CN112685478A

CN112685478A - Information processing method for cloud service and user portrait mining and cloud server

Info

Publication number: CN112685478A
Application number: CN202110201173.XA
Authority: CN
Inventors: 张雪君
Original assignee: Individual
Current assignee: Aiyunbao Shanghai Technology Co ltd
Priority date: 2021-02-23
Filing date: 2021-02-23
Publication date: 2021-04-20
Anticipated expiration: 2041-02-23
Also published as: CN112685478B

Abstract

The invention discloses an information processing method and a cloud server aiming at cloud service and user portrait mining, wherein a first service interaction state of a payment service terminal to be analyzed is analyzed to obtain first service interaction state information, identity information of the payment service terminal to be analyzed is judged and analyzed, and then after the payment service data to be analyzed comprises a service interaction prompt message of the payment service terminal to be analyzed, a second service interaction state of the payment service terminal to be analyzed is analyzed to obtain second service interaction state information, so that the user preference information of the payment service terminal to be analyzed is determined by integrating the first service interaction state information and the second service interaction state information, the accuracy of service interaction state analysis is improved, the deviation of service interaction state analysis is reduced, and the preference information of a user is accurately mined, and further, accurate business service recommendation is realized, so that network resources occupied during business service recommendation are reduced.

Description

Information processing method for cloud service and user portrait mining and cloud server

Technical Field

The disclosure relates to the technical field of cloud services and information processing, and in particular relates to an information processing method and a cloud server for cloud services and user portrait mining.

Background

Cloud services refer to cloud services provided by a cloud on demand, that is, required services are obtained in an on-demand and easily-extensible manner through a network, wherein the cloud services refer to addition, usage and interaction modes of related services based on the internet, and generally relate to providing dynamic and easily-extensible and often-virtualized resources through the internet. Briefly, the cloud is a metaphor for networks and the internet. In the past, telecommunications networks were often represented by clouds and later also by the abstraction of the internet and the underlying infrastructure. Such services may be IT and software, internet related, or other services. It means that computing power can also be circulated as a commodity over the internet.

With the rapid development of cloud services, most of offline transactions are transferred to online transactions, users usually display information such as user attributes, user preferences, living habits, user behaviors and the like in a website or an e-commerce platform in a tag form, and merchants analyze the tag information (user images) of the users and then recommend service in a targeted manner.

At present, in the related technology, when a user portrait is mined, the analysis of a service interaction state is difficult to realize, so that the preference information of the user cannot be accurately mined, further, accurate service recommendation cannot be performed, and more network resources are occupied.

Disclosure of Invention

In order to solve the technical problems in the related art, the present disclosure provides an information processing method and a cloud server for cloud service and user portrait mining.

In a first aspect, the present invention provides an information processing method for cloud services and user portrait mining, where the information processing method includes:

acquiring payment service data to be analyzed of a payment service terminal to be analyzed;

analyzing a first service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the first service interaction state;

judging whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal; if the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, judging whether the payment service data to be analyzed comprises a service interaction prompt message of the payment service terminal to be analyzed; if the payment service data to be analyzed comprises the service interaction prompt message of the payment service terminal to be analyzed, analyzing a second service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the second service interaction state;

and determining the user preference information of the payment service terminal to be analyzed based on the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state.

Optionally, the analyzing the first service interaction state of the payment service terminal to be analyzed to obtain the service interaction state information of the first service interaction state includes:

analyzing a first service interaction state of the payment service terminal to be analyzed to obtain user preference information of the first service interaction state and a first preference heat value of the user preference information of the first service interaction state;

if the payment service data to be analyzed includes the service interaction prompt message of the payment service terminal to be analyzed, analyzing a second service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the second service interaction state, including: analyzing a second service interaction state of the payment service terminal to be analyzed to obtain user preference information of the second service interaction state and a second preference heat value of the user preference information of the second service interaction state;

the determining the user preference information of the payment service terminal to be analyzed based on the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state includes: determining user preference information of the payment service terminal to be analyzed based on first user preference information and second user preference information or/and the first preference heat value and the second preference heat value;

