CN116228301A

CN116228301A - Method, device, equipment and medium for determining a target user

Info

Publication number: CN116228301A
Application number: CN202310279687.6A
Authority: CN
Inventors: 王利强; 费闯; 沈乐
Original assignee: Agricultural Bank of China
Current assignee: Agricultural Bank of China
Priority date: 2023-03-21
Filing date: 2023-03-21
Publication date: 2023-06-06

Abstract

The invention discloses a method, a device, equipment and a medium for determining a target user. The method comprises the following steps: acquiring attribute characteristic information and product sales data of a product to be recommended; obtaining associated sample data related to the product to be recommended according to the attribute characteristic information and the trained product classification model; determining a target user portrait matched with the product to be recommended according to the associated sample data and the product sales data; and determining a target recommended user of the product to be recommended according to the target user portrait and the established user list. A user portrayal model is constructed by determining associated sample data of a product to be recommended and combining product sales data of the product to be recommended so as to determine a target user portrayal, and users matched with the target user portrayal are determined in a large number of users. The method and the device realize the accurate determination of the target user of the product to be recommended, solve the problem that the target user cannot be accurately determined due to the fact that the available data of the new product is less in the early stage, and reduce the popularization difficulty of the new product.

Description

Method, device, equipment and medium for determining target user

Technical Field

The present invention relates to the field of computer data processing technologies, and in particular, to a method, an apparatus, a device, and a medium for determining a target user.

Background

The existing financial products are various in variety and quantity, and determining the corresponding marketable customer groups for different financial products is important.

In the prior art, sales samples of financial products are analyzed through machine learning, and marketable customer groups of the products are determined, so that accurate marketing to customers is realized.

However, financial products are a class of products that change very rapidly. For newly added financial products, the sales data of the early-stage products are less, and the samples are sparse, so that the problems that potential clients cannot be accurately determined and popularization is difficult exist.

Disclosure of Invention

The invention provides a method, a device, equipment and a medium for determining a target user, which are used for realizing automatic determination of the target user of a new product.

According to a first aspect of the present invention, there is provided a method for determining a target user, comprising:

acquiring attribute characteristic information and product sales data of a product to be recommended;

obtaining associated sample data related to the product to be recommended according to the attribute characteristic information and the trained product classification model;

determining a target user portrait matched with the product to be recommended according to the association sample data and the product sales data;

And determining the target recommended user of the product to be recommended according to the target user portrait and the established user list.

According to a second aspect of the present invention, there is provided a target user determination apparatus comprising:

the first acquisition module is used for acquiring attribute characteristic information and product sales data of the products to be recommended;

the second acquisition module is used for acquiring associated sample data related to the product to be recommended according to the attribute characteristic information and the trained product classification model;

the user portrait determining module is used for determining a target user portrait matched with the product to be recommended according to the associated sample data and the product sales data;

and the target user determining module is used for determining a target recommended user of the product to be recommended according to the target user portrait and the established user list.

According to a third aspect of the present invention, there is provided an electronic device comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,,

the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of determining a target user according to any one of the embodiments of the present invention.

According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute a method for determining a target user according to any embodiment of the present invention.

According to the technical scheme, the attribute characteristic information and the product sales data of the product to be recommended are obtained; obtaining associated sample data related to the product to be recommended according to the attribute characteristic information and the trained product classification model; determining a target user portrait matched with the product to be recommended according to the associated sample data and the product sales data; and determining a target recommended user of the product to be recommended according to the target user portrait and the established user list. A user portrayal model is constructed by determining associated sample data of a product to be recommended and combining product sales data of the product to be recommended so as to determine a target user portrayal, and users matched with the target user portrayal are determined in a large number of users. The method and the device realize the accurate determination of the target user of the product to be recommended, solve the problem that the target user cannot be accurately determined due to the fact that the available data of the new product is less in the early stage, and reduce the popularization difficulty of the new product.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of a method for determining a target user according to a first embodiment of the present invention;

fig. 2 is a flowchart of a method for determining a target user according to a second embodiment of the present invention;

fig. 3 is a schematic structural diagram of a determining device for a target user according to a third embodiment of the present invention;

fig. 4 is a schematic structural diagram of an electronic device implementing an embodiment of the present invention.

Detailed Description

In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Example 1

Fig. 1 is a flowchart of a method for determining a target user according to an embodiment of the present invention, where the method may be performed by a determining device of a target user, the determining device of the target user may be implemented in hardware and/or software, and the determining device of the target user may be configured in an electronic device. As shown in fig. 1, the method includes:

S110, acquiring attribute characteristic information of the product to be recommended and product sales data.

In this embodiment, the product to be recommended can be understood as a new product with less sales data. Attribute feature information may be understood as information used to summarize product features. Product sales data may be understood to include data including characteristics of users of product sales, sales volume, etc.

Specifically, the product to be recommended generally comprises an attribute feature text, the processor can acquire attribute feature information of the product to be recommended by identifying text content corresponding to the attribute feature text, the processor can acquire historical sales conditions of the product to be recommended according to a set period, and statistics is carried out on the historical sales conditions according to a set processing mode to obtain product sales data.

S120, according to the attribute characteristic information and the trained product classification model, obtaining associated sample data related to the product to be recommended.

In this embodiment, the product classification model may be understood as a model for determining a category to which a product belongs, such as: classification model of KNN neighbor algorithm, neural network, biLSTM class classification model. The associated sample data may be understood as sales data and attribute feature information of products belonging to the same category as the product to be recommended.

Specifically, the processor may input attribute feature information of the product to be recommended into a trained product classification model, where the product classification model is obtained after training attribute feature samples of other products, and determines a classification to which the product to be recommended belongs according to a classification result output by the product classification model, so as to determine other related products under the classification, that is, the related products are most similar to the product to be recommended, and the processor may query the memory and obtain related sample data corresponding to the related products.

S130, determining a target user portrait matched with the product to be recommended according to the associated sample data and the product sales data.

In this embodiment, the target user image may be understood as a labeled user model abstracted by information such as user attributes, user preferences, lifestyle, user behaviors, and the like.

Specifically, the processor can perform model training by taking the related sample data and the product sales data as samples, wherein the sample data is attribute characteristic information of the product, the sample label is a user portrait, multiple rounds of sampling are performed according to different sampling probabilities, each round of sampling is used for one round of model training to obtain multiple models, the multiple models are combined to determine a final model, the attribute characteristic information of the product to be recommended is input into the final model, and the obtained result is a target user portrait matched with the product to be recommended.

And S140, determining a target recommended user of the product to be recommended according to the target user portrait and the established user list.

In this embodiment, the user list may be understood as a list including a number of already determined users. The target recommending user may be understood as the user that most matches the product to be recommended.

Specifically, the processor can acquire an established user list through a database or a memory and the like, compare user description information of the target user portrait with attribute information of each user in the user list, and determine the user which is most matched with the target user portrait as a target recommended user of the product to be recommended.

As a first alternative embodiment of the first embodiment, on the basis of the above embodiment, further optimization may include:

and importing the attribute characteristic information into machine classification learning, and correcting the product classification model to obtain a corrected product classification model.

Specifically, if the product to be recommended may be the product to be recommended for the first time or the attribute feature information is updated, the processor may guide the attribute feature information of the product to be recommended into machine classification learning, and correct the existing product classification model to obtain a corrected product classification model.

In the first alternative embodiment of the present embodiment, the attribute feature information is imported into machine classification learning to correct the product classification model, so that dynamic update of the product classification model is achieved, and accuracy of classification results is further ensured.

Example two

Fig. 2 is a flowchart of a method for determining a target user according to a second embodiment of the present invention, where the method is further optimized for the foregoing embodiment. As shown in fig. 2, the method includes:

s210, acquiring attribute characteristic information of the product to be recommended and product sales data.

S220, according to the attribute characteristic information and the trained product classification model, obtaining associated sample data related to the product to be recommended.

S230, constructing an auxiliary domain sample set based on the associated sample data, and constructing a target domain sample set based on the product sales data.

In this embodiment, the auxiliary domain sample set may be understood as a sample set composed of a large number of associated sample data of an associated product. The target domain sample set may be understood as a sample set of a small amount of product sales data of a product to be recommended.

Specifically, the processor may construct the auxiliary domain sample set in a set format based on the associated sample data and the target domain sample set in a set format based on the product sales data.

S240, constructing a user portrait model according to the auxiliary domain sample set and the target domain sample set.

In this embodiment, the user portrayal model can be understood as a model for determining a user portrayal.

Specifically, the processor can sample in the auxiliary domain sample set and the target domain sample set according to different sampling probabilities respectively to obtain sampled samples, sample data is attribute characteristic information of a product, sample labels are user portraits, machine learning is performed based on the sampled samples, and a user portrait model is constructed.

S250, inputting the product sales data into a user portrait model to obtain a target user portrait matched with the product to be recommended.

Specifically, the processor can input the product sales data into the user portrait model to obtain an output result of the user portrait model, namely, a target user portrait matched with the product to be recommended.

S260, obtaining the importance degree duty ratio corresponding to each attribute item in the user description information associated with the target user image.

In this embodiment, the user description information may be understood as information characterizing the user's characteristics. The attribute items may be items for determining attributes of the user, such as work information, demand information, and the like of the user. The importance level duty cycle may reflect the criticality of each attribute item relative to determining the target user.

Specifically, the processor may search the memory for a key identifier corresponding to the target user image, and obtain, through an association relationship between the user image and the user description information, the user description information associated with the target user image and an importance degree ratio corresponding to each attribute item included in the user description information, where the user description information and the importance degree ratio may be preset when the user portrait is constructed, and establish the association relationship between the user description information and the corresponding user portrait.

Illustratively, the user description information of the target user portrait 1 may include: the method comprises the steps of user demand A, user risk bearing B and user input amount C, wherein the importance degree of the user demand is 60%, the importance degree of the user risk bearing is 20% and the importance degree of the user input amount is 20%. The user description information of the target user portrait 2 may include: the method comprises the steps of user requirement E, user risk bearing F and user input amount C, wherein the importance degree of the user requirement is 40%, the importance degree of the user risk bearing is 50% and the importance degree of the user input amount is 10%.

S270, according to the user description information and the importance degree duty ratio, determining similarity values of each user and the target user portrait in the user list.

In this embodiment, the similarity value may be understood as representing the matching degree of each user and the target user portrait by a numerical form.

Specifically, the processor may determine, according to the user description information and the importance ratio of the target user portrait, a similarity value between each user and the target user portrait in the user list by comparing the user description information and the importance ratio of each user in the user list.

Illustratively, the user description information of the target user portrait 1 may include: the method comprises the steps of user demand A, user risk bearing B and user input amount C, wherein the importance degree of the user demand is 60%, the importance degree of the user risk bearing is 20% and the importance degree of the user input amount is 20%. The user list includes: user 1 and user 2, wherein the user description information of user 1 is user requirement A, user risk bearing F and user input amount I, and the similarity value of user 1 is 0.6; the user description information of the user 2 is the user requirement E, the user risk bearing B and the user input amount C, and the similarity value of the user 2 is 0.4.

S280, sorting the similarity degree values according to a set sorting mode to obtain a candidate user list.

In this embodiment, the sorting manner may be in order of from small to large values or from large to small values. The candidate user list may be understood as a list of users ordered by similarity value.

Specifically, the processor may sort the corresponding users according to the order of the similarity values from large to small or from small to large, to obtain a candidate user list.

