CN116257566A

CN116257566A - Network user demand analysis method and platform based on big data mining

Info

Publication number: CN116257566A
Application number: CN202211702046.9A
Authority: CN
Inventors: 季银; 蒋俊; 蒋杰
Original assignee: Jiangsu Yongshuo Zhouyu Data Technology Co ltd
Current assignee: Jiangsu Yongshuo Zhouyu Data Technology Co ltd
Priority date: 2022-12-29
Filing date: 2022-12-29
Publication date: 2023-06-13

Abstract

The application provides a network user demand analysis method and a platform based on big data mining, and relates to the technical field of big data. In the method, a network user behavior sequence to be analyzed corresponding to a network user to be mined is constructed according to the network user behavior to be analyzed of the network user to be mined, and a plurality of network user behaviors to be analyzed are distributed according to corresponding behavior time in the network user behavior sequence to be analyzed; performing data expansion processing on the network user behavior sequence to be analyzed to construct a corresponding network user behavior expansion data set to be analyzed, wherein the data expansion processing at least comprises associated expansion of a network user behavior layer; and analyzing the neural network by using the target user requirements, and performing user requirements analysis processing on the network user behavior extension data set to be analyzed so as to output the target behavior requirements of the network user to be mined. Based on the above, the reliability of user demand analysis can be improved to some extent.

Description

Network user demand analysis method and platform based on big data mining

Technical Field

The application relates to the technical field of big data, in particular to a network user demand analysis method and platform based on big data mining.

Background

The big data technology has more application scenes, for example, a user can analyze the demands of the network user, namely analyze and mine the behavior data of the network user so as to determine the demand information of the network user based on the corresponding network behavior data, thereby carrying out corresponding pushing and the like. However, in the prior art, the behavioral data of the network user is generally directly analyzed to obtain the corresponding requirement information, so there may be a problem that the reliability of the user requirement analysis is poor.

Disclosure of Invention

In view of this, the present application aims to provide a network user demand analysis method and platform based on big data mining, so as to improve the reliability of user demand analysis to a certain extent.

In order to achieve the above purpose, the embodiment of the present application adopts the following technical scheme:

a network user demand analysis method based on big data mining comprises the following steps:

constructing a network user behavior sequence to be analyzed corresponding to the network user to be mined according to the network user behavior to be analyzed of the network user to be mined, wherein a plurality of network user behaviors to be analyzed are distributed according to corresponding behavior time in the network user behavior sequence to be analyzed;

Performing data expansion processing on the network user behavior sequence to be analyzed to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed, wherein the data expansion processing at least comprises association expansion of a network user behavior layer;

and utilizing a target user demand analysis neural network to perform user demand analysis processing on the network user behavior extension data set to be analyzed so as to output target behavior demands of the network user to be mined on the basis of the network user behaviors to be analyzed.

In a possible embodiment, in the above network user requirement analysis method based on big data mining, the step of performing data expansion processing on the network user behavior sequence to be analyzed to construct an expanded data set of network user behaviors to be analyzed corresponding to the network user behavior sequence to be analyzed includes:

extracting a target network user behavior relation distribution, wherein the target network user behavior relation distribution comprises at least one target network user behavior;

performing key data mining on the network user behavior sequence to be analyzed to form a behavior key data description vector corresponding to the network user behavior sequence to be analyzed, and performing key data mining on the target network user behavior relation distribution to form a behavior key data description vector corresponding to the target network user behavior relation distribution;

According to the behavior key data description vector corresponding to the network user behavior sequence to be analyzed, performing significance characteristic analysis operation on the behavior key data description vector corresponding to the target network user behavior relation distribution to form a significance behavior key data description vector corresponding to the target network user behavior relation distribution;

according to the behavior key data description vector corresponding to the target network user behavior relation distribution, performing significance characteristic analysis operation on the behavior key data description vector corresponding to the network user behavior sequence to be analyzed to form a significance behavior key data description vector corresponding to the network user behavior sequence to be analyzed;

according to the significance behavior key data description vector corresponding to the target network user behavior relation distribution and the significance behavior key data description vector corresponding to the network user behavior sequence to be analyzed, matching a first target network user behavior corresponding to the network user behavior to be analyzed in the network user behavior sequence to be analyzed from the at least one target network user behavior included in the target network user behavior relation distribution;

And carrying out data expansion processing on the network user behavior sequence to be analyzed according to the first target network user behavior so as to construct and form a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed.

In a possible embodiment, in the above analysis method of network user requirements based on big data mining, the step of performing key data mining on the network user behavior sequence to be analyzed to form a behavior key data description vector corresponding to the network user behavior sequence to be analyzed includes:

performing first key data mining on the network user behavior sequence to be analyzed to form a first behavior key data description vector corresponding to the network user behavior sequence to be analyzed;

performing second key data mining on the first behavior key data description vector corresponding to the network user behavior sequence to be analyzed to form a second behavior key data description vector corresponding to the network user behavior sequence to be analyzed;

and performing relevant conversion processing on the second behavior key data description vector corresponding to the network user behavior sequence to be analyzed so as to form the behavior key data description vector corresponding to the network user behavior sequence to be analyzed.