the first preference heat value comprises a preference heat value of each payment service information corresponding to the first user preference information; the second preference heat value comprises a preference heat value of each payment service information corresponding to the second user preference information; the determining the user preference information of the payment service terminal to be analyzed based on the first user preference information and the second user preference information or/and the first preference heat value and the second preference heat value comprises:

sequentially comparing whether each payment service information corresponding to the first user preference information is the same as the payment service information of the second user preference information corresponding to the service operation information;

if the payment service information is the same as the service operation information, determining the payment service information as the interactive state payment information of the service operation information corresponding to the user preference information of the payment service terminal to be analyzed;

if the preference heat value of the payment service information of the first user preference information is different from the preference heat value of the payment service information of the service operation information corresponding to the second user preference information, comparing the preference heat value of the payment service information of the first user preference information with the preference heat value of the payment service information of the second user preference information; determining payment service information with high preference heat value as interactive state payment information of service operation information corresponding to user preference information of the payment service terminal to be analyzed;

wherein the first preference heating value further comprises a preference heating value of the first service interaction state, and the second preference heating value further comprises a preference heating value of the second service interaction state; the information processing method further includes:

if the preference heat value of the payment service information of the first user preference information is equal to the preference heat value of the payment service information of the service operation information corresponding to the second user preference information, comparing the preference heat value of the first service interaction state with the preference heat value of the second service interaction state;

and determining the payment service information of the service operation information corresponding to the user preference information with high preference heat value as the interactive state payment information.

Optionally, the determining whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal includes:

judging whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal or not through an identity information analysis model of the payment service terminal; the identity information analysis model of the payment service terminal comprises an even number of analysis submodels;

the method for judging whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal through the identity information analysis model of the payment service terminal comprises the following steps:

analyzing whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal through each analysis submodel to obtain an analysis result of each analysis submodel;

determining the analysis results with large number as the judgment result of whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal;

before the step of judging whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal through the identity information analysis model of the payment service terminal, the method further comprises the following steps:

acquiring an identity information sample of a target payment service terminal and an identity information sample of a negative case payment service terminal;

dividing the identity information samples of the negative example payment service terminals into preset groups, and fusing the negative example payment service terminal samples of each group with the target payment service terminal samples respectively to obtain training samples of the preset groups; wherein the number of the preset groups is the same as the number of the analysis submodels;

respectively utilizing the even groups of training samples to train each initial analysis submodel in sequence to obtain each analysis submodel; wherein each set of training samples corresponds to one of the analysis submodels.

Optionally, before analyzing the second service interaction state of the payment service terminal to be analyzed to obtain the service interaction state information of the second service interaction state, the method includes:

performing operation thermalization processing on the payment service data to be analyzed to obtain operation thermalized payment service data corresponding to the payment service data to be analyzed;

and denoising the payment service data subjected to the operation thermalization so as to perform second service interaction state analysis on the payment service terminal to be analyzed in the payment service data subjected to the operation thermalization.

Optionally, the denoising the payment service data of the operation thermalization includes:

determining to obtain an operation heat change value in the operation thermalized payment service data according to a highest operation heat value and a lowest operation heat value in the operation thermalized payment service data, and setting a first correction parameter;

judging whether the payment service data subjected to operation thermalization needs to be subjected to denoising by using the operation heat change value and the first correction parameter;

if the payment service data with the operation thermalization needs to be subjected to denoising of the payment service data, correcting the first correction parameter to obtain a second correction parameter;

determining to obtain a corrected lowest operation heat value by using the second correction parameter and the operation heat change value;

and correcting the operation thermalization of each service item in the operation thermalized payment service data based on the corrected lowest operation thermalization value and the second correction parameter so as to perform payment service data denoising on the operation thermalized payment service data.

Optionally, the obtaining of the to-be-analyzed payment service data of the to-be-analyzed payment service terminal includes:

acquiring initial payment service data;

responding to the initial payment service data corresponding to the payment service terminal to be analyzed, and acquiring first payment channel information of the payment service terminal to be analyzed in the initial payment service data;

and generating the payment service data to be analyzed based on the initial payment service data and the first payment channel information of the payment service terminal to be analyzed.

Optionally, the analyzing the first service interaction state of the payment service terminal to be analyzed to obtain the service interaction state information of the first service interaction state further includes:

splitting the payment business data to be analyzed based on first preset business operation information of the payment business data to be analyzed to obtain a first interaction state business data block of the payment business data to be analyzed;

performing first service interaction state analysis on the first interaction state service data block to obtain service interaction state information of the first service interaction state;

the analyzing the second service interaction state of the payment service terminal to be analyzed to obtain the service interaction state information of the second service interaction state, further includes: splitting the payment business data to be analyzed based on second preset business operation information of the payment business data to be analyzed to obtain a second interaction state business data block of the payment business data to be analyzed; and performing second service interaction state analysis on the second interaction state service data block to obtain service interaction state information of the second service interaction state.

Optionally, the method further includes:

and determining a service push strategy aiming at the payment service terminal to be analyzed according to the user preference information of the payment service terminal to be analyzed.

Optionally, determining a service pushing policy for the payment service terminal to be analyzed according to the user preference information of the payment service terminal to be analyzed includes:

acquiring each piece of first push item information containing first target service content according to the interest tag corresponding to the user preference information;

for each piece of first pushed item information, determining a first matching value of second pushed item information matched with the first pushed item information in a pushed item information set and first item content description information corresponding to the second pushed item information;

selecting a first amount of third push item information from each second push item information;

for each first item content description information corresponding to each third push item information, determining an interest tendency matching value corresponding to the first item content description information according to a first matching value corresponding to each third push item information of the first item content description information;

determining a second quantity of third pushed item information corresponding to second item content description information with the highest interest tendency matching value, and determining a second matching value corresponding to the second item content description information according to the first quantity, the second quantity and a preset first item matching mode;

determining a highest third matching value in third push item information of the second item content description information, and determining a first target matching value of the first target service content according to the third matching value and the second matching value;

determining a service pushing strategy aiming at the payment service terminal to be analyzed according to the first target matching value and a predetermined matching degree threshold value;

preferably, the determining, according to the first matching value corresponding to each third pushed item information of the first item content description information, the interest tendency matching value corresponding to the first item content description information includes:

determining target matching degree records corresponding to each first matching value according to the first matching value corresponding to each third push item information of the first item content description information and each preset matching degree record;

determining fourth matching values corresponding to the first matching values respectively according to the corresponding relation between each matching degree record and the loading matching value, and determining interest tendency matching values corresponding to the first item content description information according to the fourth matching values;

preferably, the selecting a first number of third push item information from each second push item information includes:

numbering each piece of second pushed item information according to a first matching value corresponding to each piece of second pushed item information, and selecting a first number of pieces of third pushed item information according to the descending order of the first matching values;

preferably, the process of predetermining the threshold of the degree of match includes:

acquiring a first push item search index value of a third amount of fourth push item information in a preset time period;

if the search index value of the first push item is not smaller than a preset first search index threshold value, determining a matching degree threshold value corresponding to the search index value of the first push item according to a predetermined first matching degree endpoint threshold value;

if the search index value of the first push item is not larger than the preset second search index threshold value, determining a matching degree threshold value corresponding to the search index value of the first push item according to a second predetermined matching degree endpoint threshold value;

if the first push item search index value is larger than a preset second search index threshold and smaller than a preset first search index threshold, determining a matching degree threshold corresponding to the first push item search index value according to a predetermined first matching degree endpoint threshold, the second matching degree endpoint threshold, the first push item search index value, the first search index threshold and the second search index threshold; wherein the first search index threshold is greater than the second search index threshold;

preferably, the process of predetermining the first matching degree endpoint threshold and the second matching degree endpoint threshold includes:

determining a target item matching deviation rate record corresponding to the target item matching deviation rate according to the determined target item matching deviation rate and each preset item matching deviation rate record;

determining a first matching degree end point threshold value and a second matching degree end point threshold value corresponding to the target item matching deviation rate record according to the corresponding relation between each item matching deviation rate record and a third matching degree end point threshold value and a fourth matching degree end point threshold value;