S290, selecting the user with the set ranking range from the candidate user list as a target recommendation user of the product to be recommended.

In this embodiment, the setting ranking range may be understood as the number of the selected target recommended users, which may be 1-10, 1-5, or the like.

Specifically, the processor may select, according to the set ranking range, a corresponding user in the candidate user list as a target recommended user of the product to be recommended.

Further, on the basis of the above embodiment, the step of constructing the user portrait model according to the auxiliary domain sample set and the target domain sample set may further include:

a1, acquiring preset initial sampling probability and step length information.

In this embodiment, the initial sampling probability may be understood as a preset minimum or maximum sampling probability. The step size information may be understood as a step size each time the sampling probability is adjusted.

Specifically, the processor may obtain the preset initial sampling probability and step size information from the memory.

b1, sampling is carried out in the auxiliary domain sample set and the target domain sample set according to the initial sampling probability, and a first auxiliary domain sample and a first target domain sample are obtained.

Specifically, the processor may sample in the auxiliary domain sample set and the target domain sample set according to the initial sampling probability to obtain a first auxiliary domain sample and a first target domain sample. The initial sampling probabilities for the auxiliary domain sample set and the target domain sample set may be the same or different.

And c1, training the initial user portrait model according to the first auxiliary domain sample and the first target domain sample, and determining a first sub-user portrait model.

In this embodiment, the first sub-user portrayal model may be understood as a user portrayal model trained from a first auxiliary domain sample and a first target domain sample.

Specifically, the processor performs machine learning based on the first auxiliary domain sample and the first target domain sample, sample data is attribute characteristic information of the product, a sample label is a user image, and a first sub-user image model is constructed.

d1, adjusting the initial sampling probability according to the step length information to obtain a sampling probability list.

In this embodiment, the sampling probability list may be understood as a list comprising a plurality of different sampling probabilities.

Specifically, the processor may superimpose or reduce the initial sampling probability according to the step size information to obtain a first sampling probability, and superimpose or reduce the first sampling probability according to the step size information to obtain a second sampling probability, and so on, until the number of the sampling probability list is satisfied, sequentially adding the first sampling probability to the nth sampling probability to the sampling probability list.

And e1, determining a second sub-user portrait model set based on the auxiliary domain sample set, the target domain sample set and the sampling probability list.

In this embodiment, the second set of user portrayal models may be understood as a set of a plurality of second user portrayal models.

Specifically, the processor may sequentially sample from the auxiliary domain sample set and the target domain sample set based on each sampling probability included in the sampling probability list, sequentially perform machine learning on the sampled samples, wherein the sample data is attribute feature information of the product itself, the sample tag is a user portrait, sequentially construct a second sub-user portrait model, and form a second sub-user portrait model set according to all second sub-user portrait models.

f1, constructing a user portrait model according to the first sub-user portrait model and each second user portrait model included in the second user portrait model set.

For example, the processor may evaluate the accuracy of each second user portrait model included in the first sub-user portrait model and the second user portrait model set, and select the model with the highest evaluation as the user portrait model.

The step of obtaining the sampling probability list may further include:

d11, taking the initial sampling probability as the last sampling probability.

And d12, superposing the last sampling probability according to the step length information to obtain the current sampling probability, and determining the current superposition times.

In this embodiment, the last sampling probability may be understood as a sampling probability obtained by last superposition, and is an initial sampling probability for the first time.

Specifically, the processor may acquire a previous sampling probability, superimpose the previous sampling probability according to the step size information to obtain a current sampling probability, acquire a previous superimposition frequency, and determine a previous superimposition frequency plus one to obtain a current superimposition frequency.

d13, judging whether the current superposition times reach a set time threshold, if so, determining a sampling probability list according to the obtained current sampling probabilities.

