CN114742648A

CN114742648A - Product pushing method, device, equipment and medium

Info

Publication number: CN114742648A
Application number: CN202210254963.9A
Authority: CN
Inventors: 倪灵
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2022-03-15
Filing date: 2022-03-15
Publication date: 2022-07-12

Abstract

The disclosure provides a product pushing method which can be applied to the technical fields of artificial intelligence and finance. The product pushing method comprises the following steps: acquiring transaction behavior data to be identified, wherein the transaction behavior data comprises transaction data and behavior data; inputting transaction behavior data to be recognized into a pre-trained classification model, and outputting a classification result, wherein the pre-trained classification model is obtained by training based on a preset loss function, and the preset loss function is obtained by constructing according to a preset calculation distance condition; identifying a target classification result from the classification results; and pushing the associated product based on the target classification result. The disclosure also provides a product pushing device, equipment, a storage medium and a program product.

Description

Product pushing method, device, equipment and medium

Technical Field

The present disclosure relates to the field of artificial intelligence and financial technologies, and in particular, to a product pushing method, apparatus, device, medium, and program product.

Background

With the continuous development of the internet, transactions via the internet have been widely used. Various websites, applets and the like have various push advertisements for products, so that users can conveniently trade the products. However, the current push advertisement generally defines part of users by means of rule screening and the like, and part of the advertisement is directly pushed to all users, without considering the actual requirements of the users, which wastes resources that can be pushed on the home page. Therefore, how to solve the problem of the accuracy of the push marketing of the product is very important.

Disclosure of Invention

In view of the above, the present disclosure provides a product push method, apparatus, device, medium, and program product.

According to a first aspect of the present disclosure, there is provided a product pushing method, including:

acquiring transaction behavior data to be identified, wherein the transaction behavior data comprises transaction data and behavior data;

inputting transaction behavior data to be recognized into a pre-trained classification model, and outputting a classification result, wherein the pre-trained classification model is obtained by training based on a preset loss function, and the preset loss function is obtained by constructing according to a preset calculation distance condition;

identifying a target classification result from the classification results; and

and pushing the associated products based on the target classification result.

According to the embodiment of the disclosure, the construction of the preset loss function according to the preset calculation distance condition comprises:

constructing a first loss function aiming at the same classification label according to a preset calculation distance condition;

constructing second loss functions aiming at different classification labels according to a preset calculation distance condition;

and combining the first loss function and the second loss function to obtain a preset loss function.

According to an embodiment of the present disclosure, the preset calculation distance condition includes a mahalanobis distance.

According to the embodiment of the disclosure, the training of the classification model trained in advance based on the preset loss function includes:

classifying historical transaction behavior data which are generated in a preset time interval and are related to a target product according to preset conditions to obtain a plurality of historical transaction behavior data with classification labels, wherein the historical transaction behavior data comprise historical transaction data and historical behavior data;

inputting a plurality of historical transaction behavior data with classification labels into a classification model to be trained, and outputting a training classification result, wherein the classification model to be trained is constructed on the basis of a preset calculation distance condition;

inputting the training classification result and the classification label into a preset loss function, and outputting a loss result;

adjusting model parameters of the classification model to be trained according to the loss result until the loss result or the iteration times of the preset loss function meet the preset condition; and

and taking the classification model obtained when the loss result or the iteration number of the preset loss function meets the preset condition as the trained classification model.

According to an embodiment of the present disclosure, a classification model to be trained includes a calculation module and a classification module; inputting a plurality of historical transaction behavior data with classification labels into a classification model to be trained, and outputting a training classification result comprises the following steps:

inputting a plurality of historical transaction behavior data with classification labels into a calculation module, and calculating the distance between two historical transaction behavior data with classification labels according to a preset calculation distance condition to output a distance calculation result;

and inputting the distance calculation result into a classification module for clustering classification and outputting a training classification result.

According to an embodiment of the present disclosure, the historical transaction data includes: transaction information and/or transaction flow of each channel; the historical behavior data includes web page or application access behavior data.

