CN115510131A - Data item display method and device - Google Patents

Data item display method and device Download PDF

Info

Publication number
CN115510131A
CN115510131A CN202211274404.0A CN202211274404A CN115510131A CN 115510131 A CN115510131 A CN 115510131A CN 202211274404 A CN202211274404 A CN 202211274404A CN 115510131 A CN115510131 A CN 115510131A
Authority
CN
China
Prior art keywords
data
input
machine learning
classifier
training set
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
CN202211274404.0A
Other languages
Chinese (zh)
Inventor
金洋旭
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.)
Bank of China Ltd
Original Assignee
Bank of China 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 Bank of China Ltd filed Critical Bank of China Ltd
Priority to CN202211274404.0A priority Critical patent/CN115510131A/en
Publication of CN115510131A publication Critical patent/CN115510131A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/248—Presentation of query results
    • 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
    • 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

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Marketing (AREA)
  • Artificial Intelligence (AREA)
  • Technology Law (AREA)
  • Computational Linguistics (AREA)
  • Strategic Management (AREA)
  • Databases & Information Systems (AREA)
  • Economics (AREA)
  • General Business, Economics & Management (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Development Economics (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

The invention discloses a data item display method and a device, which relate to artificial intelligence, wherein the method comprises the following steps: determining an occurrence scene and an input channel of input transaction data; according to the occurrence scene and the input channel of transaction data input, matching data items to be displayed for the transaction data through the classification result of the classifier selected based on the machine learning defect prediction model; the machine learning model is used for processing a training set into balance data after a display data item expected by a user is input into the machine learning-based defect prediction model under different scenes and different input channels, and the balance data is used for the machine learning model to select a classifier. The invention can improve the display accuracy of the transaction in different channels and different scenes without manual intervention, provide accurate and friendly customer service and improve the experience of customers.

