CN113033658A - Merchant classification method and device, electronic equipment and medium - Google Patents

Merchant classification method and device, electronic equipment and medium Download PDF

Info

Publication number
CN113033658A
CN113033658A CN202110313676.6A CN202110313676A CN113033658A CN 113033658 A CN113033658 A CN 113033658A CN 202110313676 A CN202110313676 A CN 202110313676A CN 113033658 A CN113033658 A CN 113033658A
Authority
CN
China
Prior art keywords
output data
neural network
merchant
layer
inputting
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
CN202110313676.6A
Other languages
Chinese (zh)
Inventor
匡海健
郑梓悫
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.)
China Construction Bank Corp
Original Assignee
China Construction Bank Corp
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 China Construction Bank Corp filed Critical China Construction Bank Corp
Priority to CN202110313676.6A priority Critical patent/CN113033658A/en
Publication of CN113033658A publication Critical patent/CN113033658A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
    • G06F18/24133—Distances to prototypes
    • G06F18/24137—Distances to cluster centroïds
    • G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Probability & Statistics with Applications (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiment of the application discloses a merchant classification method, a merchant classification device, electronic equipment and a merchant classification medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: inputting a preset number of merchant transaction data acquired in each preset unit time into a recurrent neural network to obtain first output data; inputting the first output data into a convolutional neural network to obtain second output data; and processing the second output data based on a normalization function to obtain a merchant classification result. The scheme solves the problem that the accuracy rate is low when the commercial tenant is classified at present by adopting a manual classification method or other model classification methods, so that the correlation characteristics of commercial tenant transaction data at different times are determined through the cyclic neural network, the characteristics among the commercial tenant transaction data are further determined through the convolutional neural network, and the classification accuracy is not improved.

