CN115563544A - Method and device for determining type of guest group and server - Google Patents

Abstract

The specification provides a method, a device and a server for determining a guest group type, and is applied to the technical field of artificial intelligence. Based on the method, when the type of the target user is required to be determined, the behavior data and the attribute data of the target user can be obtained firstly; simultaneously processing behavior data and attribute data of the target user by utilizing a pre-trained pre-set guest group classification model at least comprising a pre-set convolution network layer, a pre-set attention mechanism layer and a pre-set classification layer so as to finely extract characteristics with good classification effect from data of multiple dimensions and obtain a corresponding target processing result based on the characteristics; and determining the type of the guest group of the target user according to the target processing result. Therefore, the type of the object group of the target user can be accurately determined, and the classification error is effectively reduced; and then, targeted service pushing can be performed subsequently according to the specific guest group type of the target user, so that a good service pushing effect is obtained, and meanwhile, the user experience is improved.

Description

Method and device for determining type of guest group and server

Technical Field

The specification belongs to the technical field of artificial intelligence, and particularly relates to a method, a device and a server for determining a guest group type.

Background

In many service scenarios, it is often necessary to distinguish the types of the customer groups of the users, so as to push the appropriate service to the users more specifically, so as to improve the success rate of pushing and the user experience of the pushed users.

However, when the user's guest group type is predicted based on the conventional method, there are problems that the classification error is large, the reliability of the classification result is low, and the like. The above problem is particularly apparent when classifying the guest groups of elderly users.

In view of the above problems, no effective solution has been proposed.

Disclosure of Invention

The specification provides a method, a device and a server for determining a guest group type, which can accurately determine the guest group type of a target user and effectively reduce classification errors; and then, accurate and targeted service pushing can be performed subsequently according to the specific guest group type of the target user, a good service pushing effect is obtained, and meanwhile, the user experience is improved.

An embodiment of the present specification provides a method for determining a guest group type, including:

acquiring behavior data and attribute data of a target user;

processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein, the preset guest group classification model at least comprises: the system comprises a preset convolution network layer, a preset attention mechanism layer and a preset classification layer;

and determining the type of the object group of the object user according to the object processing result.

In an embodiment, the processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result includes:

acquiring a behavior feature vector of a target user based on behavior data of the target user by utilizing a preset convolutional network layer;

combining the behavior characteristic vector and the attribute data of the target user by using a preset attention mechanism layer to obtain the combined characteristic of the target user;

and obtaining a corresponding target processing result by utilizing a preset classification layer based on the combination characteristics of the target user.

In one embodiment, the predetermined convolutional network layer includes: a convolution layer, a pooling layer, a full-link layer; the convolution layer is used for performing convolution operation on input behavior data to obtain and output a convolution operation result; the pooling layer is used for reducing the matrix size of the convolution operation result to obtain and output a pooling result; and the pooling layer is used for combining the features extracted based on the convolution layer and the features extracted based on the pooling layer according to the convolution operation result and the pooling result to obtain the behavior feature vector of the target user.

In one embodiment, the preset guest group classification model further includes a preset GRU network layer; and the preset GRU network layer is connected with the preset attention mechanism layer and the preset classification layer.

In one embodiment, the obtaining of the corresponding target processing result based on the combined features of the target user by using the preset classification layer includes:

processing the combination characteristics of the target users output by the preset attention mechanism layer by using a preset GRU network layer to obtain and output the combination characteristics processed by the target users at least carrying time characteristics;

and processing the combination characteristics processed by the target user by utilizing a preset classification layer to obtain and output a corresponding target processing result.

In one embodiment, the target users include users with an age greater than a preset age threshold.

In one embodiment, the behavior data of the target user includes: the inline behavior data and the inline down behavior data.

In one embodiment, the inline behavior data includes: the method comprises the following steps of recording the operation of the pushed service in the mobile application, recording the operation of the pushed service of a webpage and recording the operation of the pushed service of a mail.

In one embodiment, said downlinking the data comprises: manual business records handled by an offline counter and self-service business records handled by offline self-service equipment.

In one embodiment, the preset classification layer comprises a classification layer based on an RFM model.

In one embodiment, after determining the guest group type of the target user according to the target processing result, the method further comprises:

determining a target service matched with a target user according to the type of a guest group of the target user;

and pushing the target service to a target user.