Description

Data item display method and device
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a data item display method and device.
Background
Currently, there is an increasing variety of transaction types, and the sources of transaction data are different, such as financial transactions.
Under each kind of transaction, different data item display requirements are provided, and for better service, under different scenes, the display requirements of the transaction are met, the data items to be displayed need to be determined in manners of manual investigation and judgment and the like, which undoubtedly consumes a large amount of manpower and material resources.
The prior art has the defect that no technical scheme supports the transaction to display corresponding data items under different channels and different scenes.
Disclosure of Invention
The embodiment of the invention provides a data item display method, which is used for solving the problem that no technical scheme supports transactions to display corresponding data items in different channels and different scenes, and comprises the following steps:
determining an occurrence scene and an input channel of input transaction data;
according to the occurrence scene and the input channel of transaction data input, matching data items to be displayed for the transaction data through the classification result of the classifier selected based on the machine learning defect prediction model; the machine learning model is used for processing a training set into balance data after a display data item expected by a user is input into the machine learning-based defect prediction model under different scenes and different input channels, and the balance data is used for the machine learning model to select a classifier.
The embodiment of the invention also provides a data item display device, which is used for solving the problem that no technical scheme supports the transaction to display corresponding data items in different channels and different scenes, and comprises the following components:
the transaction module is used for determining the occurrence scene and the input channel of the input transaction data;
the display module is used for matching data items to be displayed for the transaction data according to the occurrence scene and the input channel of the transaction data input and the classification result of the classifier selected by the defect prediction model based on machine learning; the machine learning model is used for processing a training set into balance data after a display data item expected by a user is input into the machine learning-based defect prediction model under different scenes and different input channels, and the balance data is used for the machine learning model to select a classifier.
The embodiment of the invention also provides computer equipment which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein the processor realizes the data item display method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, and when the computer program is executed by a processor, the data item display method is realized.
An embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the computer program implements the above data item display method.
In the embodiment of the invention, compared with the technical scheme of manual research and management in the prior art, the data items which are expected to be displayed by the user under different occurrence scenes and input channels of transaction data are trained by using the defect prediction model based on machine learning, so that the data items which are suitable to be displayed under different occurrence scenes and input channels can be obtained, the classifier is adopted to perform classification and identification, and the appropriate display data items are matched, so that the optimal classification effect can be achieved on any data set, the display accuracy of the transaction under different channels and different scenes can be improved without manual intervention, the accurate and friendly customer service is provided, and the customer experience is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
FIG. 1 is a schematic flow chart illustrating an implementation of a data item presentation method according to an embodiment of the present invention;
FIG. 2 is a diagram of a data item presentation architecture according to an embodiment of the present invention;
FIG. 3 is a flow chart illustrating an implementation of data item presentation according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a data item presentation apparatus according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a computer device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
Various transaction types are continuously increased, channels of transaction data sources are different, and the financial transaction display requirements under different scenes of better service are met. The technical scheme provided by the embodiment of the invention can improve the display accuracy and friendliness of the transaction in different channels and different scenes, reduce the cost of manual research and manual management, improve the maintenance efficiency and avoid the defects.
The following description will be given with reference to examples.
First, the concept involved in the embodiments will be explained.
The transaction refers to a transaction that triggers accounting processing, such as transfer of money, deposit of money, remittance of money.
Accounting refers to the customer's accounting information, such as amount, cost, etc.
Channels refer to ports where data comes from different service ports, such as from cell phone banks, internet banking, etc.
A scenario refers to a minimum granularity event that produces a financial or non-financial transaction.
Fig. 1 is a schematic diagram of an implementation flow of a data item presentation method, as shown in fig. 1, which may include:
step 101, determining an occurrence scene and an input channel of input transaction data;
102, matching data items to be displayed for transaction data through a classification result of a classifier selected based on a machine learning defect prediction model according to an occurrence scene and an input channel of the transaction data input; the machine learning model is used for processing a training set into balance data after a display data item expected by a user is input into the machine learning-based defect prediction model under different scenes and different input channels, and the balance data is used for the machine learning model to select a classifier.
In the implementation, the training set is processed into the balance data, and the training set is processed into the balance data by using a sampling method.
In particular, the training set may be processed into balance data using a sampling method.
In implementation, when the training set is processed by using a sampling method, the SMOTE algorithm is used.
Specifically, the SMOTE algorithm can be used in a defect prediction model to solve the common unbalanced problem in defect prediction.
In implementation, the classifier selected based on the machine-learned defect prediction model is an adaptive selection classifier.
Specifically, an ASC (Adaptive Selection Classifier) may be used instead of the Classifier in the conventional machine learning model. The ASC combines various traditional classification models, and can select the most appropriate classifier for the new data according to the characteristics of the new data, so that the classification effect is improved, and the optimal scheme is selected for display.
When the scheme is implemented, firstly, a machine learning model is provided. The model mainly comprises two parts:
firstly, a defect prediction model based on machine learning is used, and the model can process common unbalance problems in defect prediction; secondly, a self-adaptive selection classifier is used, the classifier which is most suitable for the data set can be selected in a self-adaptive mode according to the characteristics of input data, and therefore the classification effect is improved.
Through the model and the input of related channels and scenes, the optimal display fields can be automatically distributed for the channels and scenes to display the content.
In specific implementation, a common transaction scenario and a desired presentation item are firstly obtained as a feature input model.
The model can solve the common unbalance problem in the defect prediction by using the SMOTE algorithm; meanwhile, by using the ASC, the most appropriate classifier can be selected according to input data, so that the classification effect is improved, and the classifier can improve the classification capability of a machine learning model. The following is an example.
Fig. 2 is a schematic diagram of a data item presentation architecture, and as shown in fig. 2, at least one functional architecture that can implement data item presentation may include: the device comprises a feature input module, a non-equilibrium data processing module and an adaptive classifier selection module, wherein:
a feature input module: the display data item is used for inputting transaction data expectation under different scenes and different data channels. As shown in table 1, table 1 is a set of data items that a bank self-service channel desires to display for a public scenario transaction list.
Table 1:
Figure BDA0003896472900000041
Figure BDA0003896472900000051
the unbalanced data processing module: after data is input into a machine learning model, aiming at the unbalanced data problem, a training set can be processed by using a sampling method to be changed into balanced data, and the commonly used sampling method is divided into oversampling (oversampling) and undersampling (undersampling).
Oversampling is based on a Bootstrap sampling method, data can be added into the minority class through modes of copying and the like, and the number of the minority class is equal to that of the majority class; in contrast, the under-sampling method subtracts data from most classes. It can be seen that oversampling may add duplicate data, resulting in data overfitting, while undersampling may result in loss of important data due to data being chopped.
The SMOTE algorithm can be used for reducing the data imbalance degree by analyzing and simulating a few types of samples and adding new samples into a data set. According to the method, for k neighbors (such as Euclidean distance used as standard calculation) of each sample x in a minority class, N samples are randomly selected to carry out random linear difference, a new minority class sample is constructed, and the new sample and original data are synthesized to generate a new training set. The method is an improved scheme based on random oversampling, and is a means for processing non-equalized data. Many parameters in the SMOTE algorithm have important influence on experimental results, the setting of the parameters often depends on the experience of people, many previous researches directly use default parameters of the algorithm, and the setting cannot be well adapted to all data sets. SMOTUNED randomly generates initial values of parameters (k, r, m) in the SMOTE algorithm by using a Differential Evolution algorithm (Differential Evolution), and gradually seeks the optimal solution of the parameters through the Evolution process of variation, intersection and selection, so that the method is a self-adaptive SMOTE method.
An adaptive classifier selection module: and selecting the most suitable classifier for the new data according to the characteristics of the new data, and selecting the best scheme for showing.
Traditional machine learning models have been successful in a number of areas, such as natural language processing, image recognition, etc., but no classifier has been found to achieve optimal classification in any situation. In this case, the present invention replaces the Classifier in the conventional machine learning model with an Adaptive Selection of Classifier ASC (Adaptive Selection of Classifier) proposed by Nucci. The ASC combines various traditional classification models, and can select the most appropriate classifier for the new data according to the characteristics of the new data, so that the classification effect is improved, and the optimal scheme is selected for display.
Fig. 3 is a schematic diagram of a data item display implementation process, as shown in fig. 3, which may include:
step 301, inputting expected display data items of transaction data under different scenes and different channels in the feature input module.
Step 302, the unbalanced data processing module inputs data into the machine learning model, and processes the training set to change the training set into balanced data aiming at unbalanced data problems.
And 303, selecting the most appropriate classifier for the new data according to the characteristics of the new data by the self-adaptive classifier selecting module, thereby improving the classification effect and selecting the optimal scheme for display.
The embodiment of the invention also provides a data item display device, which is described in the following embodiment. Because the principle of the device for solving the problems is similar to the data item display method, the implementation of the device can refer to the implementation of the data item display method, and repeated details are not repeated.
Fig. 4 is a schematic structural diagram of a data item presentation device, as shown in fig. 4, which may include:
the transaction module 401 is configured to determine an occurrence scenario and an input channel of input transaction data;
a display module 402, configured to match data items to be displayed for the transaction data according to the occurrence scenario and the input channel of the transaction data input, through a classification result of a classifier selected based on a machine learning defect prediction model; the machine learning model is used for processing a training set into balance data after a display data item expected by a user is input into a defect prediction model based on machine learning in different scenes and different input channels, and the balance data is used for the machine learning model to select a classifier.
In implementation, the display module is further configured to process the training set into balance data, and convert the training set into balance data after processing the training set by using a sampling method.
In an implementation, the presentation module is further configured to use a SMOTE algorithm when processing the training set using a sampling method.
In an implementation, the presentation module is further configured to select a classifier in the machine learning based defect prediction model, which is an adaptive selection classifier.
When the technical scheme provided by the embodiment of the invention is implemented, the implementation can be carried out as follows.
Fig. 5 is a schematic diagram of a computer device, as shown in fig. 5, the computer device includes:
the processor 500, which is used to read the program in the memory 520, executes the following processes:
determining an occurrence scene and an input channel of input transaction data;
according to an occurrence scene and an input channel of transaction data input, matching data items to be displayed for the transaction data through a classification result of a classifier selected based on a machine learning defect prediction model; the machine learning model is used for processing a training set into balance data after a display data item expected by a user is input into a defect prediction model based on machine learning under different scenes and different input channels, and the balance data is used for the machine learning model to select a classifier;
a transceiver 510 for receiving and transmitting data under the control of the processor 500.
In the implementation, the training set is processed into balance data, and the training set is processed into balance data by using a sampling method.
In implementation, when the training set is processed by using a sampling method, the SMOTE algorithm is used.
In implementation, the classifier selected based on the machine-learned defect prediction model is an adaptive selection classifier.
Wherein in fig. 5, the bus architecture may include any number of interconnected buses and bridges, with one or more processors, represented by processor 500, and various circuits, represented by memory 520, being linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The transceiver 510 may be a number of elements including a transmitter and a receiver that provide a means for communicating with various other apparatus over a transmission medium. The processor 500 is responsible for managing the bus architecture and general processing, and the memory 520 may store data used by the processor 500 in performing operations.
An embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program implements the above data item display method.
An embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the computer program implements the above data item display method.
The technical scheme provided by the embodiment of the invention is easy to realize, and can achieve the best classification effect on any data set; the transaction management is convenient, and accurate and friendly customer service is provided.
According to the technical scheme, the data acquisition, storage, use, processing and the like meet relevant regulations of national laws and regulations, and various types of data such as personal identity data, operation data, behavior data and the like related to individuals, clients, crowds and the like are authorized.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (11)