Description

Merchant classification method and device, electronic equipment and medium
Technical Field
The embodiment of the application relates to the technical field of artificial intelligence, in particular to a merchant classification method, a merchant classification device, electronic equipment and a medium.
Background
In a merchant billing transaction, the rate of charge for the billing transaction is affected by the type of merchant. At present, the merchant type is generally entered manually when the merchant file is established, a method for manually classifying merchants is adopted, the workload is too heavy, the method has certain subjectivity, and different people can generate different classification results due to different understandings of merchant classification standards, so that the accuracy of merchant classification is influenced.
In addition, the transaction area of the merchant can be determined according to the payment position information in the transaction. And combining the information of the merchant operation area with other information such as transaction scale, buyer scale and the like to determine a merchant classification model, and then classifying unknown merchants by using the trained model to realize the combination of the information of the merchant operation area and other operation information. The technical scheme depends on the positioning of payment information and the acquisition of buyer information during merchant transaction, but generally in the transaction, the information is not easy to acquire, so the scheme has certain limitation, the K neighbor algorithm used by the scheme has high calculation complexity, the value of K is worthy of selection and depends on the result of data training to a great extent, and the improper value of K can make classification difficult to achieve the expected effect.
Disclosure of Invention
The embodiment of the invention provides a merchant classification method, a merchant classification device, electronic equipment and a medium, which are used for improving accuracy of merchant classification.
Inputting a preset number of merchant transaction data acquired in each preset unit time into a recurrent neural network to obtain first output data;
inputting the first output data into a convolutional neural network to obtain second output data;
and processing the second output data based on a normalization function to obtain a merchant classification result.
In another embodiment, an embodiment of the present application further provides a merchant classification apparatus, where the apparatus includes:
the first processing module is used for inputting the preset number of merchant transaction data acquired in each preset unit time into the recurrent neural network to obtain first output data;
the second processing module is used for inputting the first output data into a convolutional neural network to obtain second output data;
and the classification module is used for processing the second output data based on the normalization function to obtain a merchant classification result.
In another embodiment, an embodiment of the present application further provides an electronic device, including: one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, the one or more programs cause the one or more processors to implement the merchant classification method of any one of the embodiments of the present application.
In yet another embodiment, the present application further provides a computer-readable storage medium, on which a computer program is stored, and the program, when executed by a processor, implements the merchant classification method according to any one of the embodiments of the present application.
In the embodiment of the application, the preset amount of merchant transaction data collected in each preset unit time is input into a recurrent neural network to obtain first output data; inputting the first output data into a convolutional neural network to obtain second output data; and processing the second output data based on a normalization function to obtain a merchant classification result. The scheme solves the problem that the classification accuracy rate obtained by adopting a manual classification method or other model classification methods is low when the merchants are classified at present, so that the correlation characteristics of the merchant transaction data at different times are determined through the cyclic neural network, the characteristics among the merchants transaction data are further determined through the convolutional neural network, and the classification accuracy is not improved.
Drawings
FIG. 1 is a flow chart of a merchant classification method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a recurrent neural network provided in accordance with an embodiment of the present invention;
FIG. 3 is a flowchart of a merchant classification method according to another embodiment of the present invention;
FIG. 4 is a schematic diagram of a convolutional neural network according to another embodiment of the present invention;
FIG. 5 is a diagram of a neural network structure of a merchant classification method according to another embodiment of the present invention;
FIG. 6 is a diagram illustrating a normalized network according to another embodiment of the present invention;
fig. 7 is a schematic structural diagram of a merchant classifying device according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of an electronic device according to an 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.
Fig. 1 is a flowchart of a merchant classification method according to an embodiment of the present invention. The merchant classification method provided by the embodiment of the application can be suitable for the condition of classifying merchants. Typically, the embodiment of the application is suitable for the case of classifying the model based on the neural network. The method may be specifically performed by a merchant classification apparatus, which may be implemented by software and/or hardware, and may be integrated in an electronic device capable of implementing the merchant classification method. Referring to fig. 1, the method of the embodiment of the present application specifically includes:
and S110, inputting the preset number of merchant transaction data collected in each preset unit time into a recurrent neural network to obtain first output data.
The preset unit time may be determined according to actual conditions, and may be, for example, 1 minute, 1 hour, 12 hours, or the like. The preset number can also be determined according to actual conditions. For example, the preset number may be 100, and the number of the merchant transaction data of one merchant in 1 hour is 1000, then 100 merchant transaction data are selected from the 1000 merchant transaction data for merchant classification. The merchant transaction data in the embodiment of the application is transaction data of a merchant of which the actual type is to be detected. Before merchant classification, the neural network is trained in advance according to merchant transaction data of different merchant types and merchant type labels as training data.
Illustratively, a preset amount of merchant transaction data collected between preset units is input into the recurrent neural network to obtain first output data. The recurrent neural network can determine the context relationship between different time sequences of the merchant transaction data, thereby accurately knowing the relationship between the merchant transaction data of different time sequences and determining the characteristics contained in the merchant transaction data.
In this embodiment of the present application, inputting the transaction data of the merchants collected in each preset unit time in a preset number into the recurrent neural network to obtain first output data, where the first output data includes: and inputting the output data of the cyclic neural network of the current layer to the input layer of the next layer of cyclic neural network, and processing the output data by the next layer of cyclic neural network to obtain first output data.
Illustratively, as shown in fig. 2, fig. 2 is a layer of recurrent neural network, and in the embodiment of the present application, more layers of recurrent neural networks may also be provided, for example, 2 layers, 3 layers, and so on. And inputting the output data of the current layer of the recurrent neural network to the input layer of the next layer of the recurrent neural network, and processing the output data by the next layer of the recurrent neural network to obtain first output data. The present layer and the next layer in the embodiments of the present application are not particularly limited to one, and may be a plurality of layers. When the layer 1 recurrent neural network is taken as the current layer recurrent neural network, the layer 2 recurrent neural network is the next layer recurrent neural network, and when the layer 2 recurrent neural network is the current layer recurrent neural network, the layer 3 recurrent neural network is the next layer recurrent neural network.
In this embodiment of the present application, inputting output data of a recurrent neural network of a current layer to an input layer of a recurrent neural network of a next layer, and obtaining first output data through processing of the recurrent neural network of the next layer, where the method includes: dividing output data of a circulating network of a current layer into groups with preset group numbers; and summarizing the output data of each group, and inputting the data into the input layer nodes of the next layer of the recurrent neural network.
For example, the number of input nodes of the next recurrent neural network may be different from the number of output nodes of the current recurrent neural network, for example, the number of input nodes of the next recurrent neural network may be less than the number of output nodes of the current recurrent neural network. When the number of input nodes of the next layer of the recurrent neural network is less than that of the output nodes of the current recurrent neural network, the output data of the current layer of the recurrent neural network are divided into groups with preset group number, and the output data of each group are aggregated and then input into the input layer nodes of the next layer of the recurrent neural network.
In this embodiment of the present application, the step of summarizing the output data of each group and inputting the summarized output data into an input layer node of a next layer of recurrent neural network includes: splicing the output data of each group to obtain a spliced input matrix; and inputting the spliced input matrix into an input layer node of the next layer of the recurrent neural network.
For example, the output data of each group of the current layer is vector-spliced, and the splicing order is not particularly limited, for example, 5 output vectors of 1 × 10 are spliced into 1 spliced input vector of 5 × 10, and the spliced input matrix is input into the input layer node of the next layer of the recurrent neural network. The perception domain of the first layer of the recurrent neural network is generally 20 time steps, that is, when the transaction running data is more in a unit time, the information is lost once the time period exceeds 20 time periods, and the transaction data characteristics in the previous node cannot be continuously transmitted backwards. The structure of the multilayer RNN can enlarge the perception domain of the recurrent neural network system, so that the transaction pipelines in the whole unit time are mutually influenced.
And S120, inputting the first output data into a convolutional neural network to obtain second output data.
The number of layers of the convolutional neural network may be determined according to actual conditions, and may be determined to be 2 layers, for example. The size of the convolution kernel can be determined according to actual conditions or an actual training process.
Illustratively, the first output data is input into the convolutional neural network to obtain second output data, so that the features of the merchant transaction data are extracted through the convolutional neural network, and merchants are classified according to the features of the merchant transaction data.
And S130, processing the second output data based on the normalization function to obtain a merchant classification result.
Illustratively, in order to enable the second output data to be represented as data in 0-1, which represents the probability that the merchant is in a certain category, the second output data needs to be normalized based on a normalization function, so that the second output data is located between 0-1, and the sum of column vectors of the second output data is equal to 1, so as to determine the merchant category of the merchant more intuitively and accurately.
In the embodiment of the application, the preset amount of merchant transaction data collected in each preset unit time is input into a recurrent neural network to obtain first output data; inputting the first output data into a convolutional neural network to obtain second output data; and processing the second output data based on a normalization function to obtain a merchant classification result. The scheme solves the problem that the classification accuracy rate obtained by adopting a manual classification method or other model classification methods is low when the merchants are classified at present, so that the correlation characteristics of the merchant transaction data at different times are determined through the cyclic neural network, the characteristics among the merchants transaction data are further determined through the convolutional neural network, and the classification accuracy is not improved.
Fig. 3 is a flowchart of a merchant classification method according to another embodiment of the present invention. For further optimization of the embodiments, details which are not described in detail in the embodiments of the present application are described in the embodiments. Referring to fig. 3, a merchant classification method provided in an embodiment of the present application may include:
and S210, expressing the preset amount of merchant transaction data collected in each preset unit time into a column vector form.
For example, if 20 merchant transaction data are collected in the first preset unit time, the 20 merchant transaction data collected in the preset unit time are identified as a column vector of 1 × 20. And the transaction data of the commercial tenants collected in other preset unit time are the same, so that the column vectors with the quantity equal to the quantity of the preset unit time are obtained.
S220, inputting the preset amount of merchant transaction data collected in the preset unit time in each column vector form into each node of the input layer to obtain first output data.
Illustratively, as shown in fig. 2, each input datum X in fig. 2 represents a preset amount of merchant transaction data collected in a preset unit time, that is, a column vector corresponding to the preset unit time. Input layer input of x1、x2、…、xt-1、xtTime-ordered merchant transaction data for the t pieces. The output of the hidden layer at time t is determined by the input at time t and the value of the previous hidden layer. Is formulated as follows:
Ot＝g(VgSt)
St＝f(Ugxt+WgSt-1)
wherein, OtRepresenting the output data of the current output node. StThe transfer values U, V and W representing the current layer can be selected according to actual conditions, and g and f are activation functions and can be determined according to actual conditions.
And S230, inputting the first output data to the convolutional layer and the pooling layer for processing to obtain second output data.
For example, as shown in fig. 4 and 5, the first output data is input to the convolution layer to extract the characteristics of the merchant transaction data, and then input to the pooling layer to perform dimension reduction, so as to extract the statistical characteristics of the merchant transaction data.
In this embodiment of the present application, inputting the first output data to the convolutional layer and the pooling layer for processing to obtain the second output data includes: and processing the first output data based on a first preset convolution kernel to obtain convolution layer output data, and performing dimensionality reduction processing on the convolution layer output data based on a second preset convolution kernel to obtain pooling layer output data serving as second output data. The convolved data can be expressed as:
Figure BDA0002990949450000081
wherein C istIs the output of the convolution, X is the original data set, subscript
Figure BDA0002990949450000082
Figure BDA0002990949450000083
For time step identification, T is the total time length, l is the length of the convolution kernel, which is ω in the formula, and f is an activation function of the result after convolution. b is a deviation variable. In the embodiment of the present application, the sizes of the first preset convolution sum and the second preset convolution sum may be determined according to actual conditions.
And S240, representing the second output data as exponential data.
Illustratively, the second output data is represented as exponential data, thereby ensuring non-negativity of the second output data and ensuring a positive probability value when subsequently calculating the probability.
In the embodiment of the present application, as shown in fig. 6, the representing the second output data as exponential data includes: to be provided with
Figure BDA0002990949450000084
Representing the second output data; wherein Z isiIs the second output data.
And S250, carrying out normalization processing on the index data to obtain a merchant classification result.
Illustratively, the index data is normalized, so that the index data is converted into probability data between 0 and 1, and the classification of the merchants is convenient to intuitively and accurately.
In this embodiment of the present application, the normalizing the index data to obtain merchant classification result data includes: determining the probability of the merchant being classified into a preset merchant list based on the following formula:
Figure BDA0002990949450000091
wherein Z isjAnd K is the number of output nodes. And K is the number of output nodes, namely the number of classified categories. The normalization function may be a Softmax function. K is derived from the preceding neural network training. The input of the Softmax function is a vector, the output of the Softmax function is a vector, each value in the output vector represents the probability that the sample belongs to each class, and the class with the highest probability is the merchant class corresponding to the running water.
In the scheme in the embodiment of the application, the transaction data of the merchants with preset number collected in each preset unit time is represented in the form of column vectors. Inputting the preset amount of merchant transaction data collected in the preset unit time in each column vector form into each node of the input layer to obtain first output data. And inputting the first output data into the convolution layer and the pooling layer for processing to obtain second output data. Representing the second output data as exponential data. And normalizing the index data to obtain a merchant classification result, so that merchants are accurately classified according to the merchant transaction data, and the problem that the classification result is easily influenced by a K value to cause inaccurate classification is solved.
In this embodiment of the present application, inputting the first output data into a convolutional neural network to obtain second output data, including: and inputting the first output data into the convolution layer, the pooling layer and the full-connection layer for processing to obtain second output data. Namely, after the pooling layer, a full connection layer is added to process the transaction data of the merchant.
Fig. 7 is a schematic structural diagram of a merchant classifying device according to an embodiment of the present invention. The device can be applied to the condition of classifying the commercial tenant. Typically, the embodiment of the application is suitable for the case of classifying the model based on the neural network. The apparatus may be implemented by software and/or hardware, and the apparatus may be integrated in an electronic device. Referring to fig. 7, the apparatus specifically includes:
the first processing module 310 is configured to input a preset number of merchant transaction data collected in each preset unit time into the recurrent neural network, so as to obtain first output data;
the second processing module 320 is configured to input the first output data into a convolutional neural network, so as to obtain second output data;
the classification module 330 is configured to process the second output data based on a normalization function to obtain a merchant classification result.
In this embodiment, the first processing module 310 includes:
the vector representation unit is used for representing the preset amount of merchant transaction data collected in each preset unit time into a column vector form;
and the input unit is used for inputting the preset amount of merchant transaction data acquired in the preset unit time in each column vector form into each node of the input layer.
In this embodiment, the first processing module 310 includes:
and the transmission unit is used for inputting the output data of the cyclic neural network of the current layer to the input layer of the next layer of cyclic neural network and obtaining first output data through the processing of the next layer of cyclic neural network.
In an embodiment of the present application, a transfer unit includes:
a grouping subunit, configured to divide output data of the cyclic network of the current layer into groups of a preset number of groups;
and the subsequent input subunit is used for summarizing the output data of each group and inputting the summarized output data into the input layer node of the next layer of the recurrent neural network.
In the embodiment of the present application, the subsequent input subunit is specifically configured to:
splicing the output data of each group to obtain a spliced input matrix;
and inputting the spliced input matrix into an input layer node of the next layer of the recurrent neural network.
In this embodiment, the second processing module 320 is specifically configured to:
and inputting the first output data into the convolution layer and the pooling layer for processing to obtain second output data.
In this embodiment, the second processing module 320 is specifically configured to:
processing the first output data based on a first preset convolution kernel to obtain convolution layer output data
And performing dimension reduction processing on the output data of the convolution layer based on a second preset convolution kernel to obtain pooled layer output data serving as second output data.
In this embodiment, the second processing module 320 is specifically configured to:
and inputting the first output data into the convolution layer, the pooling layer and the full-connection layer for processing to obtain second output data.
In an embodiment of the present application, the classification module 330 includes:
an index representing unit for representing the second output data as index data;
and the normalization processing unit is used for performing normalization processing on the index data to obtain a merchant classification result.
In the embodiment of the present application,
an index representation unit, specifically configured to:
for use in
Figure BDA0002990949450000111
Representing the second output data; wherein Z isiIs the second output data.
In an embodiment of the present application, the normalization processing unit is specifically configured to:
determining the probability of the merchant being classified into a preset merchant list based on the following formula:
Figure BDA0002990949450000121
wherein Z isjAnd K is the number of output nodes.
In an embodiment of the present application, the normalization function is a Softmax function.
The merchant classification device provided by the embodiment of the application can execute the merchant classification method provided by any embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method.
Fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present invention. FIG. 8 illustrates a block diagram of an exemplary electronic device 412 suitable for use in implementing embodiments of the present application. The electronic device 412 shown in fig. 8 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.
As shown in fig. 8, the electronic device 412 may include: one or more processors 416; the memory 428 is configured to store one or more programs, and when the one or more programs are executed by the one or more processors 416, the one or more processors 416 are enabled to implement the merchant classification method provided in the embodiment of the present application, including:
inputting a preset number of merchant transaction data acquired in each preset unit time into a recurrent neural network to obtain first output data;
inputting the first output data into a convolutional neural network to obtain second output data;
and processing the second output data based on a normalization function to obtain a merchant classification result.
The components of the electronic device 412 may include, but are not limited to: one or more processors or processors 416, a memory 428, and a bus 418 that couples the various device components including the memory 428 and the processors 416.
Bus 418 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, transaction ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.
Electronic device 412 typically includes a variety of computer device-readable storage media. These storage media may be any available storage media that can be accessed by electronic device 412 and includes both volatile and nonvolatile storage media, removable and non-removable storage media.
Memory 428 can include computer-device readable storage media in the form of volatile memory, such as Random Access Memory (RAM)430 and/or cache memory 432. The electronic device 412 may further include other removable/non-removable, volatile/nonvolatile computer device storage media. By way of example only, storage system 434 may be used to read from and write to non-removable, nonvolatile magnetic storage media (not shown in FIG. 8, and commonly referred to as a "hard drive"). Although not shown in FIG. 8, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical storage medium) may be provided. In these cases, each drive may be connected to bus 418 by one or more data storage media interfaces. Memory 428 can include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
A program/utility 440 having a set (at least one) of program modules 442 may be stored, for instance, in memory 428, such program modules 442 including, but not limited to, an operating device, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. The program modules 442 generally perform the functions and/or methodologies of the described embodiments of the invention.
The electronic device 412 may also communicate with one or more external devices 414 (e.g., keyboard, pointing device, display 424, etc.), with one or more devices that enable a user to interact with the electronic device 412, and/or with any devices (e.g., network card, modem, etc.) that enable the electronic device 412 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interfaces 422. Also, the electronic device 412 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet) through the network adapter 420. As shown in FIG. 8, network adapter 420 communicates with the other modules of electronic device 412 over bus 418. It should be appreciated that although not shown in FIG. 8, other hardware and/or software modules may be used in conjunction with the electronic device 412, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID devices, tape drives, and data backup storage devices, among others.
The processor 416 executes various functional applications and data processing, such as implementing a merchant classification method provided by embodiments of the present application, by executing at least one of the other programs stored in the memory 428.
One embodiment of the present invention provides a storage medium containing computer-executable instructions which, when executed by a computer processor, perform a merchant classification method, comprising:
inputting a preset number of merchant transaction data acquired in each preset unit time into a recurrent neural network to obtain first output data;
inputting the first output data into a convolutional neural network to obtain second output data;
and processing the second output data based on a normalization function to obtain a merchant classification result.
The computer storage media of the embodiments of the present application may take any combination of one or more computer-readable storage media. The computer readable storage medium may be a computer readable signal storage medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor device, apparatus, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In embodiments of the present application, a computer readable storage medium may be any tangible storage medium that can contain, or store a program for use by or in connection with an instruction execution apparatus, device, or apparatus.
A computer readable signal storage medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal storage medium may also be any computer readable storage medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution apparatus, device, or apparatus.
Program code embodied on a computer readable storage medium may be transmitted using any appropriate storage medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or device. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
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 (15)

1. A merchant classification method, characterized in that the method comprises:
inputting a preset number of merchant transaction data acquired in each preset unit time into a recurrent neural network to obtain first output data;
inputting the first output data into a convolutional neural network to obtain second output data;
and processing the second output data based on a normalization function to obtain a merchant classification result.
2. The method of claim 1, wherein inputting a preset number of merchant transaction data collected in each preset unit time into a recurrent neural network to obtain first output data comprises:
expressing the preset amount of merchant transaction data collected in each preset unit time into a column vector form;
and inputting the preset amount of merchant transaction data collected in the preset unit time in each column vector form into each node of the input layer.
3. The method of claim 1, wherein inputting a preset number of merchant transaction data collected in each preset unit time into a recurrent neural network to obtain first output data comprises:
and inputting the output data of the cyclic neural network of the current layer to the input layer of the next layer of cyclic neural network, and processing the output data by the next layer of cyclic neural network to obtain first output data.
4. The method of claim 3, wherein inputting the output data of the recurrent neural network of the current layer to the input layer of the recurrent neural network of the next layer, and obtaining the first output data through the processing of the recurrent neural network of the next layer, comprises:
dividing output data of a circulating network of a current layer into groups with preset group numbers;
and summarizing the output data of each group, and inputting the data into the input layer nodes of the next layer of the recurrent neural network.
5. The method of claim 4, wherein the step of summarizing the output data of each group and inputting the summarized output data into the input layer nodes of the next layer of recurrent neural network comprises the steps of:
splicing the output data of each group to obtain a spliced input matrix;
and inputting the spliced input matrix into an input layer node of the next layer of the recurrent neural network.
6. The method of claim 1, wherein inputting the first output data into a convolutional neural network, resulting in second output data, comprises:
and inputting the first output data into the convolution layer and the pooling layer for processing to obtain second output data.
7. The method of claim 6, wherein inputting the first output data into the convolutional layer and the pooling layer for processing to obtain the second output data comprises:
processing the first output data based on a first preset convolution kernel to obtain convolution layer output data
And performing dimension reduction processing on the output data of the convolution layer based on a second preset convolution kernel to obtain pooled layer output data serving as second output data.
8. The method of claim 1, wherein inputting the first output data into a convolutional neural network, resulting in second output data, comprises:
and inputting the first output data into the convolution layer, the pooling layer and the full-connection layer for processing to obtain second output data.
9. The method of claim 1, wherein processing the second output data based on a normalization function to obtain a merchant classification result comprises:
representing the second output data as exponential data;
and carrying out normalization processing on the index data to obtain a merchant classification result.
10. The method of claim 9, wherein representing the second output data as exponential data comprises:
to be provided with
Figure FDA0002990949440000021
Representing the second output data; wherein Z isiIs the second output data.
11. The method of claim 9, wherein normalizing the index data to obtain merchant classification result data comprises:
determining the probability of the merchant being classified into a preset merchant list based on the following formula:
Figure FDA0002990949440000031
wherein，ZjAnd K is the number of output nodes.
12. The method of claim 1, wherein the normalization function is a Softmax function.
13. A merchant classifying apparatus, the apparatus comprising:
the first processing module is used for inputting the preset number of merchant transaction data acquired in each preset unit time into the recurrent neural network to obtain first output data;
the second processing module is used for inputting the first output data into a convolutional neural network to obtain second output data;
and the classification module is used for processing the second output data based on the normalization function to obtain a merchant classification result.
14. An electronic device, characterized in that the electronic device comprises:
one or more processors;
a memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the merchant classification method of any of claims 1-12.
15. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the merchant classification method according to any one of claims 1 to 12.
CN202110313676.6A 2021-03-24 2021-03-24 Merchant classification method and device, electronic equipment and medium Pending CN113033658A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110313676.6A CN113033658A (en) 2021-03-24 2021-03-24 Merchant classification method and device, electronic equipment and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110313676.6A CN113033658A (en) 2021-03-24 2021-03-24 Merchant classification method and device, electronic equipment and medium

