analysis request, and acquiring the market transaction service to be processed and the target special clause carried by the resource analysis request. And obtaining a modeling factor corresponding to the market trading service according with the target special term and corresponding initial trading resources, and training a preset initial analysis model to obtain a trading resource analysis model. And based on the transaction resource analysis model, carrying out valuation analysis on the market transaction service to be processed to generate a corresponding special term transaction resource analysis result, and generating corresponding service recommendation information according to the special term transaction resource analysis result. By adopting the method, the transaction resource analysis result of the special term can be automatically calculated according to the trained transaction resource analysis model, the accuracy of the corresponding analysis result is improved, the asset safety of the investment user is ensured, and the stability of the internet financial market is maintained.

Description

Transaction resource analysis method, device, equipment and medium based on machine learning
Technical Field
The present application relates to the field of artificial intelligence technologies, and in particular, to a method, an apparatus, a device, and a medium for analyzing transaction resources based on machine learning.
Background
With the increasing development of artificial intelligence technology and the gradual popularization of internet financial services, more and more users select different types of internet financial services according to the actual requirements of the users, such as different types of bonds, investment products, financial products and the like. With the increasing abundance of bond categories, the investment is gradually refined, and in order to ensure the vital interests of investment users and maintain the stable financial market, the interest and the handicap of different bond categories are increasingly paid attention.
The traditional calculation method for bond profit is mostly based on the yield of bonds and corresponding reference yield to carry out simple calculation, taking the guarantee terms as an example, the guarantee profit can be obtained by calculating the difference between the yield of the guarantee bond and the yield of the reference curve thereof, or the guarantee profit can be obtained by calculating the difference between the yield of the main body non-guarantee bond and the yield of the main body guarantee bond, or the guarantee profit can be obtained by calculating the difference between the yield of the same-term same-rating curve and the yield of the guarantee bond.
However, the traditional calculation method for bond difference is rough, bonds of different categories are not considered comprehensively, for example, the difference with special terms or a subject with fewer surviving bonds is considered, and when the bond difference is calculated, the remaining time limits of the surviving bonds of different subjects are different, so that the reference bond searching and the time limit difference adjustment cannot be unified. Therefore, when the traditional method for calculating the profit difference of the bonds needs to calculate the profit difference of the bonds of different types, the actual conditions of the bonds cannot be considered comprehensively, the universality is low, the calculated profit difference value of the bonds has large error and the accuracy is low.
Disclosure of Invention
In view of the above, it is necessary to provide a machine learning-based transaction resource analysis method, device, apparatus and medium capable of reducing calculated bond variance error values and improving bond variance accuracy.
A machine learning based transaction resource analysis method, the method comprising:
receiving a resource analysis request, and acquiring market transaction service to be processed and target special terms carried by the resource analysis request;
obtaining a modeling factor corresponding to market transaction business according with the target special terms and corresponding initial transaction resources, and training a preset initial analysis model to obtain a transaction resource analysis model;
based on the trading resource analysis model, carrying out valuation analysis on the market trading service to be processed to generate a corresponding special clause trading resource analysis result;
and generating corresponding service recommendation information according to the special clause transaction resource analysis result.
In one embodiment, training a preset initial analysis model to obtain a transaction resource analysis model includes:
acquiring a first element factor corresponding to a market transaction service according with a target special term and a second element factor corresponding to the target special term;
performing element coding processing based on the first element factor and the second element factor to generate a corresponding modeling factor;
acquiring initial trading resources corresponding to market trading services according with target special terms;
and training the preset initial analysis model based on each modeling factor and the initial transaction resource to generate a trained transaction resource analysis model.
In one embodiment, the obtaining of the initial trading resource corresponding to the market trading service meeting the target special term includes:
extracting market transaction services conforming to the target special terms, and determining corresponding sample transaction services according to the market transaction services conforming to the target special terms;
acquiring special clause transaction resource profitability of the market transaction service of the target special clause and non-clause transaction resource profitability corresponding to the sample transaction service;
and determining initial trading resources corresponding to the market trading service according with the target special terms according to the special term trading resource profitability and the non-term trading resource profitability.
In one embodiment, the performing element encoding processing based on the first element factor and the second element factor to generate a corresponding modeling factor includes:
determining continuous variable factors and classification variable factors from the first element factors and the second element factors;
performing element coding processing based on the classification variable factors, and converting each classification variable factor into a numerical variable factor;
carrying out standardization processing based on the continuous variable factors to generate corresponding discrete data factors;
and carrying out normalization processing based on the numerical variable factor and the discrete data factor to generate a corresponding modeling factor.
In one embodiment, the manner of generating a non-clause trading resource profitability corresponding to the sample trading service comprises:
obtaining a yield curve of the same grade and the same industry and the yield of the sample transaction service;
adjusting the rate of return of the sample transaction service according to the rate of return curve of the same grade and the same industry, and generating an adjusted average value of the rate of return of the sample transaction service;
and determining the adjusted average value of the profitability of the sample transaction service as the profitability of the non-clause transaction resource corresponding to the sample transaction service.
In one embodiment, the extracting market transaction services meeting the target special terms and determining corresponding sample transaction services according to the market transaction services meeting the target special terms includes:
extracting market transaction services which accord with target special terms;
according to the market transaction service meeting the target special terms, searching the open issue transaction service with the same main body continuity;
and determining the business of the public development bank with the same main body as the corresponding sample transaction business.
A machine learning based transaction resource analysis apparatus, the apparatus comprising:
the system comprises a to-be-processed market transaction service acquisition module, a resource analysis module and a target special term acquisition module, wherein the to-be-processed market transaction service acquisition module is used for receiving a resource analysis request and acquiring the to-be-processed market transaction service and the target special term carried by the resource analysis request;
the trading resource analysis model acquisition module is used for acquiring a trading resource analysis model obtained by training a preset initial analysis model according to a modeling factor corresponding to the market trading service meeting the target special term and the corresponding initial trading resource;
the special clause transaction resource analysis result generation module is used for carrying out valuation analysis on the market transaction service to be processed based on the trained transaction resource analysis model and generating a corresponding special clause transaction resource analysis result;
and the service recommendation information generation module is used for generating corresponding service recommendation information according to the special clause transaction resource analysis result.
In one embodiment, the apparatus further comprises:
the element factor acquisition module is used for acquiring a first element factor corresponding to the market transaction service meeting target special terms and a second element factor corresponding to the target special terms;
the modeling factor generation module is used for carrying out element coding processing on the basis of the first element factor and the second element factor to generate a corresponding modeling factor;
the initial trading resource acquisition module is used for acquiring initial trading resources corresponding to the market trading service according with the target special terms;
and the transaction resource analysis model generation module is used for training the preset initial analysis model based on each modeling factor and the initial transaction resource to generate a trained transaction resource analysis model.
A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:
receiving a resource analysis request, and acquiring market transaction service to be processed and target special terms carried by the resource analysis request;
obtaining a modeling factor corresponding to market transaction business according with the target special terms and corresponding initial transaction resources, and training a preset initial analysis model to obtain a transaction resource analysis model;
based on the trading resource analysis model, carrying out valuation analysis on the market trading service to be processed to generate a corresponding special clause trading resource analysis result;
and generating corresponding service recommendation information according to the special clause transaction resource analysis result.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
receiving a resource analysis request, and acquiring market transaction service to be processed and target special terms carried by the resource analysis request;
obtaining a modeling factor corresponding to market transaction business according with the target special terms and corresponding initial transaction resources, and training a preset initial analysis model to obtain a transaction resource analysis model;
based on the trading resource analysis model, carrying out valuation analysis on the market trading service to be processed to generate a corresponding special clause trading resource analysis result;
and generating corresponding service recommendation information according to the special clause transaction resource analysis result.
In the transaction resource analysis method, the transaction resource analysis device, the computer equipment and the storage medium based on machine learning, the resource analysis request is received, the market transaction service to be processed and the target special terms carried by the resource analysis request are obtained, the preset initial analysis model is trained to obtain the transaction resource analysis model by obtaining the modeling factors corresponding to the market transaction service according with the target special terms and the corresponding initial transaction resources, the market transaction service to be processed is evaluated and analyzed based on the transaction resource analysis model to generate the corresponding special term transaction resource analysis results, and then the corresponding service recommendation information is generated according to the special term transaction resource analysis results. Due to the fact that proper reference bonds or term interest difference adjustment and other point data do not need to be found manually, good performance of the machine learning model is combined, transaction resource analysis results of special terms can be obtained through automatic calculation according to the trained transaction resource analysis model, accuracy of the corresponding analysis results is improved, corresponding business recommendation information is further generated according to the resource analysis results with high accuracy, accurate recommendation business is provided for investment users, asset safety of the investment users is guaranteed, and stability of the internet financial market is maintained.
Drawings
FIG. 1 is a diagram of an application environment of a machine learning-based transaction resource analysis method in one embodiment;
FIG. 2 is a flow diagram illustrating a method for machine learning-based transaction resource analysis in one embodiment;
FIG. 3 is a schematic diagram illustrating a process of generating a transaction resource analysis model by training a predetermined initial analysis model according to an embodiment;
FIG. 4 is a diagram illustrating a training process for a transactional resource analysis model, according to one embodiment;
FIG. 5 is a schematic flow chart illustrating a method for machine learning-based transaction resource analysis in accordance with another embodiment;
FIG. 6 is a block diagram of an embodiment of a machine learning based transaction resource analysis device;
FIG. 7 is a block diagram of another embodiment of a machine learning based transaction resource analysis apparatus;
FIG. 8 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 transaction resource analysis method based on machine learning provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 and the server 104 communicate via a network. The server 104 receives the resource analysis request sent by the terminal 102, and obtains the market transaction service to be processed and the target special terms carried in the resource analysis request. The server 104 trains a preset initial analysis model to obtain a transaction resource analysis model by obtaining a modeling factor corresponding to the market transaction service according to the target special terms and corresponding initial transaction resources, further performs valuation analysis on the market transaction service to be processed based on the transaction resource analysis model to generate a corresponding special term transaction resource analysis result, further generates corresponding service recommendation information according to the special term transaction resource analysis result, and sends the service recommendation information to the terminal 102. 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, there is provided a transaction resource analysis method based on machine learning, which is illustrated by applying the method to the server in fig. 1, and includes the following steps:
step S202, receiving the resource analysis request, and acquiring the market transaction service to be processed and the target special clause carried by the resource analysis request.
Specifically, a resource analysis request sent by a user based on a terminal is received, and the resource analysis request is analyzed, so as to obtain a to-be-processed market transaction task carried by the resource analysis request, where the to-be-processed market transaction service may be market bonds of different bond types, where the market bonds of different bond types may include: corporate debt, mid-term bills, short-cut, ultra-short cut, and orientation tools, among others.
After receiving the resource analysis request and acquiring the market transaction task to be processed carried by the resource analysis request, further acquiring preset target special terms. The market transaction service to be processed can be market bonds of different bond types, and the target special terms can be special terms corresponding to the market bonds of different bond types, such as non-public, guarantee, perpetual and the like.
And step S204, obtaining a modeling factor corresponding to the market trading service according with the target special terms and corresponding initial trading resources, and training a preset initial analysis model to obtain a trading resource analysis model.
Specifically, the preset initial analysis model is trained according to the modeling factor of the market trading service meeting the target special term and the corresponding initial trading resource, so that the trained trading resource analysis model can be obtained.
Further, the corresponding modeling factor can be generated by acquiring a first element factor corresponding to the market trading service conforming to the target special term and a second element factor corresponding to the target special term, and performing element coding processing based on the first element factor and the second element factor. And training a preset initial analysis model by acquiring initial trading resources corresponding to the market trading service according with the target special terms and further based on each modeling factor and the initial trading resources to generate a trained trading resource analysis model.
The preset initial analysis model may be an xgboost model, the whole sample is divided into a corresponding training set and a corresponding testing set by a layered random sampling manner, for example, 70% of the whole sample may be taken as the training set and 30% as the testing set, the model is trained according to the training set, the selected modeling factors are ranked according to the contribution degree, and the modeling factors of the preset number are selected for model training and parameter adjustment, so as to ensure the stability of the model. The overall sample represents the overall market bond conforming to the special terms, and can be classified according to the special terms, such as non-public, guarantee and perpetual terms.
And step S206, based on the trained trading resource analysis model, carrying out valuation analysis on the market trading service to be processed, and generating a corresponding special clause trading resource analysis result.