In a possible embodiment, in the above network user demand analysis method based on big data mining, the step of performing key data mining on the target network user behavior relationship distribution to form a behavior key data description vector corresponding to the target network user behavior relationship distribution includes:

analyzing the data of the target network user behavior relation distribution to extract target network user behavior data corresponding to the target network user behavior relation distribution and corresponding behavior correlation description data;

performing data expression form conversion processing on the target network user behavior data and the behavior correlation description data corresponding to the target network user behavior relation distribution to obtain a target data form conversion result corresponding to the target network user behavior data and a target data form conversion result corresponding to the behavior correlation description data;

performing aggregation processing on the target data form conversion result corresponding to the target network user behavior data and the target data form conversion result corresponding to the behavior correlation description data to form a corresponding aggregated target data form conversion result;

And carrying out fusion processing on the aggregated target data form conversion result and a pre-configured target fusion parameter distribution to form a behavior key data description vector corresponding to the target network user behavior relation distribution.

In a possible embodiment, in the above analysis method of network user requirements based on big data mining, the step of distributing the corresponding behavior key data description vector according to the target network user behavior relationship, and performing a salient feature analysis operation on the behavior key data description vector corresponding to the network user behavior sequence to be analyzed to form a salient behavior key data description vector corresponding to the network user behavior sequence to be analyzed includes:

analyzing and processing the behavior key data description vector corresponding to the target network user behavior relation distribution to form a representative behavior key data description vector corresponding to the target network user behavior relation distribution;

vector integration processing is carried out on the representative behavior key data description vector corresponding to the target network user behavior relation distribution so as to form an integrated behavior key data description vector corresponding to the target network user behavior relation distribution;

Vector parameter mapping operation is carried out on the integrated behavior key data description vector corresponding to the target network user behavior relation distribution so as to form a mapped behavior key data description vector corresponding to the target network user behavior relation distribution;

and performing significance characteristic analysis operation on the behavior key data description vector corresponding to the network user behavior sequence to be analyzed based on the mapping behavior key data description vector corresponding to the target network user behavior relation distribution so as to form the significance behavior key data description vector corresponding to the network user behavior sequence to be analyzed.

In a possible embodiment, in the above network user demand analysis method based on big data mining, the step of performing data expansion processing on the network user behavior sequence to be analyzed according to the first target network user behavior to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed includes:

searching the related target network user behaviors of the connection with the first target network user behaviors from the target network user behavior relation distribution;

Extracting behavior correlation description data of the related target network user behaviors, and screening the behavior correlation description data of the related target network user behaviors to output screened behavior correlation description data corresponding to the related target network user behaviors;

and carrying out data expansion processing on the network user behavior sequence to be analyzed according to the first target network user behavior, the related target network user behavior and screening behavior correlation description data corresponding to the related target network user behavior so as to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed.

In a possible embodiment, in the above network user demand analysis method based on big data mining, the step of analyzing the neural network by the target network user behavior includes:

the step of carrying out key data mining on the network user behavior sequence to be analyzed to form a behavior key data description vector corresponding to the network user behavior sequence to be analyzed, and carrying out key data mining on the target network user behavior relation distribution to form a behavior key data description vector corresponding to the target network user behavior relation distribution;

The step of performing salient feature analysis operation on the behavior key data description vector corresponding to the target network user behavior relation distribution according to the behavior key data description vector corresponding to the network user behavior sequence to be analyzed so as to form the salient behavior key data description vector corresponding to the target network user behavior relation distribution;

the step of performing salient feature analysis operation on the behavior key data description vector corresponding to the network user behavior sequence to be analyzed according to the behavior key data description vector corresponding to the target network user behavior relationship distribution so as to form the salient behavior key data description vector corresponding to the network user behavior sequence to be analyzed;

and matching a first target network user behavior corresponding to the network user behavior to be analyzed in the network user behavior sequence to be analyzed from the at least one target network user behavior included in the target network user behavior relation distribution according to the significance behavior key data description vector corresponding to the target network user behavior relation distribution and the significance behavior key data description vector corresponding to the network user behavior sequence to be analyzed.