Specifically, the processor may compare the current stacking times with a set time threshold, and when the current stacking times are the same as the set time threshold, the processor may put each current sampling probability into the sampling probability list according to a determining order of the current sampling probabilities.

And d14, if not, returning to the step of determining the current sampling probability.

Specifically, the processor may compare the current stacking times with the set times threshold, and when the current stacking times do not reach the set times threshold, continue stacking the previous sampling probability according to the step size information to obtain the current sampling probability, and add one to the current stacking times.

Wherein the step of determining the second sub-user portrayal model set based on the auxiliary domain sample set, the target domain sample set and the sampling probability list may further be optimized comprising:

and e11, sampling the auxiliary domain sample set and the target domain sample set according to the sampling probability aiming at each sampling probability included in the sampling probability list to obtain a second auxiliary domain sample and a second target domain sample.

Specifically, the processor may sequentially obtain sampling probabilities from the sampling probability list, and sample the auxiliary domain sample set and the target domain sample set according to the sampling probabilities to obtain a second auxiliary domain sample and a second target domain sample.

e12, training the initial user portrait model based on the second auxiliary domain sample and the second target domain sample to determine a second sub-user portrait model.

In this embodiment, the second sub-user portrayal model may be understood as a user portrayal model trained from the second auxiliary domain sample and the second target domain sample.

Specifically, the processor performs machine learning based on the second auxiliary domain sample and the second target domain sample, sample data is attribute characteristic information of the product, sample labels are user images, and a second sub-user image model is constructed.

e13, forming a second sub-user portrayal model set according to each second sub-user portrayal model.

Specifically, the processor may sequentially sort the second sub-user portrayal models according to the construction order of the second sub-user portrayal models to form a second sub-user portrayal model set.

According to the technical scheme, the associated product which is most similar to the product to be recommended is determined through the product classification model, corresponding associated sample data is obtained, an auxiliary domain sample set and a target domain sample set are constructed according to the associated sample data and the product sales data, different sampling probabilities can be automatically obtained according to initial sampling probability and step length information, the auxiliary domain sample set and the target domain sample set are sampled according to the different sampling probabilities, automatic determination of the sampling probabilities and automatic sampling of different samples are achieved, a plurality of sub-user image models are determined according to each sample obtained through sampling, further, a user image model is determined through the plurality of sub-user image models, accuracy of the user image model is guaranteed, attribute characteristic information of the product to be recommended is input into the user image model to obtain a target user image, and a target user is determined in a user list according to user description information associated with the target user image and importance degree proportion corresponding to each attribute item. The method and the device realize automatic determination of the products related to the products to be recommended, ensure the accuracy of the determination of the related products, solve the problem that the available data of the new products are less in the early stage, realize the accurate determination of the target users of the products to be recommended, and further reduce the popularization difficulty of the new products.

Example III

Fig. 3 is a schematic structural diagram of a determining device for a target user according to a third embodiment of the present invention. As shown in fig. 3, the apparatus includes: a first acquisition module 31, a second acquisition module 32, a user representation determination module 33 and a target user determination module 34. Wherein,,

a first obtaining module 31, configured to obtain attribute feature information of a product to be recommended and product sales data;

a second obtaining module 32, configured to obtain, according to the attribute feature information and the trained product classification model, associated sample data related to the product to be recommended;

a user representation determining module 33, configured to determine a target user representation matched with the product to be recommended according to the association sample data and the product sales data;

and the target user determining module 34 is configured to determine a target recommended user of the product to be recommended according to the target user portrait and the established user list.

Optionally, the user portrait determination module 33 includes:

the sample set construction unit is used for constructing an auxiliary domain sample set based on the associated sample data and constructing a target domain sample set based on the product sales data;

the model construction unit is used for constructing a user portrait model according to the auxiliary domain sample set and the target domain sample set;

and the user portrait determining unit is used for inputting the product sales data into the user portrait model to obtain a target user portrait matched with the product to be recommended.