Specifically, the target special terms corresponding to the market transaction business to be processed are obtained, the first element factors of the market transaction business to be processed and the second element factors corresponding to the target special terms are further obtained, and the first element factors of the market transaction business to be processed and the second element factors corresponding to the target special terms are used as input data of a trained transaction resource analysis model. Inputting a first element factor of the market transaction service to be processed and a second element factor corresponding to the target special term into a trained transaction resource analysis model, and performing valuation analysis through the trained transaction resource analysis model to generate a corresponding special term transaction resource analysis result.
In this embodiment, the market transaction business to be processed may be market bonds of different bond types, and the special term transaction resource analysis result may be a special term bond interest analysis result. Specifically, evaluation analysis can be performed on market bonds to be processed through a trained trading resource analysis model, and bond profit and difference analysis results of special terms corresponding to the market bonds are generated.
The first factor of the market transaction service to be processed may include basic factors and market factors, the basic factors include but are not limited to terms, issue size, variety, payment frequency, enterprise nature, and transaction location, and the market factors corresponding to the market transaction service meeting the target special terms include money market, equity market, primary issue data, and secondary deal situation.
Further, the special terms may include non-public, security and permanent, while the different types of bonds include different special terms, and because the market bonds of different types of bonds include different special terms, the market bonds including the corresponding target special terms may be determined according to the target special terms for which the interest analysis is required.
And step S208, generating corresponding service recommendation information according to the special clause transaction resource analysis result.
Specifically, according to the analysis result of the special transaction resources, the analysis result of the special term bond interest difference of market transaction services containing the target special term, namely market bonds of different bond types containing the target special term is obtained, and further the expected income situation of the market bonds containing the target special term can be determined. And generating recommendation information of the market bonds of the corresponding bond types according to the expected income condition of the market bonds containing the target special terms, wherein the recommendation information comprises a service recommendation result corresponding to the expected income condition reaching a preset income threshold value and a service elimination result corresponding to the expected income condition not reaching the preset income threshold value.
The service recommendation information is sent to the terminal where the user triggering the resource analysis request is located, so that the user can determine whether to buy or invest market bonds of corresponding bond types according to the service recommendation information, and can select the market bond types more meeting the requirements to reduce the asset loss.
In the transaction resource analysis method based on machine learning, a resource analysis request is received, a market transaction service to be processed and a target special term carried by the resource analysis request are obtained, a preset initial analysis model is trained to obtain a transaction resource analysis model by obtaining a modeling factor corresponding to the market transaction service according with the target special term and a corresponding initial transaction resource, the market transaction service to be processed is evaluated and analyzed based on the transaction resource analysis model to generate a corresponding special term transaction resource analysis result, and then corresponding service recommendation information is generated according to the special term transaction resource analysis result. Due to the fact that proper reference bonds or term interest difference adjustment and other point data do not need to be found manually, good performance of the machine learning model is combined, transaction resource analysis results of special terms can be obtained through automatic calculation according to the trained transaction resource analysis model, accuracy of the corresponding analysis results is improved, corresponding business recommendation information is further generated according to the resource analysis results with high accuracy, accurate recommendation business is provided for investment users, asset safety of the investment users is guaranteed, and stability of the internet financial market is maintained.
In an embodiment, as shown in fig. 3, the step of training a preset initial analysis model to obtain a transaction resource analysis model specifically includes:
step S302, a first element factor corresponding to the market transaction service according with the target special clause and a second element factor corresponding to the target special clause are obtained.
Specifically, in this embodiment, the market transaction service to be processed may be market bonds of different bond types, and the market bonds of different bond types may include: corporate debt, medium term instruments, short melt, ultra short melt, and orientation tools, etc., while special terms may include different term types, non-public, guaranteed, and permanent, etc.
The first factor of the market transaction service to be processed may include basic factors and market factors, the basic factors include but are not limited to terms, issue size, variety, payment frequency, enterprise nature, and transaction location, and the market factors corresponding to the market transaction service meeting the target special terms include money market, equity market, primary issue data, and secondary deal situation.
Furthermore, for three different special terms, namely non-public, guarantee and perpetual, respective unique factors need to be added, for example, the information of the guarantee person in the guarantee term, the guarantee mode and the like, and the right period, the interest rate adjustment mechanism and the like in the perpetual term all belong to second element factors corresponding to the target special term.
Step S304, based on the first element factor and the second element factor, element coding processing is carried out to generate a corresponding modeling factor.
Specifically, continuous variable factors and classified variable factors are determined from the first element factors and the second element factors, element coding processing is performed on the basis of the classified variable factors, each classified variable factor is converted into a numerical variable factor, normalization processing is performed on the basis of the continuous variable factors, corresponding discrete data factors are generated, normalization processing is performed on the basis of the numerical variable factors and the discrete data factors, and corresponding modeling factors can be generated.
The method comprises the steps of carrying out normalization processing on continuous variable factors, converting the continuous variable factors with large data span into discrete data by using barrel processing, and carrying out processing on abnormal values, such as filling null values of a data set with a median.
Similarly, the element encoding process is performed on the categorical variable factors, for example, one-hot encoding process is performed on the categorical variable factors, and each categorical variable factor is converted into a numerical variable factor.
Further, the processed data are normalized, dimensional data are converted into dimensionless data, and the influence of different dimensions on a preset initial analysis model is eliminated. Namely, normalization processing is carried out based on the numerical variable factor and the discrete data factor, dimensional data in the numerical variable factor and the discrete data factor are converted into dimensionless data, and a corresponding modeling factor can be further generated.
In one embodiment, the one-hot encoding process is used to convert categorical variable factors into numerical variable factors, as exemplified by the industry of enterprises: if the company is in the real estate industry, the fact that the company is in the real estate industry < - >1 and the fact that the company is not in the real estate industry < - >0 can be correspondingly generated, and the obtained trained transaction resource analysis model can identify more element factors by converting non-digital variable factors such as classification variable factors into numerical variable factors.
Step S306, obtaining the initial trading resource corresponding to the market trading service according with the target special terms.
Specifically, the market trading service meeting the target special terms is extracted, the corresponding sample trading service is determined according to the market trading service meeting the target special terms, and the special term trading resource profitability of the market trading service meeting the target special terms and the non-term trading resource profitability corresponding to the sample trading service are obtained. And further determining the initial trading resources corresponding to the market trading service according to the special clause trading resource yield and the non-clause trading resource yield.