An embodiment of the present specification further provides an apparatus for determining a guest group type, including:

the acquisition module is used for acquiring behavior data and attribute data of a target user;

the processing module is used for processing the behavior data and the attribute data of the target user by utilizing a preset guest group classification model to obtain a corresponding target processing result; wherein the preset passenger group classification model at least comprises: the system comprises a preset convolution network layer, a preset attention mechanism layer and a preset classification layer;

and the determining module is used for determining the type of the guest group of the target user according to the target processing result.

Embodiments of the present specification also provide a server, including a processor and a memory for storing processor-executable instructions, the processor performing the following steps: acquiring behavior data and attribute data of a target user; processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein, the preset guest group classification model at least comprises: the system comprises a preset convolution network layer, a preset attention mechanism layer and a preset classification layer; and determining the type of the object group of the object user according to the object processing result.

Embodiments of the present specification also provide a computer readable storage medium having stored thereon computer instructions, the instructions being executable by a processor to perform the steps of: acquiring behavior data and attribute data of a target user; processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein, the preset guest group classification model at least comprises: the method comprises the following steps of (1) presetting a convolution network layer, a presetting attention mechanism layer and a presetting classification layer; and determining the type of the object group of the object user according to the object processing result.

Embodiments of the present specification further provide a computer program product, which includes a computer program, and when the computer program is executed by a processor, the computer program implements the following steps: acquiring behavior data and attribute data of a target user; processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein the preset passenger group classification model at least comprises: the method comprises the following steps of (1) presetting a convolution network layer, a presetting attention mechanism layer and a presetting classification layer; and determining the type of the object group of the object user according to the object processing result.

Based on the method, the device and the server for determining the guest group type, when the guest group type of the target user needs to be determined, behavior data and attribute data of the target user can be obtained firstly; then, the behavior data and the attribute data of the target user are processed simultaneously by utilizing a pre-trained pre-set passenger group classification model at least comprising a pre-set convolution network layer, a pre-set attention mechanism layer and a pre-set classification layer, so that the characteristics with better classification effect are extracted from the data with multiple dimensions, and the corresponding target processing result is obtained based on the characteristics; and determining the type of the object group of the object user according to the object processing result. Therefore, the type of the object group of the target user can be accurately determined, and the classification error is effectively reduced; and then follow-up accurate and targeted service pushing can be carried out according to the specific guest group type of the target user, so that a good service pushing effect is obtained, and meanwhile, the user experience is also improved. In addition, the behavior characteristics of the old users are also considered, and the online behavior data and the offline behavior data of the target users are collected at the same time in a targeted mode to obtain comprehensive and detailed behavior data related to the target users; and then the type of the target user's guest group can be determined more accurately based on the behavior data.

Drawings

In order to more clearly illustrate the embodiments of the present specification, the drawings needed to be used in the embodiments will be briefly described below, and the drawings in the following description are only some of the embodiments described in the specification, and it is obvious to those skilled in the art that other drawings can be obtained based on the drawings without any inventive work.

Fig. 1 is a flowchart illustrating a method for determining a guest group type according to an embodiment of the present disclosure;

FIG. 2 is a diagram illustrating an example scenario in which an embodiment of a method for determining a class of a guest provided by an embodiment of the present specification is applied;

FIG. 3 is a diagram illustrating an embodiment of a method for determining a type of a guest group according to an embodiment of the present disclosure;

FIG. 4 is a diagram illustrating an embodiment of a method for determining a class of a guest provided by an embodiment of the present specification;

FIG. 5 is a diagram illustrating an embodiment of a method for determining a type of a guest group according to an embodiment of the present disclosure;

FIG. 6 is a diagram illustrating an example scenario in which an embodiment of a method for determining a class of a guest provided by an embodiment of the present specification is applied;

FIG. 7 is a schematic diagram of a server according to an embodiment of the present disclosure;

fig. 8 is a schematic structural component diagram of a guest group type determining apparatus provided in an embodiment of the present specification;

fig. 9 is a schematic diagram of an embodiment of a method for determining a class of a guest provided by an embodiment of the present specification, in an example scenario.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present specification, and not all of the embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments in the present specification without any inventive step should fall within the scope of protection of the present specification.

Referring to fig. 1, an embodiment of the present disclosure provides a method for determining a guest group type, where the method is specifically applied to a server side. In specific implementation, the method may include the following:

s101: acquiring behavior data and attribute data of a target user;

s102: processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein, the preset guest group classification model at least comprises: the system comprises a preset convolution network layer, a preset attention mechanism layer and a preset classification layer;

s103: and determining the type of the object group of the object user according to the object processing result.

According to the embodiment, the data of two dimensions, namely behavior data and attribute data of the target user are acquired and obtained, and the data of the two dimensions are processed simultaneously by using the preset guest group classification model, so that the guest group type of the target user can be determined accurately.

In some embodiments, the above method may be applied specifically to the server side.

As shown in fig. 2, the server may specifically include a background server that is applied to a side of a service platform (e.g., a banking platform) and is capable of implementing functions such as data transmission and data processing. Specifically, the server may be, for example, an electronic device having data operation, storage function and network interaction function. Alternatively, the server may be a software program running in the electronic device to support data processing, storage, and network interaction. In the present embodiment, the number of servers is not particularly limited. The server may specifically be one server, several servers, or a server cluster formed by several servers.

Further, referring to fig. 2, the service may specifically be a server of a banking platform. The server can simultaneously acquire and acquire the online behavior data of the user and the offline behavior data of the user through an online channel or an offline channel. In addition, the server is also connected with a database of the banking platform, and attribute data of the user can be acquired through the database.

It should be noted that the data related to the user are obtained and used under the knowledge and consent of the user. In the embodiments of the present specification, the acquisition, storage, use, processing, and the like of the data are in accordance with the relevant regulations of the national laws and regulations.

In some embodiments, the target user may be specifically understood as a user to be determined in the guest group type. Specifically, the target user may be a customer of a related structure, such as a bank, or a potential customer.

Further, the target users may specifically include users whose ages are greater than a preset age threshold, for example, elderly users (or silver users). The preset age threshold may be 60 years.

Of course, it should be noted that the target users may also include other types of clients, such as young users, children, and so on, according to specific application scenarios and processing requirements. The present specification is not limited to these. In this specification, a target user is mainly taken as an example of an elderly user whose age is greater than a preset age threshold. With regard to the classification of the guest group of other types of users, reference may be made to related embodiments of elderly users. Therefore, the description is omitted.

In some embodiments, the behavior data of the target user may specifically include: on-line behavior data and off-line behavior data, etc.

By the embodiment, the behavior characteristics of the old user are fully considered, and the online behavior data is collected and the offline behavior data is also collected in a targeted manner, so that more comprehensive and similar behavior data can be obtained. And then, the type of the target user can be accurately determined based on the behavior data.

In some embodiments, the inline behavior data comprises: the method comprises the following steps of recording the operation of the pushed service in the mobile application, recording the operation of the pushed service of a webpage and recording the operation of the pushed service of a mail.

The operation record of the service pushed in the mobile application may specifically include: and in a time period, the observation record sequence of the click operation of the multi-target user on the service pushed in the mobile application, or the observation record sequence of the sweep operation of the target user on the service pushed in the mobile application. Specifically, it can be expressed in the following form: { x ₁ ,x ₂ ,...,x _i ...,x _n }. Wherein x is _i The click operation observed for the ith time point in the time period.

Similarly, the operation record of the service pushed to the web page may specifically include: and in a time period, the observation record sequence of the click operation of the multi-target user on the service pushed in the mobile application, or the observation record sequence of the closing operation of the target user on the service pushed in the mobile application.

The operation record of the service pushed by the mail may specifically include an acceptance operation or a rejection operation of the service pushed by the mail.

In some embodiments, the descending of the line as data may specifically include: manual business records handled by an offline counter, self-service business records handled by offline self-service equipment and the like.

In some embodiments, the obtained behavioral data may further include: platform action account records, purchasing information of business products, and the like.

In some embodiments, the attribute data of the target user may specifically include: gender, age, place of residence, income, marital status, health status, etc. of the target user.

In some embodiments, the preset guest group classification model may be specifically understood as a deep neural network model obtained by pre-training and capable of determining a guest group type of a user based on input behavior data and attribute data of the user.

In some embodiments, after the behavior data and the attribute data of the target user are input to the preset guest group classification model, the preset guest group classification model may pre-process the input behavior data and the input attribute data of the target user. The pretreatment specifically may include one or more of the following: denoising processing, zero-filling processing, normalization processing, and the like.

In some embodiments, referring to fig. 3, the preset object classification model at least includes: the system comprises a preset convolution network layer, a preset attention mechanism layer, a preset classification layer and the like.

In some embodiments, after the behavior data and the attribute data of the target user are input into the preset guest group classification model, the preset guest group classification model may convert the behavior data of the target user into a corresponding information data matrix; and then the information data matrix replaces the behavior data of the target user and is input into a preset convolution network layer for further processing.

In some embodiments, referring to fig. 3 and 4, the processing the behavior data and the attribute data of the target user by using the preset guest group classification model to obtain the corresponding target processing result may include the following steps:

s102-1: acquiring a behavior feature vector of a target user based on behavior data of the target user by utilizing a preset convolutional network layer;

s102-2: combining the behavior characteristic vector and the attribute data of the target user by using a preset attention mechanism layer to obtain the combined characteristic of the target user;

s102-3: and obtaining a corresponding target processing result by utilizing a preset classification layer based on the combination characteristics of the target user.

Through the embodiment, the data of two dimensions, namely behavior data and attribute data of the target user can be comprehensively processed by using the preset guest group classification model, and the target processing result with higher accuracy is obtained.

In some embodiments, the preset convolutional network layer may specifically include: a convolution layer, a pooling layer, and a full-link layer. The convolution layer can be specifically used for performing convolution operation on input behavior data to obtain and output a convolution operation result; the pooling layer can be specifically used for reducing the matrix size of the convolution operation result to obtain and output a pooling result; the pooling layer may be specifically configured to combine the features extracted based on the convolution layer and the features extracted based on the pooling layer according to the convolution operation result and the pooling result, so as to obtain the behavior feature vector of the target user.

In some embodiments, specifically referring to fig. 5, the preset guest group classification model may further include a preset GRU network layer (which may be denoted as GRU Net); the preset GRU network layer is connected with a preset attention mechanism layer (which can be recorded as Soft) and a preset classification layer (which can be recorded as Class). Wherein Conv Net is a preset convolutional network layer. X represents behavior data of the target user, and Z represents attribute data of the target user.

The GRU may specifically be a GATED recovery UNIT (GATED recovery UNIT), which is a local control mechanism in the RECURRENT neural network, and includes a RESET GATE (RESET GATE) and an UPDATE GATE (UPDATE GATE), which are used to solve the problem of gradient disappearance in the standard RECURRENT neural network, and simultaneously can better handle the sequence problem of long time interval and delay, and can simultaneously retain long-term information of the sequence.

In specific implementation, the combined features of the target user processed and output by the predetermined attention mechanism layer are input into the predetermined GRU network layer. The preset GRU network layer can analyze and extract the combined features of the target users based on the time dimension, and further comprehensively process the combined features to obtain the combined features which have better classification effect and at least carry the time features and are processed by the target users.

In some embodiments, the obtaining of the corresponding target processing result by using the preset classification layer based on the combined features of the target user may include the following steps:

s1: processing the combination characteristics of the target users output by the preset attention mechanism layer by using a preset GRU network layer to obtain and output the combination characteristics processed by the target users at least carrying time characteristics;

s2: and processing the combination characteristics processed by the target user by utilizing a preset classification layer to obtain and output a corresponding target processing result.

In some embodiments, the predetermined classification layer may specifically include a classification layer based on an RFM model.

Referring to fig. 6, the RFM model can be specifically understood as a model for measuring the value of a client based on the characteristics of the client in three dimensions, i.e., the last consumption (Recency), the consumption Frequency (Frequency), and the consumption amount (money), so as to determine the importance of the client.

In some embodiments, referring to fig. 6, the guest group type of the target user may specifically include at least one of: important remaining users, important developing users, important value users, general remaining users, general developing users, general value users, and the like.

Through the embodiment, the classification layer based on the RFM model is introduced and used in the preset passenger group classification model, and the passenger group type matched with the banking business scene can be accurately determined based on the preset passenger group classification rule shown in the table 1.

TABLE 1

User grouping	R value	F value	Value of M	User ratings
					Important value user	High (a)	Height of	Height of	A
Important developing users	Height of	Is low in	Height of	B
					Important keeping user	Is low in	Height of	High (a)	B
Important saving user	Is low in	Is low with	Height of	B
					General value user	Height of	High (a)	Is low in	C
General development user	Height of	Is low in	Is low in	C
					General maintenance user	Is low in	Height of	Is low in	C
General saving of the user	Is low with	Is low with	Is low in	D

In some embodiments, after determining the type of the target user's guest group according to the target processing result, the method may further include, when being implemented:

s1: determining a target service matched with a target user according to the type of a guest group of the target user;

s2: and pushing the target service to a target user.

In some embodiments, in a specific implementation, for example, for a user with important value, a service related to more resources and higher profit can be pushed in a targeted manner. For the important saving user, the service with higher preferential activity can be pushed in a targeted manner. For a general value user, services related to common resources and common benefits can be pushed in a targeted manner. For a general reserved user, preferential general services can be pushed in a targeted mode. For important development users, the method can pertinently push services which obviously improve resources and obviously improve profits. For important keep users, more stable service can be pushed in a targeted manner. For a general development user, services related to general improvement of resources and general improvement of benefits can be pushed in a targeted manner. For general holding users, ordinary and stable services can be pushed in a targeted manner, and the like.

Through the embodiment, different users can be distinguished according to the types of the guest groups, and matched services can be pushed to different users in a targeted manner, so that a good pushing effect can be obtained, and meanwhile, the users can obtain good interaction experience.

In some embodiments, before implementation, a preset guest group classification model may be obtained through training in the following manner:

s1: acquiring behavior data of a sample old user aiming at an online service and behavior data of a sample old user aiming at an offline service to obtain sample behavior data of the sample old user; meanwhile, sample user data of a sample old user is obtained;

s2: obtaining a sample training set containing sample data of a plurality of sample old users according to the sample behavior data and the sample user data of the sample old users;

s3; according to a preset passenger group type classification rule based on an RFM model, marking sample data in the sample training set to obtain a marked sample training set;

s4: constructing an initial classification model; wherein the initial classification model comprises at least an initial convolutional network layer, an initial attention mechanism layer, and an initial classification layer;

s5: and continuously training the initial classification model by using the marked sample training set to obtain a preset passenger group classification model with the accuracy meeting the requirement.

The initial classification model may be a model based on an artificial neural network structure.

The Artificial Neural Network (ANN) may specifically refer to a basis of existing artificial intelligence application, and a statistical learning method is used to abstract high-dimensional features from original low-dimensional data and obtain effective representations of input space from massive data. Compared with a shallow model, the method has obvious advantages in feature extraction and model fitting.

Referring to fig. 9, the artificial neural network may be regarded as a multi-layer perceptron including a plurality of hidden layers, and the artificial neural network layers may be divided into 3 types, i.e., an input layer, a hidden layer, and an output layer, where nodes of adjacent layers are all connected.

During specific training, the artificial neural network firstly can utilize a forward propagation algorithm to carry out a series of linear operations and activation operations on the weight coefficient matrix W, the bias vector b and the input value x; and then, calculating layer by layer from the input layer to the back, executing operation till the output layer to obtain an output result value, and finally adjusting network parameters through a BP algorithm, so that a preset passenger group classification model meeting the requirements can be obtained through efficient and rapid training.

Through the embodiment, the data of multiple dimensions can be fully utilized, and the preset passenger group classification model with good effect and high precision can be efficiently trained.

As can be seen from the above, based on the method for determining a guest group type provided in the embodiment of the present specification, when a guest group type of a target user needs to be determined, behavior data and attribute data of the target user may be obtained first; then, the behavior data and the attribute data of the target user are processed simultaneously by utilizing a pre-trained pre-set passenger group classification model at least comprising a pre-set convolution network layer, a pre-set attention mechanism layer and a pre-set classification layer, so that the characteristics with better classification effect are extracted from the data with multiple dimensions, and the corresponding target processing result is obtained based on the characteristics; and determining the type of the guest group of the target user according to the target processing result. Therefore, the type of the object group of the target user can be accurately determined, and the classification error is effectively reduced; and then, targeted service pushing can be performed subsequently according to the specific guest group type of the target user, so that a good service pushing effect is obtained, and meanwhile, the user experience is improved. Furthermore, by introducing and using a classification layer based on the RFM model in a preset passenger group classification model, the passenger group type matched with the banking business scene can be more accurately determined. In addition, the behavior characteristics of the old users are considered, and the online behavior data and the offline behavior data of the target users are collected at the same time in a targeted mode to obtain comprehensive and detailed behavior data; and then can be based on above-mentioned action data, determine old user's guest crowd type more accurately.

Embodiments of the present specification further provide a server, including a processor and a memory for storing processor-executable instructions, where the processor, when implemented, may perform the following steps according to the instructions: acquiring behavior data and attribute data of a target user; processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein the preset passenger group classification model at least comprises: the system comprises a preset convolution network layer, a preset attention mechanism layer and a preset classification layer; and determining the type of the object group of the object user according to the object processing result.

In order to complete the above instructions more accurately, referring to fig. 7, an embodiment of the present specification further provides another specific server, where the server includes a network communication port 701, a processor 702, and a memory 703, and the above structures are connected by an internal cable, so that the structures may perform specific data interaction.

The network communication port 701 may be specifically configured to obtain behavior data and attribute data of a target user;

the processor 702 may be specifically configured to process the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein the preset passenger group classification model at least comprises: the system comprises a preset convolution network layer, a preset attention mechanism layer and a preset classification layer; determining the type of the guest group of the target user according to the target processing result;

the memory 703 may be specifically configured to store a corresponding instruction program.

In this embodiment, the network communication port 701 may be a virtual port bound to different communication protocols, so as to send or receive different data. For example, the network communication port may be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for mail data communication. In addition, the network communication port can also be a communication interface or a communication chip of an entity. For example, it may be a wireless mobile network communication chip, such as GSM, CDMA, etc.; it can also be a Wifi chip; it may also be a bluetooth chip.

In this embodiment, the processor 702 may be implemented in any suitable manner. For example, the processor may take the form of, for example, a microprocessor or processor and a computer-readable medium that stores computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, an embedded microcontroller, and so forth. The description is not intended to be limiting.

In this embodiment, the memory 703 may include multiple layers, and in a digital system, the memory may be any memory as long as it can store binary data; in an integrated circuit, a circuit without a physical form and with a storage function is also called a memory, such as a RAM, a FIFO and the like; in the system, the storage device in physical form is also called a memory, such as a memory bank, a TF card and the like.

The present specification further provides a computer-readable storage medium based on the above guest group type determination method, where the computer-readable storage medium stores computer program instructions, and when executed, the computer program instructions implement: acquiring behavior data and attribute data of a target user; processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein the preset passenger group classification model at least comprises: the method comprises the following steps of (1) presetting a convolution network layer, a presetting attention mechanism layer and a presetting classification layer; and determining the type of the object group of the object user according to the object processing result.

In this embodiment, the storage medium includes, but is not limited to, a Random Access Memory (RAM), a Read-Only Memory (ROM), a Cache (Cache), a Hard Disk Drive (HDD), or a Memory Card (Memory Card). The memory may be used to store computer program instructions. The network communication unit may be an interface for performing network connection communication, which is set in accordance with a standard prescribed by a communication protocol.

In this embodiment, the functions and effects specifically realized by the program instructions stored in the computer storage medium can be explained by comparing with other embodiments, and are not described herein again.

Embodiments of the present specification further provide a computer program product, which includes a computer program, and when the computer program is executed by a processor, the computer program implements the following steps: acquiring behavior data and attribute data of a target user; processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein, the preset guest group classification model at least comprises: the system comprises a preset convolution network layer, a preset attention mechanism layer and a preset classification layer; and determining the type of the object group of the object user according to the object processing result.

Referring to fig. 8, in a software level, an embodiment of the present specification further provides an apparatus for determining a guest group type, where the apparatus may specifically include the following structural modules:

the obtaining module 801 may be specifically configured to obtain behavior data and attribute data of a target user;

the processing module 802 may be specifically configured to process the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein, the preset guest group classification model at least comprises: the system comprises a preset convolution network layer, a preset attention mechanism layer and a preset classification layer;

the determining module 803 may be specifically configured to determine the type of the guest group of the target user according to the target processing result.

In some embodiments, when the processing module 802 is implemented specifically, the behavior data and the attribute data of the target user may be processed by using a preset guest group classification model in the following manner to obtain a corresponding target processing result: acquiring a behavior feature vector of a target user based on behavior data of the target user by utilizing a preset convolutional network layer; combining the behavior characteristic vector and the attribute data of the target user by using a preset attention mechanism layer to obtain the combined characteristic of the target user; and obtaining a corresponding target processing result by utilizing a preset classification layer based on the combined characteristics of the target user.

In some embodiments, the preset convolutional network layer may specifically include: a convolution layer, a pooling layer, and a full-link layer; the convolution layer is used for performing convolution operation on input behavior data to obtain and output a convolution operation result; the pooling layer is used for reducing the matrix size of the convolution operation result to obtain and output a pooling result; and the pooling layer is used for combining the features extracted based on the convolution layer and the features extracted based on the pooling layer according to the convolution operation result and the pooling result to obtain the behavior feature vector of the target user.

In some embodiments, the preset guest group classification model may further include a preset GRU network layer; and the preset GRU network layer is connected with the preset attention mechanism layer and the preset classification layer.

In some embodiments, when the processing module 802 is implemented, the preset classification layer may be used to obtain a corresponding target processing result based on the combined features of the target user according to the following manner: processing the combination characteristics of the target users output by the preset attention mechanism layer by using a preset GRU network layer to obtain and output the combination characteristics processed by the target users at least carrying time characteristics; and processing the combination characteristics processed by the target user by utilizing a preset classification layer to obtain and output a corresponding target processing result.

In some embodiments, the target users may specifically include users with an age greater than a preset age threshold, and the like.

In some embodiments, the behavior data of the target user may specifically include: on-line behavior data and off-line behavior data, etc.

In some embodiments, the online behavior data may specifically include: the operation records of the business pushed in the mobile application, the operation records of the business pushed by the webpage, the operation records of the business pushed by the mail and the like.

In some embodiments, the descending of the line as data may specifically include: manual business records handled by offline counters, self-service business records handled by offline self-service equipment and the like.

In some embodiments, the preset classification layer may specifically include a classification layer based on an RFM model, and the like.

In some embodiments, the apparatus may further include an application module, and in implementation, after determining the guest group type of the target user according to the target processing result, the apparatus may determine a target service matched with the target user according to the guest group type of the target user; and pushing the target service to a target user.

It should be noted that, the units, devices, modules, etc. illustrated in the above embodiments may be implemented by a computer chip or an entity, or implemented by a product with certain functions. For convenience of description, the above devices are described as being divided into various modules by functions, which are described separately. It is to be understood that, in implementing the present specification, functions of each module may be implemented in one or more pieces of software and/or hardware, or a module that implements the same function may be implemented by a combination of a plurality of sub-modules or sub-units, or the like. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of logical functional division, and other divisions may be realized in practice, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

As can be seen from the above, based on the apparatus for determining a guest group type provided in the embodiment of the present specification, when a guest group type of a target user needs to be determined, behavior data and attribute data of the target user may be obtained first; then, the behavior data and the attribute data of the target user are processed simultaneously by utilizing a pre-trained pre-set passenger group classification model at least comprising a pre-set convolution network layer, a pre-set attention mechanism layer and a pre-set classification layer, so that the characteristics with better classification effect are extracted from the data with multiple dimensions, and the corresponding target processing result is obtained based on the characteristics; and determining the type of the guest group of the target user according to the target processing result. Therefore, the type of the object group of the target user can be accurately determined, and the classification error is effectively reduced; and then, targeted service pushing can be performed subsequently according to the specific guest group type of the target user, so that a better service pushing effect is obtained, and meanwhile, the user experience is also improved. In addition, the behavior characteristics of the old users are also considered, and the online behavior data and the offline behavior data of the target users are collected at the same time in a targeted mode to obtain comprehensive and detailed behavior data; and then can be based on above-mentioned action data, determine old user's guest crowd type more accurately.

Although the present specification provides method steps as described in the examples or flowcharts, additional or fewer steps may be included based on conventional or non-inventive means. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. When an apparatus or client product in practice executes, it may execute sequentially or in parallel (e.g., in a parallel processor or multithreaded processing environment, or even in a distributed data processing environment) according to the embodiments or methods shown in the figures. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, the presence of additional identical or equivalent elements in a process, method, article, or apparatus that comprises the recited elements is not excluded. The terms first, second, etc. are used to denote names, but not to denote any particular order.

Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may therefore be considered as a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

This description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, classes, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

From the above description of the embodiments, it is clear to those skilled in the art that the present specification can be implemented by software plus necessary general hardware platform. With this understanding, the technical solutions in the present specification may be essentially embodied in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a mobile terminal, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments in the present specification.

The embodiments in the present specification are described in a progressive manner, and the same or similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. The description is operational with numerous general purpose or special purpose computing system environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multiprocessor systems, microprocessor-based systems, set top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments that include any of the above systems or devices, and the like.

While the specification has been described with examples, those skilled in the art will appreciate that there are numerous variations and permutations of the specification without departing from the spirit of the specification, and it is intended that the appended claims encompass such variations and modifications without departing from the spirit of the specification.

Claims (15)

1. A method for determining a class of a guest, comprising:

acquiring behavior data and attribute data of a target user;

processing the behavior data and the attribute data of the target user by using a preset guest group classification model to obtain a corresponding target processing result; wherein the preset passenger group classification model at least comprises: the system comprises a preset convolution network layer, a preset attention mechanism layer and a preset classification layer;

and determining the type of the object group of the object user according to the object processing result.

2. The method according to claim 1, wherein the processing behavior data and attribute data of the target user by using a preset passenger group classification model to obtain a corresponding target processing result comprises:

acquiring a behavior feature vector of a target user based on behavior data of the target user by utilizing a preset convolutional network layer;

combining the behavior characteristic vector and the attribute data of the target user by using a preset attention mechanism layer to obtain the combined characteristic of the target user;

and obtaining a corresponding target processing result by utilizing a preset classification layer based on the combined characteristics of the target user.

3. The method of claim 2, wherein the predetermined convolutional network layer comprises: a convolution layer, a pooling layer, a full-link layer; the convolution layer is used for performing convolution operation on input behavior data to obtain and output a convolution operation result; the pooling layer is used for reducing the matrix size of the convolution operation result to obtain and output a pooling result; and the pooling layer is used for combining the features extracted based on the convolution layer and the features extracted based on the pooling layer according to the convolution operation result and the pooling result to obtain the behavior feature vector of the target user.

4. The method of claim 2, wherein the predetermined guest group classification model further comprises a predetermined GRU network layer; the preset GRU network layer is connected with the preset attention mechanism layer and the preset classification layer.

5. The method of claim 4, wherein obtaining the corresponding target processing result based on the combined features of the target user by using a preset classification layer comprises:

processing the combination characteristics of the target users output by the preset attention mechanism layer by using a preset GRU network layer to obtain and output the combination characteristics processed by the target users at least carrying time characteristics;

and processing the combination characteristics processed by the target user by utilizing a preset classification layer to obtain and output a corresponding target processing result.

6. The method of claim 1, wherein the target users comprise users with an age greater than a preset age threshold.

7. The method of claim 6, wherein the target user's behavior data comprises: on-line behavior data and off-line behavior data.

8. The method of claim 7, wherein the online behavior data comprises: the method comprises the following steps of recording the operation of the pushed service in the mobile application, recording the operation of the pushed service of a webpage and recording the operation of the pushed service of a mail.

9. The method of claim 7, wherein the downlinking the data comprises: manual business records handled by an offline counter and self-service business records handled by offline self-service equipment.

10. The method of claim 6, wherein the predetermined classification level comprises an RFM model-based classification level.

11. The method of claim 1, wherein after determining a guest group type of a target user according to the target processing result, the method further comprises:

determining a target service matched with a target user according to the type of a guest group of the target user;

and pushing the target service to a target user.

12. An apparatus for determining a class of a guest, comprising:

the acquisition module is used for acquiring behavior data and attribute data of a target user;

the processing module is used for processing the behavior data and the attribute data of the target user by utilizing a preset guest group classification model to obtain a corresponding target processing result; wherein, the preset guest group classification model at least comprises: the method comprises the following steps of (1) presetting a convolution network layer, a presetting attention mechanism layer and a presetting classification layer;

and the determining module is used for determining the type of the guest group of the target user according to the target processing result.

13. A server comprising a processor and a memory for storing processor-executable instructions which, when executed by the processor, implement the steps of the method of any one of claims 1 to 11.

14. A computer-readable storage medium having stored thereon computer instructions which, when executed by a processor, carry out the steps of the method of any one of claims 1 to 11.

15. A computer program product comprising a computer program which, when executed by a processor, carries out the steps of the method according to any one of claims 1 to 11.

Images