Further, the model building unit includes:

the information acquisition subunit is used for acquiring preset initial sampling probability and step length information;

the sample determining subunit is used for sampling in the auxiliary domain sample set and the target domain sample set according to the initial sampling probability to obtain a first auxiliary domain sample and a first target domain sample;

a first model determination subunit, configured to train an initial user portrait model according to the first auxiliary domain sample and the first target domain sample, and determine a first sub-user portrait model;

a list determining subunit, configured to adjust the initial sampling probability according to the step size information, to obtain a sampling probability list;

A second model determining subunit, configured to determine a second sub-user portrait model set based on the auxiliary domain sample set, the target domain sample set, and the sampling probability list;

and the model construction subunit is used for constructing the user portrait model according to the first sub user portrait model and each second user portrait model included in the second user portrait model set.

Wherein the list determination subunit may be specifically configured to:

taking the initial sampling probability as a last sampling probability;

superposing the last sampling probability according to the step length information to obtain a current sampling probability, and determining the current superposition times;

judging whether the current superposition times reach a set times threshold, if so, determining the sampling probability list according to the obtained current sampling probabilities;

and if not, returning to the step of determining the current sampling probability.

Wherein the second model determination subunit may be specifically configured to:

sampling the auxiliary domain sample set and the target domain sample set according to the sampling probability aiming at each sampling probability included in the sampling probability list to obtain a second auxiliary domain sample and a second target domain sample;

Training an initial user portrayal model based on the second auxiliary domain sample and the second target domain sample to determine a second sub-user portrayal model;

forming the second set of sub-user portrayal models from each of the second sub-user portrayal models.

Further, the target user determination module 34 may be specifically configured to:

acquiring user description information associated with the target user portrait and the importance degree duty ratio corresponding to each attribute item in the user description information;

determining a similarity value between each user and the target user portrait in the user list according to the user description information and the importance degree duty ratio;

sorting the similarity values according to a set sorting mode to obtain a candidate user list;

and selecting the user with the set ranking range from the candidate user list as the target recommended user of the product to be recommended.

Optionally, the apparatus further comprises:

and the model correction module is used for guiding the attribute characteristic information into machine classification learning, correcting the product classification model and obtaining a corrected product classification model.

The target user determining device provided by the embodiment of the invention can execute the target user determining method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the executing method.

Example III

Fig. 4 shows a schematic diagram of an electronic device 40 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.

As shown in fig. 4, the electronic device 40 includes at least one processor 41, and a memory communicatively connected to the at least one processor 41, such as a Read Only Memory (ROM) 42, a Random Access Memory (RAM) 43, etc., in which the memory stores a computer program executable by the at least one processor, and the processor 41 may perform various suitable actions and processes according to the computer program stored in the Read Only Memory (ROM) 42 or the computer program loaded from the storage unit 48 into the Random Access Memory (RAM) 43. In the RAM43, various programs and data required for the operation of the electronic device 40 may also be stored. The processor 41, the ROM42 and the RAM43 are connected to each other via a bus 44. An input/output (I/O) interface 45 is also connected to bus 44.

Various components in electronic device 40 are connected to I/O interface 45, including: an input unit 46 such as a keyboard, a mouse, etc.; an output unit 47 such as various types of displays, speakers, and the like; a storage unit 48 such as a magnetic disk, an optical disk, or the like; and a communication unit 49 such as a network card, modem, wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The processor 41 may be various general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 41 performs the respective methods and processes described above, for example, the determination method of the target user.

In some embodiments, the method of determining the target user may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 48. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 40 via the ROM42 and/or the communication unit 49. When the computer program is loaded into the RAM43 and executed by the processor 41, one or more steps of the above-described determination method of the target user may be performed. Alternatively, in other embodiments, the processor 41 may be configured to perform the method of determining the target user in any other suitable way (e.g., by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on 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.

To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.

The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.

The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.