In this embodiment, the market transaction service may be market bonds of different bond types, the sample transaction service may be a sample bond, the market bonds conforming to the target special term are extracted, the publicly issued bonds having the same principal persistence are found according to the market bonds conforming to the target special term, and the publicly issued bonds having the same principal persistence are determined as corresponding sample bonds, that is, the sample transaction service is determined as corresponding sample bonds.
Further, in this embodiment, the special term transaction resource profitability may be a special term bond profitability for obtaining market bonds complying with the target special term, and the non-term transaction resource profitability may be a non-term bond profitability corresponding to the sample bonds.
Specifically, the initial transaction resource corresponding to the market transaction service according to the target special terms is determined according to the special term transaction resource profitability and the non-term transaction resource profitability, namely, the earning rate of the special term bond of the market bond according to the target special terms can be obtained, the earning rate of the special term bond of the market bond according to the target special terms and the earning rate of the non-term bond corresponding to the sample bond are calculated, the difference between the earning rate of the special term bond and the earning rate of the non-term bond corresponding to the sample bond is obtained through calculation, the earning rate of the special term bond of the market bond according to the target special terms is obtained, and the initial transaction resource corresponding to the market transaction service according to the target special terms is obtained.
In one embodiment, a manner of generating a non-clause trading resource profitability corresponding to a sample trading business comprises:
obtaining a yield curve of the same grade and the same industry and the yield of the sample transaction service;
adjusting the rate of return of the sample transaction service according to the rate of return curve of the same grade and the same industry, and generating an average value of the adjusted rate of return of the sample transaction service;
and determining the adjusted average value of the profitability of the sample transaction service as the profitability of the non-clause transaction resource corresponding to the sample transaction service.
Specifically, by acquiring a yield curve of the same industry with the same rating and the yield of the sample transaction service, on the yield curve of the priced bonds, because the sample bonds and the priced bonds have different residual time periods, a certain difference exists, and the yield of the sample transaction service can be adjusted according to the yield curve of the same industry with the same rating when the sample bonds and the priced bonds are finally priced.
Further, the profitability of the sample transaction business is adjusted according to the profitability curve of the same grade and the same industry, the profitability average value of the adjusted sample transaction business can be obtained, and the adjusted profitability average value of the sample transaction business is determined as the non-clause transaction resource profitability corresponding to the sample transaction business.
In one embodiment, extracting market trading traffic that meets the target special terms and determining corresponding sample trading traffic based on the market trading traffic that meets the target special terms includes:
extracting market transaction services which accord with target special terms;
according to the market transaction service which accords with the target special terms, the open issue transaction service which has the same main body continuity is searched;
and determining the existing public development business with the same main body as the corresponding sample transaction business.
Specifically, in this embodiment, the market transaction business is market bonds of different bond types, and the public development company business may be publicly issued bonds. The finding of the publicly-issued transaction service having the same principal persistence as the market transaction service conforming to the target special term can be understood as that the publicly-issued bonds having the same principal persistence are found by extracting the market bonds conforming to the target special term and according to the market bonds conforming to the target special term, and then the publicly-issued bonds having the same principal persistence are determined as corresponding sample bonds.
And S308, training a preset initial analysis model based on each modeling factor and the initial transaction resource to generate a trained transaction resource analysis model.
Specifically, each modeling factor is used as input, the initial transaction resource is used as output, and a preset initial analysis model is trained to obtain a trained transaction resource analysis model.
In this embodiment, the initial transaction resource may be the special term bond interest difference of the market bond of the target special term, the modeling factor may include, but is not limited to, term, issue size, variety, payment frequency, enterprise property, transaction place, money market, equity market, primary issue data, secondary transaction situation, etc., and may further include the information of the middle guarantor of the guarantee term, guarantee mode, etc., the right period in the persistent term, interest rate adjustment mechanism, etc.
In an embodiment, the preset initial analysis model may be an xgboost model, the whole samples are divided into corresponding training sets and test sets by means of hierarchical random sampling, for example, 70% of the whole samples may be taken as the training sets and 30% as the test sets, the model is trained according to the training sets, the selected modeling factors are ranked according to the contribution degree, and the modeling factors of the preset number are selected for model training and parameter adjustment, so as to ensure the stability of the model. For example,
the overall sample represents the overall market bond conforming to the special terms, and can be classified according to the special terms, such as non-public, guarantee and perpetual terms.
Wherein, the parameters of the model adjustment mainly comprise: (1) n _ estimators, the total number of iterations, i.e. the number of decision trees, and the parameter adjusting alternative range [10,20,30,40,50,100,120 ]; (2) max _ depth, the larger the value of the decision tree is, the more complex the tree is, and the candidate range of the parameter is adjusted [2,3,4,5,6 ]; (3) and gamma is a penalty term coefficient, a minimum loss function reduction value required by node splitting is specified, and a parameter adjusting alternative range is [0.001,0.01,0.1,0.2 ].
In one embodiment, the specific process of K-fold cross validation includes:
dividing the training set into K parts, taking one part as a verification set each time, taking the rest K-1 parts as a test set, and repeating the cross verification for K times. And performing one round of cross validation on each parameter to be optimized, and obtaining an average score by a cross validation scoring method, wherein the parameter with the highest score is the value of the optimal parameter, and obtaining all the optimal parameters. Wherein cross-validation may reduce overfitting to some extent.
As shown in fig. 4, referring to fig. 4, the process of classifying the whole samples and training the preset initial analysis model, that is, the training process of the trading resource analysis model, as can be seen from fig. 4, the whole samples are classified into the sample sets of the preset shares, for example, the whole samples D are divided into 10 sample sets including D1 to D10, further, all of D1 to D9 are used as training sets, and D10 is used as a test set, and cross validation is performed to obtain the corresponding first test result.
Similarly, D1 through D8, and D10 may be used as training sets, D9 may be used as test set, and cross-validation may be performed to obtain a second corresponding test result, and so on, until D2 through D10 are used as training sets, and D1 is used as test set, and a tenth corresponding test result is obtained.
Further, an average test result is obtained by averaging the first test result, the second test result, … …, and the tenth test result, and the test result with the largest score value higher than the average test result is determined from the first test result to the tenth test result based on the average test result. And determining the parameter value corresponding to the test result with the maximum score value higher than the average test result as the value of the optimal parameter so as to finish the parameter adjusting process of the model.