In one possible embodiment, in the above network user demand analysis method based on big data mining, the forming process of the target network user behavior analysis neural network includes:

extracting an exemplary data set and an initial network user behavior analysis neural network, wherein the exemplary data set comprises an exemplary network user behavior sequence to be analyzed and an exemplary target network user behavior relation distribution;

performing key data mining on the exemplary network user behavior sequence to be analyzed through the initial network user behavior analysis neural network to form a behavior key data description vector corresponding to the exemplary network user behavior sequence to be analyzed, and performing key data mining on the exemplary target network user behavior relation distribution through the initial network user behavior analysis neural network to form a behavior key data description vector corresponding to the exemplary target network user behavior relation distribution;

performing salient feature analysis operation on the behavior key data description vector corresponding to the behavior relation distribution of the exemplary target network user according to the behavior key data description vector corresponding to the behavior sequence of the exemplary network user to be analyzed through the initial network user behavior analysis neural network so as to form the salient behavior key data description vector corresponding to the behavior relation distribution of the exemplary target network user;

Performing salient feature analysis operation on the behavior key data description vector corresponding to the exemplary network user behavior sequence to be analyzed according to the behavior key data description vector corresponding to the exemplary target network user behavior relation distribution through the initial network user behavior analysis neural network so as to form the salient behavior key data description vector corresponding to the exemplary network user behavior sequence to be analyzed;

and performing network optimization processing on the initial network user behavior analysis neural network through the salient behavior key data description vector corresponding to the exemplary target network user behavior relation distribution and the salient behavior key data description vector corresponding to the exemplary network user behavior sequence to be analyzed so as to form the target network user behavior analysis neural network.

In a possible embodiment, in the above network user demand analysis method based on big data mining, the step of constructing a network user behavior sequence to be analyzed corresponding to the network user to be mined according to the network user behavior to be analyzed of the network user to be mined includes:

extracting each original network user behavior of a network user to be mined to obtain a plurality of original network user behaviors, and respectively determining the behavior duration of each original network user behavior to output the behavior duration corresponding to each original network user behavior;

Respectively calculating the difference value between the behavior duration time corresponding to each two adjacent original network user behaviors to obtain a corresponding behavior duration time difference value;

marking each original network user behavior with a corresponding behavior duration greater than or equal to a preconfigured behavior duration reference value as a network user behavior to be analyzed, and determining whether to mark the original network user behavior as a corresponding network user behavior to be analyzed according to a behavior duration difference value corresponding to the original network user behavior, wherein the corresponding original network user behavior is marked as a corresponding network user behavior to be analyzed when the behavior duration difference value is less than or equal to a preconfigured behavior duration comparison value, or determining not to mark the corresponding original network user behavior as a corresponding network user behavior to be analyzed when the behavior duration difference value is greater than the behavior duration comparison value;

and constructing a network user behavior sequence to be analyzed corresponding to the network user to be mined according to the corresponding behavior time based on each marked network user behavior to be analyzed.

The application also provides a network user demand analysis platform based on big data mining, which comprises a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program so as to realize the network user demand analysis method based on big data mining.

According to the network user demand analysis method and the network user demand analysis platform based on big data mining, a network user behavior sequence to be analyzed corresponding to the network user to be mined can be constructed according to the network user behaviors to be analyzed of the network user to be mined, and a plurality of network user behaviors to be analyzed are distributed according to corresponding behavior time in the network user behavior sequence to be analyzed; performing data expansion processing on the network user behavior sequence to be analyzed to construct a corresponding network user behavior expansion data set to be analyzed, wherein the data expansion processing at least comprises associated expansion of a network user behavior layer; and analyzing the neural network by using the target user requirements, and performing user requirements analysis processing on the network user behavior extension data set to be analyzed so as to output the target behavior requirements of the network user to be mined. Based on the method, before user demand analysis, the network user behavior sequence to be analyzed is subjected to data expansion processing, so that the data for user demand analysis is more sufficient, the reliability of the obtained result is higher, namely the reliability of the user demand analysis is improved to a certain extent, and the existing defects are overcome.

In order to make the above objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

Fig. 1 is a structural block diagram of a network user demand analysis platform based on big data mining according to an embodiment of the present application.

Fig. 2 is a flowchart illustrating steps included in the big data mining-based network user demand analysis method according to the embodiment of the present application.

Fig. 3 is a flowchart illustrating steps included in the big data mining-based network information pushing method according to the embodiment of the present application.

Detailed Description

For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations.

Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.

Referring to fig. 1, there is shown a big data mining based network user demand analysis platform, such as a server with data processing capabilities, wherein the big data mining based network user demand analysis platform comprises a bus or other communication component for communicating information, and a processor coupled to the bus for processing information.

The large data mining-based network user demand analysis platform also includes a main memory, such as a Random Access Memory (RAM) or other dynamic storage device, coupled to the bus for storing information and instructions to be executed by the processor. The main memory may also be used for storing location information, temporary variables, or other intermediate information during execution of instructions by the processor. The big data mining based network user demand analysis platform may further include a Read Only Memory (ROM) or other static storage device coupled to the bus for storing static information and instructions for the processor. The storage device (such as a solid state device, magnetic disk, or optical disk) may be coupled to the bus for persistently storing information and instructions.

In some implementations, the big data mining based network user demand analysis platform may be coupled via a bus to a display (such as a liquid crystal display, or an active matrix display) for displaying information to a user. An input device, such as a keyboard including alphanumeric and other keys, may be coupled to the bus for communicating information and command selections to the processor. In another embodiment, the input device has a touch screen display.

In some embodiments, the big data mining based network user requirements analysis platform may include a communications adapter (such as a network adapter). A communications adapter may be coupled to the bus and configured to enable communication with a computing or communications network and/or other computing system. In various illustrative embodiments, any type of networking configuration may be implemented using a communications adapter, such as peer-to-peer, LAN, WAN, etc.

It is to be understood that while an example platform has been described in fig. 1, embodiments of the subject matter and functional operations described in this specification can be implemented using other types of digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them, as corresponding methods described below.

With reference to fig. 2, the embodiment of the application further provides a network user demand analysis method based on big data mining, which can be applied to the network user demand analysis platform based on big data mining. The method steps defined by the flow related to the network user demand analysis method based on big data mining can be realized by the network user demand analysis platform based on big data mining.

The specific flow shown in fig. 2 will be described in detail.

Step S110, according to the network user behavior to be analyzed of the network user to be mined, a network user behavior sequence to be analyzed corresponding to the network user to be mined is constructed.

In the embodiment of the invention, the network user demand analysis platform based on big data mining can construct a network user behavior sequence to be analyzed corresponding to the network user to be mined according to the network user behavior to be analyzed of the network user to be mined. And in the network user behavior sequence to be analyzed, a plurality of network user behaviors to be analyzed (such as network browsing behaviors, network interaction behaviors, network shopping behaviors and the like) are distributed according to corresponding behavior time.

And step S120, performing data expansion processing on the network user behavior sequence to be analyzed to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed.

In the embodiment of the invention, the network user demand analysis platform based on big data mining can perform data expansion processing on the network user behavior sequence to be analyzed so as to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed. The data expansion processing at least comprises the association expansion of the network user behavior layer.

And step S130, utilizing a target user demand analysis neural network to perform user demand analysis processing on the network user behavior extension data set to be analyzed so as to output target behavior demands of the network user to be mined on the basis of the network user behaviors to be analyzed.

In the embodiment of the present invention, the network user demand analysis platform based on big data mining may perform user demand analysis processing on the network user behavior extension data set to output a target behavior demand of the network user to be mined based on the network user behavior to be analyzed (the target user demand analysis neural network may be formed by performing network optimization based on relevant exemplary data, where the relevant exemplary data includes an exemplary network user behavior extension data set to be analyzed and a corresponding real behavior demand, so that the target user demand analysis neural network may learn a relationship between the two, so that based on the relationship, user demand analysis processing may be performed on the network user behavior extension data set to be analyzed to obtain a target behavior demand of the network user to be mined based on the network user behavior to be analyzed, and a specific network optimization process may refer to a relevant prior art, which is not limited herein).

Based on the above, before user demand analysis, the network user behavior sequence to be analyzed is subjected to data expansion processing, so that the data for user demand analysis is more sufficient, the reliability of the obtained result is higher, namely the reliability of the user demand analysis is improved to a certain extent, and the existing defects (namely the problem of low reliability of the user demand analysis) are overcome.

It will be appreciated that for step S110 included above, in some specific embodiments, it may further include the following:

marking each original network user behavior of which the corresponding behavior duration is longer than or equal to a preconfigured behavior duration reference value (the behavior duration reference value can be configured according to actual requirements, such as different network environments or different network behavior types, the corresponding behavior duration reference value can be different), as a network user behavior to be analyzed, marking the corresponding original network user behavior as a corresponding network user behavior to be analyzed under the condition that the corresponding behavior duration is smaller than the behavior duration reference value, or marking the corresponding original network user behavior as a corresponding network user behavior to be analyzed under the condition that the corresponding behavior duration is larger than the behavior duration reference value (the behavior duration difference can be the average value of the differences between the behavior durations corresponding to each adjacent original network user behavior, such as the network user duration C can be determined to be the network user behavior to be analyzed under the condition that the corresponding behavior is not longer than the duration C, the network user behavior can be compared under the condition that the corresponding behavior is larger than the duration C can be the network user behavior, and the network duration C can be determined to be less than the network user duration is equal to the original user behavior to be analyzed, thus, it is possible to screen out; in other embodiments, screening may also be performed in combination with other factors).

It will be appreciated that for step S120 included hereinabove, in some specific embodiments it may further include the following:

extracting a target network user behavior relation distribution, wherein the target network user behavior relation distribution comprises at least one target network user behavior (illustratively, the target network user behavior relation distribution can also comprise behavior correlation description data, the behavior correlation description data can be used for comprising correlation relation data between the target network user behaviors and behavior description data corresponding to the target network user behaviors, so that the target network user behaviors can be clearly defined, the problem that one network user behavior has multiple meanings is avoided, and based on the target network user behavior relation distribution, the data expansion processing is performed on the network user behavior sequence to be analyzed, so that some matched and clearly defined target network user behaviors can be added in the network user behavior sequence to be analyzed, so that the network user behaviors to be analyzed in the network user behavior sequence to be analyzed are constrained, namely, clear definition is realized, the follow-up requirement analysis is performed according to the increase, the constraint is performed, and the reliability of the analysis is higher;

It may be appreciated that, for the step of performing the key data mining on the network user behavior sequence to be analyzed to form a behavior key data description vector corresponding to the network user behavior sequence to be analyzed, in some specific embodiments, the method may further include the following steps:

the first key data mining is performed on the network user behavior sequence to be analyzed to form a first behavior key data description vector corresponding to the network user behavior sequence to be analyzed (for example, a configured filter or a filtering matrix can be used for carrying out filtering processing on the network user behavior sequence to be analyzed to form a first behavior key data description vector corresponding to the network user behavior sequence to be analyzed, namely, key data in the first behavior key data description vector is screened out and represented by continuous vectors);

performing second key data mining on the first behavior key data description vector corresponding to the network user behavior sequence to be analyzed to form a second behavior key data description vector corresponding to the network user behavior sequence to be analyzed (illustratively, nonlinear mapping can be performed on the first behavior key data description vector to realize second key data mining, so as to obtain a second behavior key data description vector corresponding to the network user behavior sequence to be analyzed);

Performing a correlation conversion process on the second behavioral key data description vector corresponding to the network user behavior sequence to be analyzed to form a behavioral key data description vector corresponding to the network user behavior sequence to be analyzed (for example, performing a correlation conversion process on the second behavioral key data description vector through configured correlation conversion distribution parameters to form a behavioral key data description vector corresponding to the network user behavior sequence to be analyzed, wherein the correlation conversion distribution parameters can be obtained by learning based on initial correlation conversion distribution parameters, and in addition, a specific correlation conversion processing mode can be that weighting processing is performed on the second behavioral key data description vector based on the correlation conversion distribution parameters, namely, multiplication operation is performed on the second behavioral key data description vector and the second behavioral key data description vector).

It may be appreciated that, for the step of performing key data mining on the target network user behavior relationship distribution to form a behavior key data description vector corresponding to the target network user behavior relationship distribution, which may further include the following in some specific embodiments:

The target network user behavior relation distribution is subjected to data analysis to extract target network user behavior data corresponding to the target network user behavior relation distribution and corresponding behavior correlation description data (the target network user behavior data can refer to included target network user behaviors by way of example, the behavior correlation description data can refer to included correlation data and behavior description data as described above, or only the correlation data can be extracted;

performing data expression form conversion processing on the target network user behavior data and the behavior correlation description data corresponding to the target network user behavior relation distribution to obtain a target data form conversion result corresponding to the target network user behavior data and a target data form conversion result corresponding to the behavior correlation description data (for example, the data expression form conversion processing may refer to integral transformation such as fourier transformation);

aggregating the target data form conversion result corresponding to the target network user behavior data and the target data form conversion result corresponding to the behavior correlation description data to form a corresponding aggregate target data form conversion result (illustratively, the target data form conversion result corresponding to the target network user behavior data and the target data form conversion result corresponding to the behavior correlation description data may be spliced to obtain an aggregate target data form conversion result);

The aggregate target data form conversion result and the pre-configured target fusion parameter distribution are fused to form a behavior key data description vector corresponding to the target network user behavior relation distribution (for example, the target fusion parameter distribution may be obtained by learning related exemplary data based on an initial fusion parameter distribution, in addition, the fusion process may refer to weighting processing, that is, multiplication operation is performed on the aggregate target data form conversion result and the target fusion parameter distribution to obtain the behavior key data description vector, and it may be understood that, in other examples, the target network user behavior relation distribution may also be subjected to key data mining in a manner of performing key data mining on the network user behavior sequence to be analyzed as described above).

It may be understood that, for the behavior key data description vector corresponding to the network user behavior sequence to be analyzed included in the foregoing, performing a salient feature analysis operation on the behavior key data description vector corresponding to the target network user behavior relationship distribution to form a salient behavior key data description vector corresponding to the target network user behavior relationship distribution, in some specific embodiments, the method may further include the following:

Vector integration processing is carried out on the behavior key data description vector corresponding to the network user behavior sequence to be analyzed so as to form an integrated behavior key data description vector corresponding to the network user behavior sequence to be analyzed;

vector parameter mapping operation is carried out on the integrated behavior key data description vector corresponding to the network user behavior sequence to be analyzed (for example, vector parameters included in the integrated behavior key data description vector can be mapped to an interval 0-1 to form a mapped behavior key data description vector) so as to form a mapped behavior key data description vector corresponding to the network user behavior sequence to be analyzed;

and performing significance characteristic analysis operation on the behavior key data description vector corresponding to the target network user behavior relation distribution according to the mapping behavior key data description vector corresponding to the network user behavior sequence to be analyzed so as to form the significance behavior key data description vector corresponding to the target network user behavior relation distribution.

It may be understood that, for the step of performing vector integration processing on the behavior key data description vector corresponding to the network user behavior sequence to be analyzed, so as to form an integrated behavior key data description vector corresponding to the network user behavior sequence to be analyzed, in some specific embodiments, the method may further include the following steps:

Counting the number of the target network user behaviors included in the target network user behavior relation distribution to output a behavior number statistic value corresponding to the target network user behavior relation distribution;

determining a corresponding vector integration processing parameter distribution and a corresponding vector shift parameter according to the behavior quantity statistic (illustratively, the distribution dimension of the vector integration processing parameter distribution and the parameter dimension of the vector shift parameter may have a positive correlation relationship with the behavior quantity statistic, for example, the three may be equal to each other);

according to the vector integration processing parameter distribution, carrying out weighting processing on the behavior key data description vector corresponding to the network user behavior sequence to be analyzed to form a weighted behavior key data description vector of the network user behavior sequence to be analyzed (the vector integration processing parameter distribution can be a weight array for mapping the vector integration processing parameter distribution to obtain the weighted behavior key data description vector in an exemplary manner);

and performing shift processing on the weighted behavior key data description vector corresponding to the network user behavior sequence to be analyzed according to the vector shift parameter to form an integrated behavior key data description vector corresponding to the network user behavior sequence to be analyzed (the vector shift parameter may be a vector, for example, bias in the related art, so as to adjust the weighted behavior key data description vector, for example, the vector shift parameter and the weighted behavior key data description vector may be summed to obtain the integrated behavior key data description vector).

It may be understood that the mapped behavior key data description vector corresponding to the network user behavior sequence to be analyzed includes a plurality of first description vector parameters, and the behavior key data description vector corresponding to the target network user behavior relation distribution includes a plurality of second description vector parameters, based on which, for the mapped behavior key data description vector corresponding to the network user behavior sequence to be analyzed included in the foregoing, performing a salient feature analysis operation on the behavior key data description vector corresponding to the target network user behavior relation distribution to form a salient behavior key data description vector corresponding to the target network user behavior relation distribution, in some specific embodiments, the method may further include the following steps:

performing correlation fusion on the second description vector parameters corresponding to the target network user behavior relation distribution and the corresponding first description vector parameters in the network user behavior sequence to be analyzed to form corresponding salient second description vector parameters (illustratively, the first description vector parameters and the second description vector parameters with the same or corresponding distribution positions can be multiplied to obtain corresponding salient second description vector parameters);

And determining a salient behavior key data description vector corresponding to the target network user behavior relation distribution based on each salient second description vector parameter (that is, the salient second description vector parameters can be combined to form the salient behavior key data description vector).

It may be appreciated that, for the behavior key data description vector corresponding to the distribution of the behavior relation according to the target network user behavior, performing a salient feature analysis operation on the behavior key data description vector corresponding to the network user behavior sequence to be analyzed to form a salient behavior key data description vector corresponding to the network user behavior sequence to be analyzed, in some specific embodiments, the method may further include the following steps:

analyzing the behavior key data description vector corresponding to the target network user behavior relation distribution to form a representative behavior key data description vector corresponding to the target network user behavior relation distribution (for example, because the information quantity of the representative behavior key data description vector corresponding to the target network user behavior relation distribution is relatively large, the behavior key data description vector corresponding to the exemplary target network user behavior relation distribution can be analyzed to obtain the representative behavior key data description vector corresponding to the target network user behavior relation distribution;

Vector integration processing (refer to the related description) is carried out on the representative behavior key data description vector corresponding to the target network user behavior relation distribution so as to form an integrated behavior key data description vector corresponding to the target network user behavior relation distribution;

vector parameter mapping operation (refer to the related description) is carried out on the integrated behavior key data description vector corresponding to the target network user behavior relation distribution so as to form a mapped behavior key data description vector corresponding to the target network user behavior relation distribution;

and performing significance characteristic analysis operation on the behavior key data description vector corresponding to the network user behavior sequence to be analyzed based on the mapping behavior key data description vector corresponding to the target network user behavior relation distribution to form a significance behavior key data description vector corresponding to the network user behavior sequence to be analyzed (illustratively, the significance characteristic analysis operation may refer to adding and calculating vector parameters of the same distribution position or corresponding distribution position between the mapping behavior key data description vector corresponding to the target network user behavior relation distribution and the behavior key data description vector corresponding to the network user behavior sequence to be analyzed).

Wherein it can be understood that, for the salient behavior key data description vector corresponding to the target network user behavior relation distribution and the salient behavior key data description vector corresponding to the network user behavior sequence to be analyzed included in the foregoing, the step of matching the first target network user behavior corresponding to the network user behavior to be analyzed in the network user behavior sequence to be analyzed from the at least one target network user behavior included in the target network user behavior relation distribution may further include, in some specific embodiments, the following steps:

vector aggregation processing is performed on the salient behavior key data description vector corresponding to the target network user behavior relation distribution and the salient behavior key data description vector corresponding to the network user behavior sequence to be analyzed to form a corresponding aggregated salient behavior key data description vector (illustratively, vector splicing may be performed on the salient behavior key data description vector corresponding to the target network user behavior relation distribution and the salient behavior key data description vector corresponding to the network user behavior sequence to be analyzed, or vector multiplication and other processing may be performed to obtain a corresponding aggregated salient behavior key data description vector);

Performing a probability parameter conversion process on the aggregate salient behavioral key data description vector to form corresponding probability parameter distribution information (illustratively, the aggregate salient behavioral key data description vector may be subjected to a conversion process based on a configured conversion process distribution parameter so that a vector parameter included in the aggregate salient behavioral key data description vector belongs to 0-1 and is used for characterizing a matching probability to obtain corresponding probability parameter distribution information;

according to the likelihood parameter distribution information, a first target network user behavior corresponding to a network user behavior to be analyzed in the network user behavior sequence to be analyzed is matched from at least one target network user behavior included in the target network user behavior relation distribution (for example, the likelihood parameter distribution information may reflect a likelihood that the network user behavior to be analyzed in the network user behavior sequence to be analyzed matches each target network user behavior in the target network user behavior relation distribution).

It may be appreciated that, for the step of performing data expansion processing on the network user behavior sequence to be analyzed according to the first target network user behavior included in the foregoing to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed, in some specific embodiments, the method may further include the following steps:

finding a relevant target network user behavior of a connection with the first target network user behavior from the target network user behavior relationship distribution (that is, in the target network user behavior relationship distribution, whether the two target network user behaviors are relevant or not can be reflected by whether the two target network user behaviors are connected or not);

extracting behavior correlation description data of the related target network user behavior, filtering the behavior correlation description data of the related target network user behavior to output filtered behavior correlation description data corresponding to the related target network user behavior (illustratively, the filtering step may be not performed in some embodiments, that is, the data expansion processing is directly performed on the network user behavior sequence to be analyzed according to the first target network user behavior, the related target network user behavior and the behavior correlation description data corresponding to the related target network user behavior to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed, and in addition, a specific filtering mode is not limited, for example, in some embodiments, the correlation description data may be filtered in a correlation manner based on the behavior data in the network user behavior sequence to be analyzed to screen out data without any correlation, etc.);

According to the first target network user behavior, the related target network user behavior and the screening behavior correlation description data corresponding to the related target network user behavior, data expansion processing is performed on the network user behavior sequence to be analyzed to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed (illustratively, the screening behavior correlation description data corresponding to the first target network user behavior, the related target network user behavior and the related target network user behavior can be directly combined into the network user behavior sequence to be analyzed to realize data expansion processing, or, considering that the content of the first target network user behavior is possibly consistent with the corresponding network user behavior to be analyzed in the network user behavior sequence to be analyzed, the screening behavior correlation description data corresponding to the first target network user behavior, the related target network user behavior and the related target network user behavior can be replaced by the corresponding to the first target network user behavior, or corresponding to the related target network user behavior can be realized, so that the corresponding behavior can have a defined meaning or the related behavior, namely, the correlation between the related behavior and the related target network user behavior is defined, and the related target network user behavior is the relevant network user behavior, and the relevant target network user behavior is defined by the relevant network user behavior, and the relevant target network user behavior is correspondingly defined by the network user behavior.

It will be appreciated that for the various steps included hereinabove, in some particular embodiments, they may be performed by a target network user behavior analysis neural network, for example, the following steps may be performed by the target network user behavior analysis neural network:

It will be appreciated that for the target network user behavior analysis neural network described above, in some particular embodiments, the formation process may include the following:

extracting an exemplary data set and an initial network user behavior analysis neural network, wherein the exemplary data set comprises an exemplary network user behavior sequence to be analyzed and an exemplary target network user behavior relation distribution; performing key data mining on the exemplary network user behavior sequence to be analyzed through the initial network user behavior analysis neural network to form a behavior key data description vector corresponding to the exemplary network user behavior sequence to be analyzed, and performing key data mining on the exemplary target network user behavior relation distribution through the initial network user behavior analysis neural network to form a behavior key data description vector corresponding to the exemplary target network user behavior relation distribution; performing salient feature analysis operation on the behavior key data description vector corresponding to the behavior relation distribution of the exemplary target network user according to the behavior key data description vector corresponding to the behavior sequence of the exemplary network user to be analyzed through the initial network user behavior analysis neural network so as to form the salient behavior key data description vector corresponding to the behavior relation distribution of the exemplary target network user; performing salient feature analysis operation on the behavior key data description vector corresponding to the exemplary network user behavior sequence to be analyzed according to the behavior key data description vector corresponding to the exemplary target network user behavior relation distribution through the initial network user behavior analysis neural network so as to form the salient behavior key data description vector corresponding to the exemplary network user behavior sequence to be analyzed; the method comprises the steps of performing network optimization processing on the initial network user behavior analysis neural network to form the target network user behavior analysis neural network by using a corresponding significance behavior key data description vector of the exemplary target network user behavior relation distribution and a significance behavior key data description vector corresponding to the exemplary target network user behavior relation distribution to form a corresponding aggregate significance behavior key data description vector, determining whether the network user behavior to be analyzed in the initial network user behavior sequence is consistent with the target network user behavior in the exemplary target network user behavior relation distribution or not through a classification model in the initial network user behavior analysis neural network, if so, determining the corresponding consistency or matching possibility, then comparing the consistency or matching possibility with a pre-configured label to analyze a corresponding learning cost value based on the consistency or matching possibility, then optimizing the target network user behavior in the initial network user behavior sequence such as the target network user behavior relation A and the target network to obtain a corresponding learning cost value based on the learning cost value, otherwise optimizing the target network user behavior in the initial network user behavior sequence such as the target network user behavior relation A, the network user behavior C to be analyzed matches the target network user behavior D).

With reference to fig. 3, the embodiment of the application also provides a network information pushing method based on big data mining, which can be applied to the network user demand analysis platform based on big data mining. The method steps defined by the flow related to the network information pushing method based on big data mining can be realized by the network user demand analysis platform based on big data mining.

The network information pushing method based on big data mining can comprise the following steps:

constructing a network user behavior sequence to be analyzed corresponding to the network user to be mined according to the network user behavior to be analyzed of the network user to be mined (the whole content of the step S110 can be included, and the details are not repeated here);

performing data expansion processing on the network user behavior sequence to be analyzed to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed (all the contents of the step S120 can be included, and are not described in detail here);

performing user demand analysis processing on the network user behavior extension data set to be analyzed by utilizing a target user demand analysis neural network to output target behavior demands of the network user to be mined on the basis of the network user behaviors to be analyzed (all contents of the step S130 can be included and are not described in detail herein);

And determining matched target network pushing information (for example, comparing the target behavior requirement with each candidate network pushing information to determine target network pushing information matched with the target behavior requirement) based on the target behavior requirement, and then sending the target network pushing information to the network terminal equipment corresponding to the network user to be mined.

In summary, according to the network user demand analysis method and platform based on big data mining, a network user behavior sequence to be analyzed corresponding to a network user to be mined can be constructed according to the network user behaviors to be analyzed of the network user to be mined, and a plurality of network user behaviors to be analyzed included in the network user behavior sequence to be analyzed are distributed according to corresponding behavior time; performing data expansion processing on the network user behavior sequence to be analyzed to construct a corresponding network user behavior expansion data set to be analyzed, wherein the data expansion processing at least comprises associated expansion of a network user behavior layer; and analyzing the neural network by using the target user requirements, and performing user requirements analysis processing on the network user behavior extension data set to be analyzed so as to output the target behavior requirements of the network user to be mined. Based on the above, before user demand analysis, the network user behavior sequence to be analyzed is subjected to data expansion processing, so that the data for user demand analysis is more sufficient, the reliability of the obtained result is higher, namely the reliability of the user demand analysis is improved to a certain extent, and the existing defects (namely the problem of low reliability of the user demand analysis) are overcome.

The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the same, but rather, various modifications and variations may be made by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. The network user demand analysis method based on big data mining is characterized by comprising the following steps of:

2. The method for analyzing network user requirements based on big data mining according to claim 1, wherein the step of performing data expansion processing on the network user behavior sequence to be analyzed to construct a network user behavior expansion data set to be analyzed corresponding to the network user behavior sequence to be analyzed includes:

3. The method for analyzing network user requirements based on big data mining according to claim 2, wherein the step of performing key data mining on the network user behavior sequence to be analyzed to form a behavior key data description vector corresponding to the network user behavior sequence to be analyzed includes:

4. The method for analyzing network user requirements based on big data mining according to claim 2, wherein the step of performing key data mining on the target network user behavior relation distribution to form a behavior key data description vector corresponding to the target network user behavior relation distribution includes:

5. The method for analyzing network user requirements based on big data mining according to claim 2, wherein the step of distributing the corresponding behavior key data description vectors according to the target network user behavior relation, performing a salient feature analysis operation on the behavior key data description vectors corresponding to the network user behavior sequence to be analyzed to form the salient behavior key data description vectors corresponding to the network user behavior sequence to be analyzed includes:

6. The method for analyzing network user requirements based on big data mining according to claim 2, wherein the step of performing data expansion processing on the network user behavior sequence to be analyzed according to the first target network user behavior to construct an expanded data set of network user behaviors to be analyzed corresponding to the network user behavior sequence to be analyzed includes:

7. The method of claim 2, wherein the step of analyzing the neural network by the target network user behavior comprises:

8. The method for analyzing network user demands based on big data mining according to claim 7, wherein the forming process of the target network user behavior analysis neural network comprises:

9. The method for analyzing network user requirements based on big data mining according to any one of claims 1 to 8, wherein the step of constructing a sequence of network user behaviors to be analyzed corresponding to the network user to be mined according to the network user behaviors to be analyzed of the network user to be mined includes:

10. A big data mining based network user demand analysis platform, comprising a processor and a memory, the memory for storing a computer program, the processor for executing the computer program to implement the method of any of claims 1-9.