1. A method for presenting data items, comprising:
determining an occurrence scene and an input channel of input transaction data;
according to an occurrence scene and an input channel of transaction data input, matching data items to be displayed for the transaction data through a classification result of a classifier selected based on a machine learning defect prediction model; the machine learning model is used for processing a training set into balance data after a display data item expected by a user is input into the machine learning-based defect prediction model under different scenes and different input channels, and the balance data is used for the machine learning model to select a classifier.
2. The method of claim 1, wherein the training set is processed into the balance data by processing the training set using a sampling method.
3. The method of claim 2, wherein the training set is processed using a sampling method using a SMOTE algorithm.
4. The method of claim 1, wherein the classifier selected based on the machine-learned defect prediction model is an adaptively selected classifier.
5. A data item presentation device, comprising:
the transaction module is used for determining the occurrence scene and the input channel of the input transaction data;
the display module is used for matching data items to be displayed for the transaction data according to the occurrence scene and the input channel of the transaction data input and the classification result of the classifier selected by the defect prediction model based on machine learning; the machine learning model is used for processing a training set into balance data after a display data item expected by a user is input into a defect prediction model based on machine learning in different scenes and different input channels, and the balance data is used for the machine learning model to select a classifier.
6. The apparatus of claim 5, wherein the presentation module is further configured to process the training set into the balance data by processing the training set using a sampling method.
7. The apparatus of claim 6, wherein the presentation module is further configured to use a SMOTE algorithm when processing the training set using a sampling method.
8. The apparatus of claim 5, wherein the presentation module is further configured to select the classifier in the machine learning based defect prediction model as an adaptively selected classifier.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of any of claims 1 to 4 when executing the computer program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program which, when executed by a processor, implements the method of any of claims 1 to 4.
11. A computer program product, characterized in that the computer program product comprises a computer program which, when being executed by a processor, carries out the method of any one of claims 1 to 4.
CN202211274404.0A 2022-10-18 2022-10-18 Data item display method and device Pending CN115510131A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211274404.0A CN115510131A (en) 2022-10-18 2022-10-18 Data item display method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211274404.0A CN115510131A (en) 2022-10-18 2022-10-18 Data item display method and device