preferably, the process of predetermining the corresponding relationship between each item matching deviation rate record and the third matching degree endpoint threshold and the fourth matching degree endpoint threshold includes:

respectively acquiring fifth push item information of a fourth quantity of preset second push item search index values and sixth push item information of a fifth quantity of preset third push item search index values, wherein the second push item search index values are larger than the third push item search index values;

for each item matching deviation rate record, determining a third matching degree endpoint threshold corresponding to the item matching deviation rate record based on the fourth quantity of fifth pushed item information;

and determining a fourth matching degree endpoint threshold corresponding to the item matching deviation rate record based on the fifth quantity of sixth pushed item information.

In a second aspect, the present invention also provides a cloud server, including a processor, a memory, and a bus; the processor and the memory are connected through the bus in a communication mode, and the processor is used for reading the computer program from the memory and executing the computer program to realize the method.

The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects.

The invention discloses an information processing method and a cloud server aiming at cloud service and user portrait mining, wherein a first service interaction state of a payment service terminal to be analyzed is analyzed to obtain service interaction state information of the first service interaction state, identity information of the payment service terminal to be analyzed is judged and analyzed, after further judgment that payment service data to be analyzed comprises service interaction prompt information of the payment service terminal to be analyzed, a second service interaction state of the payment service terminal to be analyzed is analyzed to obtain service interaction state information of the second service interaction state, and finally user preference information of the payment service terminal to be analyzed is determined based on the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state, so that the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state are combined to determine user preference information of the payment service terminal to be analyzed, and accordingly the user preference information of the payment service terminal to be analyzed is determined through the service interaction state information of the first Analyzing the user preference information of the payment service terminal to improve the accuracy of service interaction state analysis and reduce the deviation of the service interaction state analysis, thereby accurately excavating the preference information of the user and further realizing accurate service recommendation, thereby reducing the network resources occupied by the service recommendation.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.

Fig. 1 is a flowchart illustrating an information processing method according to an embodiment of the present invention.

Fig. 2 is a schematic flowchart of an information processing method according to another embodiment of the present invention.

Fig. 3 is a schematic diagram of a training process of the identity information analysis model of the payment service terminal in step S23.

Fig. 4 is a flowchart illustrating an embodiment of denoising of payment service data according to the present embodiment.

Fig. 5 is a block diagram of an information processing apparatus according to an embodiment of the present invention.

Fig. 6 is a schematic diagram of a hardware structure of a cloud server according to an embodiment of the present invention.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

In order to solve the problems in the background art, embodiments of the present invention provide an information processing method and a cloud server for cloud services and user portrait mining, where the method may be summarized as follows: acquiring payment service data to be analyzed of a payment service terminal to be analyzed; analyzing different service interaction states of the payment service terminal to be analyzed based on the payment service data to be analyzed to obtain different service interaction state information; and determining the user preference information of the payment service terminal to be analyzed according to different service interaction state information.

Further, analyzing different service interaction states of the payment service terminal to be analyzed based on the payment service data to be analyzed in the content to obtain different service interaction state information; determining the user preference information of the payment service terminal to be analyzed according to different service interaction state information, which can be realized by the following steps: analyzing a first service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the first service interaction state; judging whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal; if the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, judging whether the payment service data to be analyzed comprises a service interaction prompt message of the payment service terminal to be analyzed; if the payment service data to be analyzed comprises the service interaction prompt message of the payment service terminal to be analyzed, analyzing a second service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the second service interaction state; and determining the user preference information of the payment service terminal to be analyzed based on the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state.

It will be appreciated that in order to facilitate the explanation of the above solution, further description will be given below with reference to the accompanying drawings.

Referring to fig. 1, fig. 1 is a flowchart illustrating an information processing method according to an embodiment of the invention.

Step S11: and acquiring the payment service data to be analyzed of the payment service terminal to be analyzed.

In this step, the payment service data to be analyzed of the payment service terminal to be analyzed is obtained first. In a specific embodiment, the payment service data to be analyzed may be dynamic payment service data obtained by data acquisition of the payment service terminal in the current payment service state by the dynamic data acquisition device, payment service data obtained by data acquisition of the payment service terminal in the current payment service state by the static data acquisition device, or payment service data including an identifier of the payment service terminal to be analyzed, which is not limited herein.