Publications (1)

Publication Number Publication Date
CN113033658A true CN113033658A (en) 2021-06-25

Family

ID=76473444

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110313676.6A Pending CN113033658A (en) 2021-03-24 2021-03-24 Merchant classification method and device, electronic equipment and medium

Country Status (1)

Country Link
CN (1) CN113033658A (en)

Similar Documents

Publication Publication Date Title
CN106951484B (en) Picture retrieval method and device, computer equipment and computer readable medium
CN108171260B (en) Picture identification method and system
CN112633419B (en) Small sample learning method and device, electronic equipment and storage medium
CN110263821B (en) Training of transaction feature generation model, and method and device for generating transaction features
CN112990294B (en) Training method and device of behavior discrimination model, electronic equipment and storage medium
CN111950279B (en) Entity relationship processing method, device, equipment and computer readable storage medium
CN108229287A (en) Image-recognizing method and device, electronic equipment and computer storage media
CN112613584A (en) Fault diagnosis method, device, equipment and storage medium
CN112988840A (en) Time series prediction method, device, equipment and storage medium
CN111291936B (en) Product life cycle prediction model generation method and device and electronic equipment
CN108647714A (en) Acquisition methods, terminal device and the medium of negative label weight
CN114090601B (en) Data screening method, device, equipment and storage medium
CN109740156B (en) Feedback information processing method and device, electronic equipment and storage medium
CN111488517A (en) Method and device for training click rate estimation model
CN111259975B (en) Method and device for generating classifier and method and device for classifying text
WO2023185125A1 (en) Product resource data processing method and apparatus, electronic device and storage medium
CN113033658A (en) Merchant classification method and device, electronic equipment and medium
CN115017385A (en) Article searching method, device, equipment and storage medium
CN112328899B (en) Information processing method, information processing apparatus, storage medium, and electronic device
CN115049546A (en) Sample data processing method and device, electronic equipment and storage medium
CN114861758A (en) Multi-modal data processing method and device, electronic equipment and readable storage medium
CN110516024B (en) Map search result display method, device, equipment and storage medium
CN110852392A (en) User grouping method, device, equipment and medium
CN114637921B (en) Item recommendation method, device and equipment based on modeling accidental uncertainty
CN115471717B (en) Semi-supervised training and classifying method device, equipment, medium and product of model

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