Publications (1)

Publication Number Publication Date
CN115510131A true CN115510131A (en) 2022-12-23

Family

ID=84509525

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211274404.0A Pending CN115510131A (en) 2022-10-18 2022-10-18 Data item display method and device

Country Status (1)

Country Link
CN (1) CN115510131A (en)

Similar Documents

Publication Publication Date Title
CN111428217B (en) Fraudulent party identification method, apparatus, electronic device and computer readable storage medium
CN111931055A (en) Object recommendation method, object recommendation device and electronic equipment
CN112765230B (en) Payment big data analysis method and big data analysis system based on internet finance
CN110930038A (en) Loan demand identification method, loan demand identification device, loan demand identification terminal and loan demand identification storage medium
CN113282623A (en) Data processing method and device
CN111144409A (en) Order following, accepting and examining processing method and system
US20210349920A1 (en) Method and apparatus for outputting information
Valizadegan et al. Learning to trade off between exploration and exploitation in multiclass bandit prediction
CN111967973B (en) Bank customer data processing method and device
US20190114639A1 (en) Anomaly detection in data transactions
CN111245815B (en) Data processing method and device, storage medium and electronic equipment
US20170300937A1 (en) System and method for inferring social influence networks from transactional data
CN115510131A (en) Data item display method and device
Liang et al. ASE: Anomaly scoring based ensemble learning for highly imbalanced datasets
CN112365343B (en) Banking service system and method based on mobile robot
CN112101952B (en) Bank suspicious transaction evaluation and data processing method and device
CN114997958A (en) Financial product recommendation processing method and device
CN115375487A (en) Recommendation processing method and device for fund products
CN115423020A (en) Method and device for generating mobile banking transaction account book
CN115936832A (en) Credit risk assessment method and device
CN115205051A (en) Bank product information recommendation method and device
CN118710391A (en) Risk object identification method and device, storage medium and electronic equipment
CN116976906A (en) Transaction risk assessment method, device, electronic equipment and medium
CN115375331A (en) Credit card outbound customer list screening method and device
CN115796424A (en) Object clustering method and related device

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination