CN114037516A - Model feature extraction method, device, equipment and medium for credit - Google Patents
Model feature extraction method, device, equipment and medium for credit Download PDFInfo
- Publication number
- CN114037516A CN114037516A CN202111337002.6A CN202111337002A CN114037516A CN 114037516 A CN114037516 A CN 114037516A CN 202111337002 A CN202111337002 A CN 202111337002A CN 114037516 A CN114037516 A CN 114037516A
- Authority
- CN
- China
- Prior art keywords
- merchant
- transaction
- data
- target user
- credit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- 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
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
- G06Q10/063114—Status monitoring or status determination for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The embodiment of the invention discloses a method, a device, equipment and a medium for extracting model features for credit. The method comprises the following steps: acquiring transaction behavior sequence data of a merchant; converting the transaction behavior sequence data into merchant feature vectors based on a set item vector conversion model; and constructing a user feature vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant feature vector corresponding to the merchant. The embodiment of the invention calculates the feature vector of the user through the feature vector of the merchant, thereby depicting the implicit transaction features of the user, improving the recognition level of the credit risk of the user, and improving the user quota improvement and service interception level in credit.
Description
Technical Field
The embodiment of the invention relates to the technical field of financial security, in particular to a method, a device, equipment and a medium for extracting model features for credit.
Background
The development of internet technology also brings new development opportunities for traditional finance. In the existing financial field, with the development of social consumption level and the continuous update of technology, various credit scenes and credit models are increasingly popularized, and the problem therewith is how to better evaluate the overdue risk of a user so as to avoid economic loss.
The characteristics of the current model for credit and central transaction are mostly quantitative statistics of user data by analyzing the latest 1-time consumption time interval (Recency), consumption Frequency (Frequency) and consumption amount (money) of a user, the behavior of the user in credit is depicted, and the model for credit and central transaction can be used as the dominant characteristic of the user to evaluate overdue risk in credit of the user.
However, the feature extraction method based on the classification of the merchants only describes the dominant similarity of the merchants and cannot capture the recessive similarity of the merchants, so that the overdue risk recognition level of users in credit is not high, and the normal development of credit business is difficult to support effectively.
Disclosure of Invention
The invention provides a model feature extraction method, a device, equipment and a medium for credit, which are used for effectively depicting hidden transaction features of a user and improving the recognition level of credit risks of the user.
In a first aspect, an embodiment of the present invention provides a method for extracting model features for credit and credit, where the method includes:
acquiring transaction behavior sequence data of a merchant;
converting the transaction behavior sequence data into merchant feature vectors based on a set item vector conversion model;
and constructing a user characteristic vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant characteristic vector corresponding to the merchant.
In a second aspect, an embodiment of the present invention further provides an apparatus for extracting model features for credit system, where the apparatus includes:
the transaction data acquisition module is used for acquiring transaction behavior sequence data of a merchant;
the feature vector conversion module is used for converting the transaction behavior sequence data into merchant feature vectors based on a set project vector conversion model;
the characteristic vector construction module is used for constructing a user characteristic vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant characteristic vector corresponding to the merchant.
In a third aspect, an embodiment of the present invention further provides a credit system model feature extraction device, where the device includes:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method for model feature extraction for credit as described in any of the first aspects.
In a fourth aspect, an embodiment of the present invention further provides a storage medium containing computer-executable instructions stored thereon, where the computer-executable instructions, when executed by a computer processor, implement any of the above methods for extracting model features for credit.
According to the invention, the characteristic vector of the user is calculated through the characteristic vector of the merchant, so that the implicit transaction characteristics of the user are described, the recognition level of credit risk of the user is improved, and the user's quota improvement and service interception level in credit are improved.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments made with reference to the following drawings:
fig. 1 is a flowchart of a method for extracting features of a credit model according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for extracting features of a credit system model according to a second embodiment of the present invention
Fig. 3 is a block diagram illustrating a structure of a model feature extraction apparatus for credit system according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a model feature extraction device for credit system according to a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a method for extracting features of a model for credit system according to an embodiment of the present invention, where this embodiment is applicable to a situation where user features need to be extracted from existing transaction data when a credit model is used in a financial service, and this method may be implemented by an apparatus for extracting features of a model for credit system according to an embodiment of the present invention, and the apparatus may be implemented in software and/or hardware. The apparatus may be configured in a corresponding device or server, and the method specifically includes:
s110, acquiring transaction behavior sequence data of a merchant;
the merchant is a participant who provides trading goods or services in trading behaviors, and can be a specific store, such as a hot pot store A; or a certain trading platform, and the merchant in the trading action with the platform is the trading platform, but not a specific shop on the trading platform. For example, a takeaway transaction with a certain takeaway platform hot pot store a, for which the merchant is the takeaway platform, but not the hot pot store a.
Specifically, the specific user may be determined according to the target user who wants to obtain the feature vector, and the representative merchant with a large number of transactions in a period of time may also be selected from the database. After the merchant name is obtained, the name of the merchant and the name of the corresponding transaction participation user are spliced to obtain transaction behavior sequence data of the merchant corresponding to the corresponding merchant represented by the name.
S120, converting the transaction behavior sequence data into merchant feature vectors based on a set item vector conversion model;
wherein the item vector conversion model is set to a preset model for converting the user behavior sequence data into data in the form of a feature vector processed by the existing credit model. The merchant feature vector can be understood as an embedded representation of transaction behavior sequence data in a high-dimensional space.
Specifically, in the embodiment of the invention, an Item vector conversion model can be constructed by using an Item2Vec method, and the sequence data vectorization representation obtained by the Item2Vec method can better extract the similarity characteristic between data, so that the recessive transaction characteristic of a user is mined.
In this embodiment of the present invention, preferably, converting the transaction behavior sequence data into a merchant feature vector based on a set item vector conversion model includes: processing the transaction behavior sequence data based on a serialization feature extraction model to obtain processed transaction behavior sequence data; and inputting the processed transaction behavior sequence data into the set item vector conversion model to obtain a merchant feature vector.
Specifically, because the transaction behavior sequence data is obtained from the database, the name of each data may be complex, and direct input of the model is inefficient. Therefore, in the embodiment of the invention, the transaction behavior sequence data is firstly input into the serialized feature extraction model for processing, and the data input into the setting item vector conversion model is ensured to be the simple character form sequence data. For example, if the merchant user corresponding sequence before input is [ 'Mingming', 'Xiaowang', 'Xiaozhang' ], the transaction behavior sequence after processing is obtained as [ '0001', '0002', '0003', ] and then the simpler transaction behavior sequence after processing is input into the set item vector conversion model, so that the processing pressure of the set item vector conversion model is reduced, and the acquisition speed of the merchant feature vector is improved.
S120, constructing a user feature vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant feature vector corresponding to the merchant.
The target user may be a buyer of the corresponding goods or services as a participant of the transaction.
Specifically, the name of the merchant who transacts with the target user, and the corresponding transaction data such as transaction times, time, frequency and the like can be known according to the transaction behavior data. At this time, according to the transaction information and the feature vector of the corresponding merchant, the user feature vector of the target user calculated on the basis of the merchant vector is constructed.
The embodiment acquires transaction behavior sequence data of a merchant; converting the transaction behavior sequence data into merchant feature vectors based on a set item vector conversion model; according to the transaction behavior data of the target user and at least one merchant and the merchant feature vector corresponding to the merchant, the user feature vector of the target user is constructed, and the feature vector of the user is calculated through the feature vector of the merchant, so that the implicit transaction features of the user are described, the recognition level of the credit risk of the user is improved, and the user quota and the service interception level of the user in credit are improved.
On the basis of the foregoing embodiment, preferably, constructing the user feature vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant feature vector corresponding to the merchant includes:
determining the transaction times of a target user and each merchant according to the transaction behavior data of at least one merchant; multiplying the feature vector of each merchant by the corresponding transaction times to obtain an intermediate feature vector of each merchant; and taking a vector obtained by adding all the merchant intermediate feature vectors as the user feature vector of the target user.
For example, if the target user mingming is taken as an example, if the mingming is consumed twice by the mcdonald merchant, consumed once by the first takeout shop in the united states, and consumed once by more than one shop, the vector of the mingming of the target user is represented as:
and the target user Xiaoming feature vector is 2 × McDanbrou commercial tenant feature vector +1 × Mei-Tuo feature commercial tenant vector +1 × piece-multiple commercial tenant feature vector.
The embodiment of the invention has the advantages that: by means of the mode of further accumulating and summing after the merchant vector is multiplied by the transaction times, on the basis that the user feature vector of the target user is obtained by the merchant feature vector to depict the recessive transaction features of the target user, the efficiency of depicting the recessive transaction features is further improved, transaction information in transaction data is not excessively introduced, and the stability and the effectiveness of the recessive transaction features are guaranteed.
Example two
Fig. 2 is a flowchart of a model feature extraction method for credit and credit systems according to a second embodiment of the present invention, and this embodiment preferably explains the obtained transaction behavior sequence data of the merchant on the basis of the foregoing embodiments. Specifically, referring to fig. 2, the method may include:
s210, acquiring transaction data of the target user within a preset time, and reading a merchant in the transaction data; and taking the transaction data as a data source of transaction behavior sequence data of the merchant.
The preset time can be selected by the system according to the real-time operation load degree, or can be preset by a worker. The preset time may be half a year, etc., and embodiments of the present invention are not described herein in detail. The longer the preset time is, the higher the reference value and the practicability of the user feature vector correspondingly obtained. In the implementation of the present invention, preferably, the minimum unit granularity of the preset time length is one complete transaction day, so as to ensure that the data can be acquired in order.
Specifically, a target user is determined, and then the name of the merchant needing to obtain the vector feature is determined according to the transaction data of the corresponding target user.
In this embodiment of the present invention, before the transaction data is used as a source of the transaction behavior sequence data of the merchant, the method further includes:
identifying a transaction day for which the number of transaction merchants is less than a first number threshold, removing data for the identified transaction day from the transaction data; identifying a transaction day for which the number of transactions is greater than a second quantity threshold, truncating transaction data that exceeds the second quantity threshold within the identified transaction day.
The first quantity threshold and the second quantity threshold are set by workers according to requirements, or the system can count according to the existing historical data, the number of daily transaction merchants and the daily transaction times under the scene with the highest historical income are respectively used as the first quantity threshold and the second quantity threshold so as to maximize the economic income, and the first quantity threshold and the second quantity threshold can be set according to the daily consumption habits of each target user.
Specifically, due to instability of the transaction of the target user, for example, the target user is sick and busy in a certain time, the number of merchants transacting the target user in a certain day is too small, the data of the transaction day lacks representativeness, or the user performs frequent consumption exceeding the normal level in a transaction day, and the situations of bill swiping, embezzlement and the like may exist, and the corresponding transaction data cannot guarantee authenticity. The embodiment of the invention removes the transaction date with the number of daily transaction merchants smaller than the first number threshold value from the data, and truncates the transaction data with the daily transaction frequency larger than the second number threshold value within the transaction date, so as to realize efficient screening of transaction behavior sequences and ensure the representativeness and the authenticity of the data for generating the vector.
And S220, acquiring transaction behavior sequence data of the merchant.
And S230, converting the transaction behavior sequence data into merchant feature vectors based on a set item vector conversion model.
S240, constructing a user feature vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant feature vector corresponding to the merchant.
S250, training an original credit model according to the user feature vector of the target user; and taking the original credit model after training as a new credit model.
The original credit model is a built credit model, and whether the original credit model is used or not or is perfected, the built credit model for judging the credit capability of the user belongs to the protection scope of the original credit model of the application.
Specifically, in order to effectively train the credit model, a large number of target users may be selected, and feature vector samples corresponding to the target users may be obtained, and the specific method for obtaining the feature vectors of the target users may refer to the above embodiments, which are not described herein in detail. And when the feature vector of the target user is obtained, overdue performance of the target user can be collected, and the original credit model is trained and further improved according to the user feature vector of the target user and past overdue performance.
In this embodiment of the present invention, preferably, training the credit model according to the feature vector of the target user includes: combining the user characteristic vector of the target user with the original characteristic vector of the target user to obtain a combined characteristic vector of the target user; and inputting the combined feature vector of the target user into an original credit model for training.
The original feature vector is the feature vector which only describes the dominant feature of the user at present and does not consider the recessive trade feature of the user.
Specifically, in the embodiment of the present invention, the original feature vector obtained by using the prior art is combined with the user feature vector in the embodiment of the present invention, so as to obtain the combined feature vector of the target user. The combined feature vector simultaneously depicts explicit and implicit transaction features of a target user, and a new credit model obtained by training the combined training feature vector can more accurately reflect actual consumption habits and credit risks of the user.
On the basis of the embodiment, the embodiment of the invention further reduces the consumption of computing resources, improves the processing capacity of equipment, realizes the pre-screening of transaction behavior sequences, ensures the representativeness and the authenticity of data for generating vectors, and utilizes a new credit model obtained by training combined training feature vectors, thereby reflecting the actual consumption habits and credit risks of users more accurately.
EXAMPLE III
Fig. 3 is a block diagram of a model feature extraction device for credit and central processing, which is provided in a third embodiment of the present invention, and is capable of executing the method for extracting model feature for credit and central processing provided in any embodiment of the present invention, and has corresponding functional modules and beneficial effects of the execution method. As shown in fig. 3, the apparatus may include:
a transaction data obtaining module 310, configured to obtain transaction behavior sequence data of a merchant;
a feature vector conversion module 320, configured to convert the transaction behavior sequence data into a merchant feature vector based on a set item vector conversion model;
the feature vector construction module 330 is configured to construct a user feature vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant feature vector corresponding to the merchant.
The product can execute the method for extracting the characteristics of the model for credit-based mail provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
Optionally, the feature vector conversion module 320 is specifically configured to process the transaction behavior sequence data based on a serialized feature extraction model to obtain processed transaction behavior sequence data; and inputting the processed transaction behavior sequence data into the set item vector conversion model to obtain a merchant feature vector.
Optionally, the feature vector constructing module 330 is specifically configured to determine transaction times of the target user and each merchant according to transaction behavior data of at least one merchant; multiplying the feature vector of each merchant by the corresponding transaction times to obtain an intermediate feature vector of each merchant; and taking a vector obtained by adding all the merchant intermediate feature vectors as the user feature vector of the target user.
Optionally, the model feature extraction device for credit system further includes: target user determination module 340.
The target user determination module 340 is specifically configured to acquire transaction data of the target user within a preset time, and read a merchant in the transaction data; and taking the transaction data as a data source of transaction behavior sequence data of the merchant.
Optionally, the target user determining module 340 is further configured to identify a transaction day when the number of the transaction merchants is smaller than the first number threshold, and remove data of the identified transaction day from the transaction data; identifying a transaction day for which the number of transactions is greater than a second quantity threshold, truncating transaction data that exceeds the second quantity threshold within the identified transaction day.
Optionally, the credit system model feature extraction apparatus further includes a credit model training module 350.
The credit model training module 350 is specifically configured to train an original credit model according to the user feature vector of the target user; and taking the original credit model after training as a new credit model.
Optionally, the credit model training module 350 is further configured to combine the user feature vector of the target user with the original feature vector of the target user to obtain a combined feature vector of the target user; and inputting the combined feature vector of the target user into an original credit model for training.
The product further described above can also execute the data processing method for extracting the model features for credit and credit provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example four
Fig. 4 is a schematic structural diagram of a model feature extraction apparatus for credit system according to a fourth embodiment of the present invention, as shown in fig. 4, the apparatus includes a processor 40, a memory 41, an input device 42, and an output device 43; the number of processors 40 in the device may be one or more, and one processor 40 is taken as an example in fig. 4; the processor 40, the memory 41, the input means 42 and the output means 43 in the device may be connected by a bus or other means, as exemplified by the bus connection in fig. 4.
The memory 41 is a computer-readable storage medium, and can be used for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the credit crediting model feature extraction method in the embodiment of the present invention (for example, the transaction data acquisition module 310, the feature vector conversion module 320, and the feature vector construction module 330 in the credit crediting model feature extraction apparatus). The processor 40 executes various functional applications of the apparatus and data processing by executing software programs, instructions and modules stored in the memory 41, namely, the method for extracting model features for credit as described above.
The memory 41 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 41 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, memory 41 may further include memory located remotely from processor 40, which may be connected to the device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 42 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function control of the terminal. The output device 43 may include a display device such as a display screen.
EXAMPLE five
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, where the computer-executable instructions are executed by a computer processor to perform a method for extracting model features for credit, the method including:
acquiring transaction behavior sequence data of a merchant;
converting the transaction behavior sequence data into merchant feature vectors based on a set item vector conversion model;
and constructing a user characteristic vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant characteristic vector corresponding to the merchant.
Of course, the storage medium provided by the embodiments of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the operations of the method described above, and may also perform related operations in the method for extracting model features for credit and credit provided by any embodiments of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the above search apparatus, each included unit and module are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A credit system model feature extraction method is characterized by comprising the following steps:
acquiring transaction behavior sequence data of a merchant;
converting the transaction behavior sequence data into merchant feature vectors based on a set item vector conversion model;
and constructing a user characteristic vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant characteristic vector corresponding to the merchant.
2. The method of claim 1, wherein prior to obtaining transaction behavior sequence data for the merchant, further comprising:
acquiring transaction data of the target user within preset time, and reading a merchant in the transaction data;
and taking the transaction data as a data source of transaction behavior sequence data of the merchant.
3. The method of claim 2, wherein prior to sourcing the transaction data as transaction behavior sequence data for the merchant, further comprising:
identifying a transaction day for which the number of transaction merchants is less than a first number threshold, removing data for the identified transaction day from the transaction data;
identifying a transaction day for which the number of transactions is greater than a second quantity threshold, truncating transaction data that exceeds the second quantity threshold within the identified transaction day.
4. The method of claim 1, wherein converting the transaction behavior sequence data into merchant feature vectors based on a set-item vector conversion model comprises:
processing the transaction behavior sequence data based on a serialization feature extraction model to obtain processed transaction behavior sequence data;
and inputting the processed transaction behavior sequence data into the set item vector conversion model to obtain a merchant feature vector.
5. The method of claim 1, wherein constructing the user feature vector of the target user according to transaction behavior data of the target user and at least one merchant and a merchant feature vector corresponding to the merchant comprises:
determining the transaction times of a target user and each merchant according to the transaction behavior data of at least one merchant;
multiplying the feature vector of each merchant by the corresponding transaction times to obtain an intermediate feature vector of each merchant;
and taking a vector obtained by adding all the merchant intermediate feature vectors as the user feature vector of the target user.
6. The method according to claim 1, wherein after constructing the user feature vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant feature vector corresponding to the merchant, the method further comprises:
training an original credit model according to the user feature vector of the target user;
and taking the original credit model after training as a new credit model.
7. The method of claim 6, wherein training a credit model based on the feature vectors of the target users comprises:
combining the user characteristic vector of the target user with the original characteristic vector of the target user to obtain a combined characteristic vector of the target user;
and inputting the combined feature vector of the target user into an original credit model for training.
8. A model feature extraction apparatus for credit, the apparatus comprising:
the transaction data acquisition module is used for acquiring transaction behavior sequence data of a merchant;
the feature vector conversion module is used for converting the transaction behavior sequence data into merchant feature vectors based on a set project vector conversion model;
the characteristic vector construction module is used for constructing a user characteristic vector of the target user according to the transaction behavior data of the target user and at least one merchant and the merchant characteristic vector corresponding to the merchant.
9. A model feature extraction apparatus for credit, the apparatus comprising:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method for model feature extraction for credit reporting as claimed in any of claims 1-7.
10. A storage medium containing computer-executable instructions for performing the method for model feature extraction for credit according to any one of claims 1-7 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111337002.6A CN114037516A (en) | 2021-11-12 | 2021-11-12 | Model feature extraction method, device, equipment and medium for credit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111337002.6A CN114037516A (en) | 2021-11-12 | 2021-11-12 | Model feature extraction method, device, equipment and medium for credit |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114037516A true CN114037516A (en) | 2022-02-11 |
Family
ID=80144096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111337002.6A Pending CN114037516A (en) | 2021-11-12 | 2021-11-12 | Model feature extraction method, device, equipment and medium for credit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114037516A (en) |
-
2021
- 2021-11-12 CN CN202111337002.6A patent/CN114037516A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110046929B (en) | Fraudulent party identification method and device, readable storage medium and terminal equipment | |
| CN110378575B (en) | Overdue event refund collection method and device and computer readable storage medium | |
| CN109035013A (en) | Detect the device and method and computer readable storage medium of abnormal financial transaction | |
| CN110807699B (en) | Overdue event payment collection method and device and computer readable storage medium | |
| CN111047426A (en) | Method, system and storage medium for analyzing suspicious account based on fund transaction | |
| CN111967807A (en) | Method and device for generating risk event judgment rule executed by computer | |
| CN114358778A (en) | Method, device, equipment and storage medium for preventing payment account from being stolen | |
| CN113887214A (en) | Artificial intelligence based wish presumption method and related equipment thereof | |
| CN113159924A (en) | Method and device for determining trusted client object | |
| CN113112347A (en) | Determination method of hasty collection decision, related device and computer storage medium | |
| CN114037516A (en) | Model feature extraction method, device, equipment and medium for credit | |
| CN115965468A (en) | Transaction data-based abnormal behavior detection method, device, equipment and medium | |
| CN114756685A (en) | Complaint risk identification method and device for complaint sheet | |
| CN111429257A (en) | Transaction monitoring method and device | |
| CN112488748A (en) | Data identifier matching method and device, storage medium and computing equipment | |
| CN112150268A (en) | Computer equipment and intelligent customer acquisition system | |
| CN111160916A (en) | Risk transaction identification method and device | |
| CN115018638B (en) | Method and device for determining service limit | |
| CN114155038B (en) | Epidemic situation affected user identification method | |
| CN111738834B (en) | Data processing method and device | |
| CN116882998B (en) | Virtual currency fund analysis method | |
| CN110782342B (en) | Method and device for verifying correctness of new channel feature engineering based on binary classification model | |
| CN113032442A (en) | Method and device for determining target user, electronic equipment and storage medium | |
| CN117076537A (en) | Feature derivation method and device of transaction flow data and electronic equipment | |
| CN114626940A (en) | Data analysis method and device and electronic 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 |