In one embodiment, the transaction resource analysis model can identify all models by specially processing the input factors into a format meeting the requirements of the models, namely, the models are all suitable numbers, so that the universality of the models is greatly improved, and for priced bonds, the related factors are input and can be directly applied to obtain an analysis result.
The model obtains a relatively ideal result through training of a large number of samples and factors thereof, when the samples are few, the model effect can be weakened, training can be performed through increasing the number of the factors, certain improvement is performed, and a better model analysis effect is achieved. The trained trading resource model scans most credit bonds through the market, covers various conditions of three special terms, and can obtain corresponding interest difference according to factor analysis of the special term bonds even if the special term interest difference is calculated without proper reference sample bonds under the condition that the special term bonds are known.
In this embodiment, the element coding processing is performed to generate the corresponding modeling factor based on the first element factor and the second element factor by obtaining the first element factor corresponding to the market transaction service conforming to the target special term and the second element factor corresponding to the target special term. And training a preset initial analysis model by acquiring initial trading resources corresponding to the market trading service according with the target special terms and further based on each modeling factor and the initial trading resources to generate a trained trading resource analysis model. Due to the fact that proper reference bonds or term interest difference adjustment and other point data do not need to be found manually, the first element factors and the second element factors are directly combined with the good performance of the machine learning model, the preset initial analysis model is trained to generate a trained transaction resource analysis model, the trained transaction resource analysis model can be used for automatically calculating transaction resource analysis results of special terms, and the accuracy of corresponding analysis results is improved.
In one embodiment, as shown in fig. 5, a transaction resource analysis method based on machine learning is provided, which specifically includes:
1) and acquiring a first element factor corresponding to the market transaction service according with the target special clause and a second element factor corresponding to the target special clause.
2) From the first factor and the second factor, a continuous variable factor and a categorical variable factor are determined.
3) And performing element coding processing based on the categorical variable factors, and converting each categorical variable factor into a numerical variable factor.
4) And carrying out standardization processing based on the continuous variable factors to generate corresponding discrete data factors.
5) And carrying out normalization processing based on the numerical variable factor and the discrete data factor to generate a corresponding modeling factor.
6) And extracting the market transaction service according with the target special terms, and searching the open issue transaction service which has the same main body and is persistent according to the market transaction service according with the target special terms.
7) And determining the existing public development business with the same main body as the corresponding sample transaction business.
8) And acquiring the special clause transaction resource profitability of the market transaction service of the target special clause, the same-grade same-industry profitability curve and the profitability of the sample transaction service.
9) And adjusting the rate of return of the sample transaction service according to the rate of return curve of the same grade and the same industry to generate an adjusted average value of the rate of return of the sample transaction service.
10) And determining the adjusted average value of the profitability of the sample transaction service as the profitability of the non-clause transaction resource corresponding to the sample transaction service.
11) And determining the initial trading resources corresponding to the market trading service according with the target special terms according to the special term trading resource profitability and the non-term trading resource profitability.
12) And training a preset initial analysis model based on each modeling factor and the initial transaction resource to generate a trained transaction resource analysis model.
13) And receiving a resource analysis request, and acquiring the market transaction service to be processed and the target special clause carried by the resource analysis request.
14) And based on the trading resource analysis model, carrying out valuation analysis on the market trading service to be processed to generate a corresponding special clause trading resource analysis result.
15) And generating corresponding service recommendation information according to the special clause transaction resource analysis result.
In the machine-learned transaction resource analysis method, the resource analysis request is received, the market transaction service to be processed and the target special terms carried by the resource analysis request are obtained, the preset initial analysis model is trained to obtain the transaction resource analysis model by obtaining the modeling factor corresponding to the market transaction service according with the target special terms and the corresponding initial transaction resources, the market transaction service to be processed is evaluated and analyzed based on the transaction resource analysis model to generate the corresponding special term transaction resource analysis result, and then the corresponding service recommendation information is generated according to the special term transaction resource analysis result. Due to the fact that proper reference bonds or term interest difference adjustment and other point data do not need to be found manually, good performance of the machine learning model is combined, transaction resource analysis results of special terms can be obtained through automatic calculation according to the trained transaction resource analysis model, accuracy of the corresponding analysis results is improved, corresponding business recommendation information is further generated according to the resource analysis results with high accuracy, accurate recommendation business is provided for investment users, asset safety of the investment users is guaranteed, and stability of the internet financial market is maintained.
It should be understood that, although the steps in the flowcharts related to the above embodiments are shown in sequence as indicated by the arrows, the steps are not necessarily executed in sequence 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 a part of the steps in each flowchart related to the above embodiments may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.
In one embodiment, as shown in fig. 6, there is provided a machine learning-based transaction resource analysis apparatus, including: a pending market transaction service obtaining module 602, a transaction resource analysis model obtaining module 604, a special clause transaction resource analysis result generating module 606, and a service recommendation information generating module 608, wherein:
the to-be-processed market trading service obtaining module 602 is configured to receive the resource analysis request, and obtain the to-be-processed market trading service and the target special term carried in the resource analysis request.
The trading resource analysis model obtaining module 604 is configured to obtain a trading resource analysis model obtained by training a preset initial analysis model according to a modeling factor corresponding to a market trading service meeting a target special term and a corresponding initial trading resource.
The special clause transaction resource analysis result generation module 606 is configured to perform valuation analysis on the market transaction service to be processed based on the trained transaction resource analysis model, and generate a corresponding special clause transaction resource analysis result.
The service recommendation information generating module 608 is configured to generate corresponding service recommendation information according to the special clause transaction resource analysis result.
In the machine-learned trading resource analysis device, the resource analysis request is received, the market trading service to be processed and the target special terms carried by the resource analysis request are obtained, the preset initial analysis model is trained to obtain the trading resource analysis model by obtaining the modeling factor corresponding to the market trading service according with the target special terms and the corresponding initial trading resources, the market trading service to be processed is evaluated and analyzed based on the trading resource analysis model to generate the corresponding special term trading resource analysis result, and then the corresponding service recommendation information is generated according to the special term trading resource analysis result. Due to the fact that proper reference bonds or term interest difference adjustment and other point data do not need to be found manually, good performance of the machine learning model is combined, transaction resource analysis results of special terms can be obtained through automatic calculation according to the trained transaction resource analysis model, accuracy of the corresponding analysis results is improved, corresponding business recommendation information is further generated according to the resource analysis results with high accuracy, accurate recommendation business is provided for investment users, asset safety of the investment users is guaranteed, and stability of the internet financial market is maintained.
In one embodiment, as shown in fig. 7, there is provided a transaction resource analysis device based on machine learning, further comprising: an element factor obtaining module 702, a modeling factor generating module 704, an initial transaction resource obtaining module 706, and a transaction resource analysis model generating module 708, wherein:
the factor acquiring module 702 is configured to acquire a first factor corresponding to the market transaction service meeting the target special term, and a second factor corresponding to the target special term.
And a modeling factor generation module 704, configured to perform element coding processing based on the first element factor and the second element factor to generate a corresponding modeling factor.
An initial trading resource obtaining module 706, configured to obtain an initial trading resource corresponding to the market trading service that meets the target special term.
The transaction resource analysis model generation module 708 is configured to train a preset initial analysis model based on each modeling factor and the initial transaction resource, and generate a trained transaction resource analysis model.
In the transaction resource analysis device based on machine learning, by acquiring the first element factor corresponding to the market transaction service conforming to the target special term and the second element factor corresponding to the target special term, element coding processing is performed based on the first element factor and the second element factor, and a corresponding modeling factor is generated. And training a preset initial analysis model by acquiring initial trading resources corresponding to the market trading service according with the target special terms and further based on each modeling factor and the initial trading resources to generate a trained trading resource analysis model. Due to the fact that proper reference bonds or term interest difference adjustment and other point data do not need to be found manually, the first element factors and the second element factors are directly combined with the good performance of the machine learning model, the preset initial analysis model is trained to generate a trained transaction resource analysis model, the trained transaction resource analysis model can be used for automatically calculating transaction resource analysis results of special terms, and the accuracy of corresponding analysis results is improved.
In one embodiment, the initial transaction resource acquisition module is further configured to:
extracting market transaction services conforming to the target special terms, and determining corresponding sample transaction services according to the market transaction services conforming to the target special terms; acquiring special clause transaction resource profitability of the market transaction service of the target special clause and non-clause transaction resource profitability corresponding to the sample transaction service; and determining the initial trading resources corresponding to the market trading service according with the target special terms according to the special term trading resource profitability and the non-term trading resource profitability.
In one embodiment, the modeling factor generation module is further configured to:
determining continuous variable factors and classification variable factors from the first element factors and the second element factors; performing element coding processing based on the categorical variable factors, and converting each categorical variable factor into a numerical variable factor; carrying out standardization processing based on the continuous variable factors to generate corresponding discrete data factors; and carrying out normalization processing based on the numerical variable factor and the discrete data factor to generate a corresponding modeling factor.
In one embodiment, the initial transaction resource acquisition module is further configured to:
obtaining a yield curve of the same grade and the same industry and the yield of the sample transaction service; adjusting the rate of return of the sample transaction service according to the rate of return curve of the same grade and the same industry, and generating an average value of the adjusted rate of return of the sample transaction service; and determining the adjusted average value of the profitability of the sample transaction service as the profitability of the non-clause transaction resource corresponding to the sample transaction service.
In one embodiment, the initial transaction resource acquisition module is further configured to:
extracting market transaction services which accord with target special terms; according to the market transaction service which accords with the target special terms, the open issue transaction service which has the same main body continuity is searched; and determining the existing public development business with the same main body as the corresponding sample transaction business.
For specific limitations of the transaction resource analysis device based on machine learning, reference may be made to the above limitations of the transaction resource analysis method based on machine learning, and details thereof are not repeated here. The modules in the machine learning-based transaction resource analysis device can be wholly or partially implemented 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, and its internal structure diagram may be as shown in fig. 8. 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 equipment is used for storing data such as market transaction business to be processed, a trained transaction resource analysis model, modeling factors, initial transaction resources, transaction resource analysis results of special terms and the like. 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 machine learning based transaction resource analysis method.
Those skilled in the art will appreciate that the architecture shown in fig. 8 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:
receiving a resource analysis request, and acquiring market transaction service to be processed and target special terms carried by the resource analysis request;
obtaining a modeling factor corresponding to market transaction business according with the target special terms and corresponding initial transaction resources, and training a preset initial analysis model to obtain a transaction resource analysis model;
based on the trading resource analysis model, carrying out valuation analysis on the market trading service to be processed to generate a corresponding special clause trading resource analysis result;
and generating corresponding service recommendation information according to the special clause transaction resource analysis result.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
acquiring a first element factor corresponding to the market transaction service according with the target special clause and a second element factor corresponding to the target special clause;
performing element coding processing based on the first element factor and the second element factor to generate a corresponding modeling factor;
acquiring initial trading resources corresponding to market trading services according with target special terms;
and training a preset initial analysis model based on each modeling factor and the initial transaction resource to generate a trained transaction resource analysis model.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
extracting market transaction services conforming to the target special terms, and determining corresponding sample transaction services according to the market transaction services conforming to the target special terms;
acquiring special clause transaction resource profitability of the market transaction service of the target special clause and non-clause transaction resource profitability corresponding to the sample transaction service;
and determining the initial trading resources corresponding to the market trading service according with the target special terms according to the special term trading resource profitability and the non-term trading resource profitability.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
determining continuous variable factors and classification variable factors from the first element factors and the second element factors;
performing element coding processing based on the categorical variable factors, and converting each categorical variable factor into a numerical variable factor;
carrying out standardization processing based on the continuous variable factors to generate corresponding discrete data factors;
and carrying out normalization processing based on the numerical variable factor and the discrete data factor to generate a corresponding modeling factor.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
obtaining a yield curve of the same grade and the same industry and the yield of the sample transaction service;
adjusting the rate of return of the sample transaction service according to the rate of return curve of the same grade and the same industry, and generating an average value of the adjusted rate of return of the sample transaction service;
and determining the adjusted average value of the profitability of the sample transaction service as the profitability of the non-clause transaction resource corresponding to the sample transaction service.
In one embodiment, the processor, when executing the computer program, further performs the steps of:
extracting market transaction services which accord with target special terms;
according to the market transaction service which accords with the target special terms, the open issue transaction service which has the same main body continuity is searched;
and determining the existing public development business with the same main body as the corresponding sample transaction business.
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:
receiving a resource analysis request, and acquiring market transaction service to be processed and target special terms carried by the resource analysis request;
obtaining a modeling factor corresponding to market transaction business according with the target special terms and corresponding initial transaction resources, and training a preset initial analysis model to obtain a transaction resource analysis model;
based on the trading resource analysis model, carrying out valuation analysis on the market trading service to be processed to generate a corresponding special clause trading resource analysis result;
and generating corresponding service recommendation information according to the special clause transaction resource analysis result.
In one embodiment, the computer program when executed by the processor further performs the steps of:
acquiring a first element factor corresponding to the market transaction service according with the target special clause and a second element factor corresponding to the target special clause;
performing element coding processing based on the first element factor and the second element factor to generate a corresponding modeling factor;
acquiring initial trading resources corresponding to market trading services according with target special terms;
and training a preset initial analysis model based on each modeling factor and the initial transaction resource to generate a trained transaction resource analysis model.
In one embodiment, the computer program when executed by the processor further performs the steps of:
extracting market transaction services conforming to the target special terms, and determining corresponding sample transaction services according to the market transaction services conforming to the target special terms;
acquiring special clause transaction resource profitability of the market transaction service of the target special clause and non-clause transaction resource profitability corresponding to the sample transaction service;
and determining the initial trading resources corresponding to the market trading service according with the target special terms according to the special term trading resource profitability and the non-term trading resource profitability.
In one embodiment, the computer program when executed by the processor further performs the steps of:
determining continuous variable factors and classification variable factors from the first element factors and the second element factors;
performing element coding processing based on the categorical variable factors, and converting each categorical variable factor into a numerical variable factor;
carrying out standardization processing based on the continuous variable factors to generate corresponding discrete data factors;
and carrying out normalization processing based on the numerical variable factor and the discrete data factor to generate a corresponding modeling factor.
In one embodiment, the computer program when executed by the processor further performs the steps of:
obtaining a yield curve of the same grade and the same industry and the yield of the sample transaction service;
adjusting the rate of return of the sample transaction service according to the rate of return curve of the same grade and the same industry, and generating an average value of the adjusted rate of return of the sample transaction service;
and determining the adjusted average value of the profitability of the sample transaction service as the profitability of the non-clause transaction resource corresponding to the sample transaction service.
In one embodiment, the computer program when executed by the processor further performs the steps of:
extracting market transaction services which accord with target special terms;
according to the market transaction service which accords with the target special terms, the open issue transaction service which has the same main body continuity is searched;
and determining the existing public development business with the same main body as the corresponding sample transaction business.
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 related to 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 examples 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 method for machine learning-based transaction resource analysis, the method comprising:
receiving a resource analysis request, and acquiring market transaction service to be processed and target special terms carried by the resource analysis request;
obtaining a modeling factor corresponding to market transaction business according with the target special terms and corresponding initial transaction resources, and training a preset initial analysis model to obtain a transaction resource analysis model;
based on the trading resource analysis model, carrying out valuation analysis on the market trading service to be processed to generate a corresponding special clause trading resource analysis result;
and generating corresponding service recommendation information according to the special clause transaction resource analysis result.
2. The method of claim 1, wherein training the predetermined initial analysis model to obtain the transaction resource analysis model comprises:
acquiring a first element factor corresponding to a market transaction service according with a target special term and a second element factor corresponding to the target special term;
performing element coding processing based on the first element factor and the second element factor to generate a corresponding modeling factor;
acquiring initial trading resources corresponding to market trading services according with target special terms;
and training the preset initial analysis model based on each modeling factor and the initial transaction resource to generate a trained transaction resource analysis model.
3. The method of claim 2, wherein obtaining the initial trading resource corresponding to the market trading service meeting the target special term comprises:
extracting market transaction services conforming to the target special terms, and determining corresponding sample transaction services according to the market transaction services conforming to the target special terms;
acquiring special clause transaction resource profitability of the market transaction service of the target special clause and non-clause transaction resource profitability corresponding to the sample transaction service;
and determining initial trading resources corresponding to the market trading service according with the target special terms according to the special term trading resource profitability and the non-term trading resource profitability.
4. The method of claim 2, wherein performing element encoding processing based on the first and second element factors to generate corresponding modeling factors comprises:
determining continuous variable factors and classification variable factors from the first element factors and the second element factors;
performing element coding processing based on the classification variable factors, and converting each classification variable factor into a numerical variable factor;
carrying out standardization processing based on the continuous variable factors to generate corresponding discrete data factors;
and carrying out normalization processing based on the numerical variable factor and the discrete data factor to generate a corresponding modeling factor.
5. The method of claim 3, wherein generating a non-clause trading resource profitability corresponding to the sample trading service comprises:
obtaining a yield curve of the same grade and the same industry and the yield of the sample transaction service;
adjusting the rate of return of the sample transaction service according to the rate of return curve of the same grade and the same industry, and generating an adjusted average value of the rate of return of the sample transaction service;
and determining the adjusted average value of the profitability of the sample transaction service as the profitability of the non-clause transaction resource corresponding to the sample transaction service.
6. The method of claim 3, wherein the extracting market transaction services that meet target specific terms and determining corresponding sample transaction services based on the market transaction services that meet target specific terms comprises:
extracting market transaction services which accord with target special terms;
according to the market transaction service meeting the target special terms, searching the open issue transaction service with the same main body continuity;
and determining the business of the public development bank with the same main body as the corresponding sample transaction business.
7. A machine learning based transaction resource analysis apparatus, the apparatus comprising:
the system comprises a to-be-processed market transaction service acquisition module, a resource analysis module and a target special term acquisition module, wherein the to-be-processed market transaction service acquisition module is used for receiving a resource analysis request and acquiring the to-be-processed market transaction service and the target special term carried by the resource analysis request;
the trading resource analysis model acquisition module is used for acquiring a trading resource analysis model obtained by training a preset initial analysis model according to a modeling factor corresponding to the market trading service meeting the target special term and the corresponding initial trading resource;
the special clause transaction resource analysis result generation module is used for carrying out valuation analysis on the market transaction service to be processed based on the trained transaction resource analysis model and generating a corresponding special clause transaction resource analysis result;
and the service recommendation information generation module is used for generating corresponding service recommendation information according to the special clause transaction resource analysis result.
8. The apparatus of claim 7, further comprising:
the element factor acquisition module is used for acquiring a first element factor corresponding to the market transaction service meeting target special terms and a second element factor corresponding to the target special terms;
the modeling factor generation module is used for carrying out element coding processing on the basis of the first element factor and the second element factor to generate a corresponding modeling factor;
the initial trading resource acquisition module is used for acquiring initial trading resources corresponding to the market trading service according with the target special terms;
and the transaction resource analysis model generation module is used for training the preset initial analysis model based on each modeling factor and the initial transaction resource to generate a trained transaction resource analysis 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 6.
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 6.
CN202110732054.7A 2021-06-29 2021-06-29 Transaction resource analysis method, device, equipment and medium based on machine learning Pending CN113362115A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110732054.7A CN113362115A (en) 2021-06-29 2021-06-29 Transaction resource analysis method, device, equipment and medium based on machine learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110732054.7A CN113362115A (en) 2021-06-29 2021-06-29 Transaction resource analysis method, device, equipment and medium based on machine learning

Publications (1)

Publication Number Publication Date
CN113362115A true CN113362115A (en) 2021-09-07

Family

ID=77537311

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110732054.7A Pending CN113362115A (en) 2021-06-29 2021-06-29 Transaction resource analysis method, device, equipment and medium based on machine learning

Country Status (1)

Country Link
CN (1) CN113362115A (en)

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108549996A (en) * 2018-04-26 2018-09-18 上海普兰金融服务有限公司 A kind of bond analysis on Achievements method
CN109598606A (en) * 2018-12-04 2019-04-09 泰康保险集团股份有限公司 Data processing method, device, medium and electronic equipment
CN109886809A (en) * 2019-01-24 2019-06-14 平安科技（深圳）有限公司 Transaction data intelligent analysis method, electronic device and computer readable storage medium
CN109885695A (en) * 2019-01-28 2019-06-14 平安科技（深圳）有限公司 Assets suggest generation method, device, computer equipment and storage medium
CN109961368A (en) * 2019-03-18 2019-07-02 京东数字科技控股有限公司 Data processing method and device based on machine learning
CN111105311A (en) * 2019-12-19 2020-05-05 第一创业证券股份有限公司 Dynamic credit rating method and device for bond body
CN111161080A (en) * 2019-12-10 2020-05-15 中国建设银行股份有限公司 Information processing method and device
CN111340236A (en) * 2020-03-03 2020-06-26 中债金融估值中心有限公司 Bond default prediction method based on bond valuation data and integrated machine learning
CN111383091A (en) * 2020-02-21 2020-07-07 中国建设银行股份有限公司 Asset securitization pricing method and device
CN112991043A (en) * 2021-03-02 2021-06-18 北京同邦卓益科技有限公司 Debt rating data processing method and device, electronic equipment and storage medium

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108549996A (en) * 2018-04-26 2018-09-18 上海普兰金融服务有限公司 A kind of bond analysis on Achievements method
CN109598606A (en) * 2018-12-04 2019-04-09 泰康保险集团股份有限公司 Data processing method, device, medium and electronic equipment
CN109886809A (en) * 2019-01-24 2019-06-14 平安科技（深圳）有限公司 Transaction data intelligent analysis method, electronic device and computer readable storage medium
CN109885695A (en) * 2019-01-28 2019-06-14 平安科技（深圳）有限公司 Assets suggest generation method, device, computer equipment and storage medium
CN109961368A (en) * 2019-03-18 2019-07-02 京东数字科技控股有限公司 Data processing method and device based on machine learning
CN111161080A (en) * 2019-12-10 2020-05-15 中国建设银行股份有限公司 Information processing method and device
CN111105311A (en) * 2019-12-19 2020-05-05 第一创业证券股份有限公司 Dynamic credit rating method and device for bond body
CN111383091A (en) * 2020-02-21 2020-07-07 中国建设银行股份有限公司 Asset securitization pricing method and device
CN111340236A (en) * 2020-03-03 2020-06-26 中债金融估值中心有限公司 Bond default prediction method based on bond valuation data and integrated machine learning
CN112991043A (en) * 2021-03-02 2021-06-18 北京同邦卓益科技有限公司 Debt rating data processing method and device, electronic equipment and storage medium

Similar Documents

Publication Publication Date Title
CN109165840B (en) Risk prediction processing method, risk prediction processing device, computer equipment and medium
CN108876133B (en) Risk assessment processing method, device, server and medium based on business information
CN111369342A (en) Loan approval method, device, equipment and storage medium based on machine learning
US20220253856A1 (en) System and method for machine learning based detection of fraud
CN109063921B (en) Optimization processing method and device for client risk early warning, computer equipment and medium
CN111160737A (en) Adaptation method of resource allocation scheme and related equipment
CN114782201A (en) Stock recommendation method and device, computer equipment and storage medium
CN112950347B (en) Resource data processing optimization method and device, storage medium and terminal
CN116681519A (en) Breach prediction method, breach prediction apparatus, electronic device, medium, and program product
CN116821759A (en) Identification prediction method and device for category labels, processor and electronic equipment
CN115759283A (en) Model interpretation method and device, electronic equipment and storage medium
CN113362115A (en) Transaction resource analysis method, device, equipment and medium based on machine learning
Rösch et al. Estimating credit contagion in a standard factor model
CN114881761A (en) Determination method of similar sample and determination method of credit limit
CN114626940A (en) Data analysis method and device and electronic equipment
CN115860889A (en) Financial loan big data management method and system based on artificial intelligence
CN114170000A (en) Credit card user risk category identification method, device, computer equipment and medium
CN113627997A (en) Data processing method and device, electronic equipment and storage medium
CN113222767A (en) Data processing method and device for indexing securities combination
CN111429257A (en) Transaction monitoring method and device
Sabkha et al. Forecasting sovereign CDS volatility: A comparison of univariate GARCH-class models
CN113313587A (en) Credit risk analysis method, device, equipment and medium based on artificial intelligence
CN117994017A (en) Method for constructing retail credit risk prediction model and online credit service Scoredelta model
CN116862645A (en) Credit risk assessment method, device, equipment and medium
CN114565460A (en) Information pushing method based on delayed conversion prediction model and related equipment

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