Claims

1. A method for determining a target user, comprising:

Obtain attribute feature information and product sales data of products to be recommended;

Obtain associated sample data related to the product to be recommended according to the attribute feature information and the trained product classification model;

determining a target user portrait matching the product to be recommended according to the associated sample data and the product sales data;

According to the target user portrait and the established user list, determine the target recommendation user of the product to be recommended.

2. The method according to claim 1, characterized in that, according to the associated sample data and the product sales data, determining the target user portrait matching the product to be recommended comprises:

Constructing an auxiliary domain sample set based on the associated sample data, and constructing a target domain sample set based on the product sales data;

Constructing a user portrait model according to the auxiliary domain sample set and the target domain sample set;

The product sales data is input into the user portrait model to obtain a target user portrait matching the product to be recommended.

3. The method according to claim 2, wherein said constructing a user portrait model based on said auxiliary domain sample set and said target domain sample set comprises:

Obtain the preset initial sampling probability and step size information;

performing sampling in the auxiliary domain sample set and the target domain sample set according to the initial sampling probability to obtain a first auxiliary domain sample and a first target domain sample;

training an initial user portrait model according to the first auxiliary domain sample and the first target domain sample, and determining a first sub-user portrait model;

adjusting the initial sampling probability according to the step size information to obtain a sampling probability list;

determining a second sub-user portrait model set based on the auxiliary domain sample set, the target domain sample set, and the sampling probability list;

Constructing the user portrait model according to the first sub-user portrait model and each second user portrait model included in the second user portrait model set.

4. The method according to claim 3, wherein said initial sampling probability is adjusted according to said step size information to obtain a sampling probability list, comprising:

Using the initial sampling probability as the previous sampling probability;

superimposing the last sampling probability according to the step size information to obtain the current sampling probability, and determining the current superposition times;

Judging whether the current number of times of superimposition reaches the set number of times threshold, if so, determining the sampling probability list according to the obtained current sampling probabilities;

If not, return to the step of determining the current sampling probability.

5. The method according to claim 3, wherein said determining a second sub-user portrait model set based on said auxiliary domain sample set, said target domain sample set and said sampling probability list comprises:

For each sampling probability included in the sampling probability list, sample the auxiliary domain sample set and the target domain sample set according to the sampling probability to obtain a second auxiliary domain sample and a second target domain sample;

training an initial user portrait model based on the second auxiliary domain sample and the second target domain sample, and determining a second sub-user portrait model;

The second sub-user portrait model set is formed according to each of the second sub-user portrait models.

6. The method according to claim 1, wherein the target recommendation user of the product to be recommended is determined according to the target user portrait and the established user list:

Obtaining the user description information associated with the target user portrait and the importance ratio corresponding to each attribute item in the user description information;

According to the user description information and the importance ratio, determine the similarity value between each user and the target user portrait included in the user list;

Sorting each of the similarity values according to a set sorting method to obtain a list of candidate users;

Selecting users with a set ranking range from the candidate user list as target recommendation users of the product to be recommended.

7. The method of claim 1, further comprising:

The attribute feature information is imported into machine classification learning, and the product classification model is corrected to obtain a corrected product classification model.

8. A device for determining a target user, comprising:

The first obtaining module is used to obtain attribute feature information and product sales data of the product to be recommended;

The second acquisition module is used to obtain associated sample data related to the product to be recommended according to the attribute feature information and the trained product classification model;

A user portrait determination module, configured to determine a target user portrait matching the product to be recommended according to the associated sample data and the product sales data;

The target user determination module is configured to determine the target recommendation user of the product to be recommended according to the target user portrait and the established user list.

9. An electronic device, characterized in that the electronic device comprises:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

The memory stores a computer program executable by the at least one processor, the computer program is executed by the at least one processor, so that the at least one processor can perform any one of claims 1-7 The method for determining the target user.

10. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method described in any one of claims 1-7 when executed. How to identify target users.