Step S12: and analyzing the first service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the first service interaction state.

And analyzing the first service interaction state of the payment service terminal to be analyzed corresponding to the payment service data to be analyzed to obtain service interaction state information of the first service interaction state. In a specific embodiment, the service interaction state information of the first service interaction state includes, but is not limited to: the first service interaction state corresponds to payment channel service operation information in the payment service data to be analyzed, service interaction state identity information of the first service interaction state, user preference information of the first service interaction state and a first preference heat value of the first service interaction state.

In a specific embodiment, the first service interaction state may be analyzed based on a machine learning algorithm to obtain service interaction state information of the first service interaction state, and a specific analysis method is not limited in this embodiment.

Step S13: and judging whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal.

In this embodiment, when the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, that is, when the payment service terminal to be analyzed is provided with the service interaction state identification information, that is, when the payment service terminal to be analyzed is provided with the second service interaction state different from the first service interaction state, the payment service terminal to be analyzed needs to perform subsequent second service interaction state analysis. Therefore, in this step, it is first determined whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal.

Step S14: and if the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, judging whether the payment service data to be analyzed comprises a service interaction prompt message of the payment service terminal to be analyzed.

When the determination result in the step S13 is that the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, it is further determined whether the payment service data to be analyzed includes the service interaction prompt message of the payment service terminal to be analyzed.

In a specific embodiment, when the payment service data to be analyzed includes a service interaction prompting message of the payment service terminal to be analyzed, the service interaction prompting of the payment service terminal to be analyzed includes a first service interaction state and a second service interaction state. In a specific embodiment, when the payment service data to be analyzed includes a service verification prompt message of the payment service terminal to be analyzed, the service interaction verification prompt message of the payment service terminal to be analyzed includes a first service interaction state.

Step S15: and if the payment service data to be analyzed comprises the service interaction prompt message of the payment service terminal to be analyzed, analyzing the second service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the second service interaction state.

And if the payment service data to be analyzed comprises the service interaction prompt message of the payment service terminal to be analyzed, the service interaction prompt of the payment service terminal to be analyzed in the payment service data to be analyzed comprises a first service interaction state and a second service interaction state. And analyzing the second service interaction state in the service interaction prompt of the payment service terminal to be analyzed to obtain service interaction state information of the second service interaction state.

In a specific embodiment, the service interaction state information of the second service interaction state includes, but is not limited to: the payment channel service operation information of the second service interaction state in the payment service data to be analyzed, the service interaction state identity information of the second service interaction state, the user preference information of the second service interaction state and the second preference heat value of the second service interaction state.

Step S16: and determining user preference information of the payment service terminal to be analyzed based on the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state.

After the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state are obtained, the user preference information of the payment service terminal to be analyzed is determined by integrating the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state, and therefore service interaction state analysis corresponding to the payment service terminal to be analyzed is achieved.

Through the method, the information processing method of the embodiment analyzes the first service interaction state of the payment service terminal to be analyzed to obtain the service interaction state information of the first service interaction state, judges and analyzes the identity information of the payment service terminal to be analyzed, further judges that the payment service data to be analyzed comprises the service interaction prompt message of the payment service terminal to be analyzed, analyzes the second service interaction state of the payment service terminal to be analyzed to obtain the service interaction state information of the second service interaction state, and finally determines the user preference information of the payment service terminal to be analyzed based on the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state, so that the user preference information of the payment service terminal to be analyzed is determined by integrating the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state The method and the device improve the accuracy of the business interaction state analysis and reduce the deviation of the business interaction state analysis, thereby accurately excavating the preference information of the user and further realizing accurate business service recommendation.

Referring to fig. 2, fig. 2 is a flow chart illustrating an information processing method according to another embodiment of the invention.