A second aspect of the present disclosure provides a product pushing device, including:

the system comprises an acquisition module, a recognition module and a recognition module, wherein the acquisition module is used for acquiring transaction behavior data to be recognized, and the transaction behavior data comprises transaction data and behavior data;

the classification module is used for inputting transaction behavior data to be recognized into a pre-trained classification model and outputting a classification result, wherein the pre-trained classification model is obtained based on a preset loss function through training, and the preset loss function is obtained through construction according to a preset calculation distance condition;

the identification module is used for identifying a target classification result from the classification results; and

and the pushing module is used for pushing the associated products based on the target classification result.

A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the product push method described above.

The fourth aspect of the present disclosure also provides a computer-readable storage medium having stored thereon executable instructions, which when executed by a processor, cause the processor to execute the above product pushing method.

A fifth aspect of the present disclosure also provides a computer program product comprising a computer program which, when executed by a processor, implements the above-mentioned product push method.

According to the embodiment of the disclosure, the transaction data and the behavior data are used, and the personal attribute of the user is not involved; and meanwhile, constructing a preset loss function based on a preset calculation distance condition, and training to obtain a pre-trained classification model for identifying transaction behaviors. And pushing the associated products according to the identified target classification result. The method and the device realize accurate pushing of associated products for users entering different channels, and are beneficial to automatic and accurate marketing.

Drawings

The foregoing and other objects, features and advantages of the disclosure will be apparent from the following description of embodiments of the disclosure, which proceeds with reference to the accompanying drawings, in which:

fig. 1 schematically illustrates an application scenario diagram of a product push method, apparatus, device, medium and program product according to an embodiment of the present disclosure;

FIG. 2 schematically illustrates a flow diagram of a product push method according to an embodiment of the disclosure;

FIG. 3 schematically illustrates a flowchart of a method for constructing a predetermined loss function according to a predetermined calculated distance condition according to an embodiment of the present disclosure;

FIG. 4 is a flow chart of a method for training a pre-trained classification model based on a predetermined loss function according to an embodiment of the present disclosure;

FIG. 5 is a flow chart of a method for inputting a plurality of historical transaction behavior data with class labels to a classification model to be trained and outputting a training classification result according to an embodiment of the present disclosure;

fig. 6 schematically shows a block diagram of a structure of a product pushing device according to an embodiment of the present disclosure; and

fig. 7 schematically shows a block diagram of an electronic device adapted for a product push method according to an embodiment of the present disclosure.

Detailed Description

Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is illustrative only and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present disclosure.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.

All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.

Where a convention analogous to "at least one of A, B and C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B and C" would include but not be limited to systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).

In the technical scheme of the disclosure, the acquisition, storage, application and the like of the personal information of the related user all accord with the regulations of related laws and regulations, necessary security measures are taken, and the customs of the public order is not violated.

In the technical scheme of the embodiment of the disclosure, before the personal information of the user is obtained or collected, the authorization or the consent of the user is obtained.

The possibility of purchasing related products by the user can be predicted by using a large number of user personal basic attributes and transaction behaviors, but the prediction accuracy still needs to be improved. Based on this, different from the past, consider that the user has relatively high willingness to consume the related products after certain behaviors occur. Therefore, by introducing a new model, the characteristics of the model only relate to transaction and behavior data, the consumption degree of the product by the user is distinguished, the user with higher degree is selected for accurate pushing, the marketing efficiency of automatic marketing is increased, and the disturbance to the unintended user is reduced.

An embodiment of the present disclosure provides a product pushing method, including: acquiring transaction behavior data to be identified, wherein the transaction behavior data comprises transaction data and behavior data; inputting transaction behavior data to be recognized into a pre-trained classification model, and outputting a classification result, wherein the pre-trained classification model is obtained by training based on a preset loss function, and the preset loss function is obtained by constructing according to a preset calculation distance condition; identifying a target classification result from the classification results; and pushing the associated product based on the target classification result.

Fig. 1 schematically illustrates an application scenario diagram of a product push method, apparatus, device, medium, and program product according to an embodiment of the present disclosure.

As shown in fig. 1, the application scenario 100 according to this embodiment may include

terminal devices

101, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the

terminal devices

101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The user may use the

terminal devices

101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The

terminal devices

101, 102, 103 may have installed thereon various communication client applications, such as a financial product type application, a shopping type application, a web browser application, a search type application, an instant messaging tool, a mailbox client, social platform software, etc. (by way of example only).

The

terminal devices

101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.

The server 105 may be a server providing various services, such as a background management server (for example only) providing support for websites browsed by users using the

terminal devices

101, 102, 103. The backend management server may analyze and process the received data such as the user request, and feed back a processing result (for example, a web page, information, or data obtained or generated according to the user request) to the terminal device.

It should be noted that the product pushing method provided by the embodiment of the present disclosure may be generally executed by the server 105. Accordingly, the product pushing device provided by the embodiments of the present disclosure may be generally disposed in the server 105. The product pushing method provided by the embodiment of the present disclosure may also be performed by a server or a server cluster different from the server 105 and capable of communicating with the

terminal devices

101, 102, 103 and/or the server 105. Correspondingly, the product pushing apparatus provided by the embodiment of the present disclosure may also be disposed in a server or a server cluster different from the server 105 and capable of communicating with the

terminal devices

101, 102, 103 and/or the server 105.

It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.

The product push method of the disclosed embodiment will be described in detail below with fig. 2 to 5 based on the scenario described in fig. 1.

Fig. 2 schematically shows a flow chart of a product push method according to an embodiment of the present disclosure.

As shown in fig. 2, the product push method 200 of this embodiment includes operations S201 to S204.

In operation S201, transaction behavior data to be identified is acquired, wherein the transaction behavior data includes transaction data and behavior data.

According to the embodiment of the disclosure, the transaction behavior data to be identified can be transaction information generated by personal consumption of credit loan products by a plurality of user banks and behavior data of accessing bank APP; or transaction information generated when a plurality of users make a purchase of a financial product and behavior data of accessing a web page. But also transaction data and behavior data generated by the consumption of other physical products by a plurality of users.

For example, for a bank's personal home consumption loan product, transaction information and behavior data generated after a transaction is performed for the product user may be obtained.

In operation S202, transaction behavior data to be recognized is input to a pre-trained classification model, and a classification result is output, where the pre-trained classification model is obtained by training based on a preset loss function, and the preset loss function is constructed according to a preset calculation distance condition.

According to the embodiment of the disclosure, the pre-trained classification model is obtained by training based on the preset loss function, and can be used for inputting the user transaction data and the behavior data and then outputting the classification result. The preset calculated distance condition may include a mahalanobis distance.

For example, transaction information and behavior data generated after a user who consumes loan products in a bank's personal home is transacted are input into a pre-trained classification model, the user who consumes the products can be classified into three types according to time periods, and classification results of the corresponding data corresponding to the user are output.

In operation S203, a target classification result is identified from the classification results.

According to the embodiment of the disclosure, one of the classification results can be identified as a target classification result.

For example, the users who recognize the personal home consumption loan products for the consumption banks are classified into three types according to time periods, and one of the classification results of the users corresponding to the corresponding data is output as the target classification result.

In operation S204, the associated product is pushed based on the target classification result.

According to the embodiment of the disclosure, the related products are pushed to the user corresponding to the target classification result.

For example, for users who consume loan products at home by the personal of a consumption bank, some consumption loan products are pushed to the users according to the target classification result. And the push is not carried out on other types of users, so that the disturbance on the users is reduced.

According to the embodiment of the disclosure, the data used are transaction data and behavior data, and the personal attributes of the user are not involved; and meanwhile, constructing a preset loss function based on a preset calculation distance condition, and training to obtain a classification model trained in advance for identifying transaction behaviors. And pushing the associated products according to the identified target classification result. The method and the device realize accurate pushing of associated products for users entering different channels, and are beneficial to automatic and accurate marketing.

Fig. 3 schematically illustrates a flowchart of a method for determining a preset loss function according to a preset calculated distance condition according to an embodiment of the present disclosure.

As shown in fig. 3, the method 300 for constructing the preset loss function according to the preset calculated distance condition in this embodiment includes operations S301 to S303.

In operation S301, a first loss function for the same classification label is constructed according to a preset calculation distance condition.

According to the embodiment of the disclosure, according to the preset calculation distance condition, for any historical trading behavior data, the difference between the absolute distance value of the historical trading behavior data with different two tags and the absolute distance value of the historical trading behavior data with the same two tags is greater than or equal to 1. A first penalty function for the same class label is defined accordingly.

For example, according to the preset calculation distance condition, for any three historical transaction behavior data

And

the labels are identical, i.e. yi ═ yj, and

and

if the labels are different, i.e. yi ≠ ym, the following formula (1) should be satisfied:

defining a first loss function epsilon for the same class label according to equation (1)_pullIt can be shown that the distance between the historical transaction behavior data of the same category label should be as close to 1 as possible, the first loss function epsilon_pullAs shown in formula (2):

in operation S302, a second loss function for different classification labels is constructed according to a preset calculation distance condition.

According to the embodiment of the disclosure, according to the preset calculation distance condition, for any historical trading behavior data, the difference between the absolute distance value of the historical trading behavior data with different two tags and the absolute distance value of the historical trading behavior data with the same two tags is greater than or equal to 1. A second penalty function for different class labels is defined accordingly.

For example, a second loss function ε for different class labels is defined according to equation (1) above_pushIt can be shown that the distance between the historical transaction behavior data of different category labels should be as close to 0 as possible, the second loss function epsilon_pushAs shown in formula (3):

in operation S303, the first loss function and the second loss function are combined to obtain a preset loss function.

According to the embodiment of the disclosure, the first loss function and the second loss function can be combined by bringing parameters of the preset loss function into the first loss function and the second loss function, so that the preset loss function is obtained.

For example, the combined predetermined loss function can be expressed as formula (4):

ε(L)＝(1-μ)ε_pull(L)+με_push(L) (4)

the value range of the parameter μ of the preset loss function can be 0-1, and is preferably 0.5.

According to the embodiment of the disclosure, the distance of the same classification label is minimized and the distance of different classification labels is maximized by presetting a calculation distance condition, so that the accuracy of the trained model is relatively high.

Fig. 4 schematically shows a flowchart of a method for training a pre-trained classification model based on a preset loss function according to an embodiment of the present disclosure.

As shown in fig. 4, the method 400 for training the pre-trained classification model of this embodiment based on the preset loss function includes operations S401 to S405.

In operation S401, historical transaction behavior data related to a target product, which is generated within a predetermined time interval, is classified according to a preset condition, so as to obtain a plurality of historical transaction behavior data with classification tags, where the historical transaction behavior data includes historical transaction data and historical behavior data.

According to embodiments of the present disclosure, the historical transaction data may include: the transaction information and/or transaction flow of each channel may be, for example, transaction information and/or transaction flow generated by a mobile internet bank APP, or transaction information and/or transaction flow generated by a browser webpage. The historical behavior data may include web page or application access behavior data.

According to the embodiment of the disclosure, transaction information and/or transaction flow of each channel and webpage or application program access behavior data related to the target product generated in a preset time interval are subjected to time series aggregation, bucket division and other processing. The classification processing according to the preset condition may be classification processing according to the transaction time with the target product, so as to obtain a plurality of historical transaction behavior data with classification tags. The predetermined time interval may be a history for a certain period of time. The target product can be a kind of financial product or other entity products.

For example, after transaction information and/or transaction flow and webpage or application access behavior data generated by all users on a certain type of financial products in mobile phone internet banking APP and through browser webpages in the last two years are subjected to time series aggregation, barreling and the like, the last two years are divided into three time periods according to the transaction time on the certain type of financial products, and three classifications are performed to obtain the transaction information and/or transaction flow and webpage or application access behavior data with three classification labels.

In operation S402, a plurality of historical transaction behavior data with classification labels are input to a classification model to be trained, and a training classification result is output, where the classification model to be trained is constructed based on a preset calculation distance condition.

According to the embodiment of the disclosure, the classification model to be trained is constructed based on the Mahalanobis distance, wherein the Mahalanobis distance is the distance between one point and one subsection, the similarity of two unknown sample sets can be effectively calculated, the relation of various characteristics is considered, and the method is irrelevant to the measurement scale.

In operation S403, the training classification result and the classification label are input into a preset loss function, and a loss result is output, where the preset loss function is determined according to a preset calculation distance condition.

According to an embodiment of the present disclosure, the preset loss function may be determined according to mahalanobis distance. The distance between the two historical transaction behavior data can be calculated by using the mahalanobis distance, and the same classification label can be calculated as 1 according to the classification labels of the two historical transaction behavior data, and the different classification labels can be calculated as 0. And determining a preset loss function according to the condition.

For example, the distance between any two historical trading behavior data is represented by yjl, and then the following equation (5) is satisfied:

where yi-yj represents that the classification labels of the two historical transaction behavior data are the same.

In operation S404, the model parameter of the classification model to be trained is adjusted according to the loss result until the loss result or the iteration number of the preset loss function satisfies the preset condition.

According to the embodiment of the disclosure, the model parameters of the classification model to be trained can be adjusted by using a gradient descent algorithm according to the loss result.

In operation S405, a classification model obtained when the loss result or the iteration number of the preset loss function satisfies the preset condition is used as a trained classification model.

According to the embodiment of the disclosure, a model obtained when the preset loss function is converged can be used as a trained classification model, and a model obtained when the iteration times meet the preset conditions can be used as a trained classification model.

According to the embodiment of the disclosure, data used for model training are user transaction data and behavior data, and do not relate to user personal attributes; meanwhile, a classification model to be trained is constructed based on a preset calculation distance condition, so that the distance of the same classification label can be minimized, and the distance of different classification labels can be maximized. The classification model trained by the model training method can be used for identifying target classification results, and related products can be pushed accurately to users entering from different channels according to the target classification results, so that automatic and accurate marketing can be realized.

Fig. 5 schematically shows a flowchart of a method for inputting a plurality of historical transaction behavior data with class labels to a classification model to be trained and outputting a training classification result according to an embodiment of the present disclosure.

As shown in fig. 5, the method 500 for inputting a plurality of historical transaction behavior data with class labels to a classification model to be trained and outputting a training classification result according to this embodiment includes operations S501 to S502.

In operation S501, a plurality of historical transaction behavior data with classification tags are input to a calculation module, which is used to calculate a distance between two historical transaction behavior data with classification tags according to a preset calculation distance condition, and output a distance calculation result.

According to the embodiment of the disclosure, the preset calculation distance condition may be mahalanobis distance, and each two pieces of historical transaction behavior data with the same classification tag are calculated as 1, and each two pieces of historical transaction behavior data with different classification tags are calculated as 0.

For example, a user who has historical transaction behavior of a financial product according to historical transaction behavior data of the user for the financial product in the last three months may be recorded as a class a, a user who has the historical transaction behavior of the product in the last three months to two years may be recorded as a class B, and a user who has no historical transaction behavior of the product in the last two years may be recorded as a class C. Calculating the distance between two historical transaction behavior data with the A-type label, the C-type label or the B-type label as 1, and calculating the distance between the two historical transaction behavior data of the A-type label and the C-type label, or the A-type label and the B-type label, or the B-type label and the C-type label as 0.

In operation S502, the distance calculation result is input to the classification module for cluster classification and a training classification result is output.

According to an embodiment of the present disclosure, the classification module may be a cluster classification module. The training classification result may be a training classification result of a user corresponding to the historical transaction behavior data.

According to the embodiment of the disclosure, the distance between two historical transaction behavior data is measured, the distance between the same classification label is minimized, and the distance between different classification labels is maximized, so that accurate classification is performed.

Based on the product pushing method, the disclosure also provides a product pushing device. The apparatus will be described in detail below with reference to fig. 6.

Fig. 6 schematically shows a block diagram of a product pushing device according to an embodiment of the present disclosure.

As shown in fig. 6, the product pushing apparatus 600 of this embodiment includes an obtaining module 610, a classifying module 620, an identifying module 630, and a pushing module 640.

The obtaining module 610 is configured to obtain transaction behavior data to be identified, where the transaction behavior data includes transaction data and behavior data. In an embodiment, the obtaining module 610 may be configured to perform the operation S201 described above, which is not described herein again.

The classification module 620 is configured to input transaction behavior data to be recognized into a pre-trained classification model, and output a classification result, where the pre-trained classification model is obtained by training based on a preset loss function, and the preset loss function is obtained by constructing according to a preset calculation distance condition. In an embodiment, the classification module 620 may be configured to perform the operation S202 described above, which is not described herein again.

The recognition module 630 is used to recognize the target classification result from the classification results. In an embodiment, the identifying module 630 may be configured to perform the operation S203 described above, which is not described herein again.

The pushing module 640 is configured to push the associated product based on the target classification result. In an embodiment, the pushing module 640 may be configured to perform the operation S204 described above, which is not described herein again.

According to the embodiment of the present disclosure, any plurality of the obtaining module 610, the classifying module 620, the identifying module 630 and the pushing module 640 may be combined into one module to be implemented, or any one of the modules may be split into a plurality of modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of the other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the obtaining module 610, the classifying module 620, the identifying module 630, and the pushing module 640 may be implemented at least partially as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented by hardware or firmware in any other reasonable manner of integrating or packaging a circuit, or implemented by any one of three implementations of software, hardware, and firmware, or any suitable combination of any of them. Alternatively, at least one of the obtaining module 610, the classifying module 620, the identifying module 630 and the pushing module 640 may be at least partially implemented as a computer program module, which when executed may perform a corresponding function.

Fig. 7 schematically shows a block diagram of an electronic device adapted to implement the product push method according to an embodiment of the present disclosure.

As shown in fig. 7, an electronic device 700 according to an embodiment of the present disclosure includes a processor 701, which can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)702 or a program loaded from a storage section 708 into a Random Access Memory (RAM) 703. The processor 701 may include, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor and/or associated chipset, and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), among others. The processor 701 may also include on-board memory for caching purposes. The processor 701 may comprise a single processing unit or a plurality of processing units for performing the different actions of the method flows according to embodiments of the present disclosure.

In the RAM 703, various programs and data necessary for the operation of the electronic apparatus 700 are stored. The processor 701, the ROM702, and the RAM 703 are connected to each other by a bus 704. The processor 701 performs various operations of the method flows according to the embodiments of the present disclosure by executing programs in the ROM702 and/or the RAM 703. Note that the programs may also be stored in one or more memories other than the ROM702 and the RAM 703. The processor 701 may also perform various operations of method flows according to embodiments of the present disclosure by executing programs stored in the one or more memories.

Electronic device 700 may also include input/output (I/O) interface 705, which input/output (I/O) interface 705 is also connected to bus 704, according to an embodiment of the present disclosure. The electronic device 700 may also include one or more of the following components connected to the I/O interface 705: an input portion 706 including a keyboard, a mouse, and the like; an output section 707 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 708 including a hard disk and the like; and a communication section 709 including a network interface card such as a LAN card, a modem, or the like. The communication section 709 performs communication processing via a network such as the internet. A drive 710 is also connected to the I/O interface 705 as needed. A removable medium 711 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 710 as necessary, so that a computer program read out therefrom is mounted into the storage section 708 as necessary.

The present disclosure also provides a computer-readable storage medium, which may be contained in the apparatus/device/system described in the above embodiments; or may exist separately and not be assembled into the device/apparatus/system. The computer-readable storage medium carries one or more programs which, when executed, implement the method according to an embodiment of the disclosure.

According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example but is not limited to: 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), 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 the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present disclosure, a computer-readable storage medium may include the above-described ROM702 and/or RAM 703 and/or one or more memories other than the ROM702 and RAM 703.

Embodiments of the present disclosure also include a computer program product comprising a computer program containing program code for performing the method illustrated in the flow chart. When the computer program product runs in a computer system, the program code is used for causing the computer system to realize the product pushing method provided by the embodiment of the disclosure.

The computer program performs the above-described functions defined in the system/apparatus of the embodiments of the present disclosure when executed by the processor 701. The systems, apparatuses, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the present disclosure.

In one embodiment, the computer program may be hosted on a tangible storage medium such as an optical storage device, a magnetic storage device, or the like. In another embodiment, the computer program may also be transmitted in the form of a signal on a network medium, distributed, downloaded and installed via the communication section 709, and/or installed from the removable medium 711. The computer program containing program code may be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 709, and/or installed from the removable medium 711. The computer program, when executed by the processor 701, performs the above-described functions defined in the system of the embodiment of the present disclosure. The systems, devices, apparatuses, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the present disclosure.

In accordance with embodiments of the present disclosure, program code for executing computer programs provided by embodiments of the present disclosure may be written in any combination of one or more programming languages, and in particular, these computer programs may be implemented using high level procedural and/or object oriented programming languages, and/or assembly/machine languages. The programming language includes, but is not limited to, programming languages such as Java, C + +, python, the "C" language, or the like. The program code may execute entirely on the user's computing device, partly on the user's device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

Those skilled in the art will appreciate that various combinations and/or combinations of features recited in the various embodiments and/or claims of the present disclosure can be made, even if such combinations or combinations are not expressly recited in the present disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments and/or claims of the present disclosure may be made without departing from the spirit or teaching of the present disclosure. All such combinations and/or associations are within the scope of the present disclosure.

The embodiments of the present disclosure are described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described separately above, this does not mean that the measures in the embodiments cannot be used in advantageous combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be devised by those skilled in the art without departing from the scope of the present disclosure, and such alternatives and modifications are intended to be within the scope of the present disclosure.

Claims

1. A product push method, comprising:

inputting the transaction behavior data to be recognized into a pre-trained classification model, and outputting a classification result, wherein the pre-trained classification model is obtained based on a preset loss function through training, and the preset loss function is obtained through construction according to a preset calculation distance condition; and

identifying a target classification result from the classification results;

and pushing the associated products based on the target classification result.

2. The method of claim 1, wherein the predetermined loss function is constructed according to a predetermined calculation distance condition and comprises:

constructing a first loss function aiming at the same classification label according to the preset calculation distance condition;

constructing second loss functions aiming at different classification labels according to the preset calculation distance condition;

and combining the first loss function and the second loss function to obtain the preset loss function.

3. The method of claim 1, wherein the preset calculated distance condition comprises a mahalanobis distance.

4. The method of claim 1, wherein the pre-trained classification model is trained based on a predetermined loss function and comprises:

inputting the historical transaction behavior data with the classification labels into a classification model to be trained, and outputting a training classification result, wherein the classification model to be trained is constructed on the basis of the preset calculation distance condition;

inputting the training classification result and the classification label into the preset loss function, and outputting a loss result;

adjusting model parameters of the classification model to be trained according to the loss result until the loss result or the iteration times of the preset loss function meet a preset condition; and

and taking the classification model obtained when the loss result or the iteration times of the preset loss function meet the preset condition as the trained classification model.

5. The method of claim 4, wherein the classification model to be trained comprises a calculation module and a classification module; the inputting the historical transaction behavior data with the classification labels into a classification model to be trained, and outputting a training classification result comprises:

inputting the historical transaction behavior data with the classification labels into the calculation module, and calculating the distance between the two historical transaction behavior data with the classification labels according to the preset calculation distance condition, and outputting a distance calculation result;

and inputting the distance calculation result into the classification module for clustering classification and outputting the training classification result.

6. The method of claim 4, wherein the historical transaction data comprises: transaction information and/or transaction flow of each channel; the historical behavior data comprises web page or application access behavior data.

7. A product pusher device comprising:

the system comprises an acquisition module, a recognition module and a processing module, wherein the acquisition module is used for acquiring transaction behavior data to be recognized, and the transaction behavior data comprises transaction data and behavior data;

the classification module is used for inputting the transaction behavior data to be recognized into a pre-trained classification model and outputting a classification result, wherein the pre-trained classification model is obtained by training based on a preset loss function, and the preset loss function is obtained by constructing according to a preset calculation distance condition;

8. An electronic device, comprising:

one or more processors;

a storage device for storing one or more programs,

wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-6.

9. A computer readable storage medium having stored thereon executable instructions which, when executed by a processor, cause the processor to perform the method of any one of claims 1 to 6.

10. A computer program product comprising a computer program which, when executed by a processor, implements a method according to any one of claims 1 to 6.