Step S21: the method comprises the steps of obtaining initial payment service data, responding to the fact that the initial payment service data correspond to a payment service terminal to be analyzed, obtaining first payment channel information of the payment service terminal to be analyzed in the initial payment service data, and generating the payment service data to be analyzed based on the initial payment service data and the first payment channel information of the payment service terminal to be analyzed.

Initial payment service data is acquired first. In a specific embodiment, the initial payment service data may be dynamic payment service data obtained by performing data acquisition on the current payment service state by using a dynamic data acquisition device, payment service data obtained by performing data acquisition on the current payment service state by using a static data acquisition device, or payment service data including service interaction state analysis to be performed, which is not limited herein.

And detecting the initial payment service data through the payment service terminal detection model so as to detect whether the initial payment service data corresponds to the payment service terminal, and when the initial payment service data does not correspond to the payment service terminal to be analyzed, no service interaction state analysis is performed on the initial payment service data. When the initial payment service data corresponds to the payment service terminal to be analyzed, the payment service terminal to be analyzed corresponding to the initial payment service data is continuously detected through the payment service terminal detection model, so that first payment channel information of the payment service terminal to be analyzed in the initial payment service data, namely payment mode information corresponding to the payment service terminal to be analyzed, is detected. The to-be-analyzed payment service data of the embodiment is generated based on the initial payment service data and the first payment channel information of the to-be-analyzed payment service terminal. In a specific embodiment, the payment service terminal detection model may be a machine learning algorithm of a convolutional neural network.

Step S22: and analyzing the first service interaction state of the payment service terminal to be analyzed to obtain the user preference information of the first service interaction state and a first preference heat value of the user preference information of the first service interaction state.

The method comprises the steps of screening payment service data to be analyzed, and screening redundant data in a payment service terminal in a considerable part of first service interaction states, so that when the payment service data to be analyzed is split, screening the redundant data on the payment service data to be analyzed corresponding to the payment service terminal is firstly carried out, the first service interaction states can be obtained based on further analysis processing after the payment service data to be analyzed corresponding to the payment service terminal is split, and the condition that the state data corresponding to the service interaction states are interfered by noise due to inaccurate detection of the payment service terminal is reduced. In this embodiment, the first service interaction state refers to a state in which service interaction is performed in the payment service terminal.

In a specific embodiment, the payment service data to be analyzed is split based on the first preset service operation information of the screened payment service data to be analyzed, so as to obtain a first interaction state service data block of the payment service data to be analyzed.

And screening and splitting the payment service data to be analyzed to obtain a first interaction state service data block of the first service interaction state of the payment service data to be analyzed.

And detecting a first service interaction state of the payment service terminal to be analyzed based on the first interaction state service data block, and analyzing the detected first service interaction state to obtain user preference information of the first service interaction state and a first preference heat value of the user preference information of the first service interaction state.

In a specific implementation manner, the first interaction state service data block may be detected through the service interaction state detection model, so as to obtain the second payment channel information of the first service interaction state in the first interaction state service data block, the service interaction state identity information of the first service interaction state, and the preference heat value of the first service interaction state. And analyzing each payment service information of the first service interaction state through an analysis network based on the first interaction state service data block and second payment channel information of the first service interaction state in the first interaction state service data block to obtain the first user preference information and a preference heat value of each payment service information corresponding to the first user preference information. In a specific embodiment, the service interaction state detection model may be a pre-trained machine learning model, which is not limited herein.

In this step, after obtaining the second payment channel information of the first service interaction state in the first interaction state service data block, the service interaction state identity information of the first service interaction state, the preference heat value of the first service interaction state, and the preference heat value of each payment service information corresponding to the first user preference information and the first user preference information, the analysis of the first service interaction state is completed.

Step S23: judging whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal or not through an identity information analysis model of the payment service terminal; the identity information analysis model of the payment service terminal comprises an even number of analysis submodels.

Inputting the payment service data to be analyzed obtained in step S21 into the identity information analysis model of the payment service terminal, and determining whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal. The identity information analysis model of the payment service terminal comprises an even number of analysis submodels.

In this embodiment, when the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, that is, when the service interaction state identification information is set in the payment service terminal to be analyzed, the subsequent second service interaction state analysis needs to be performed on the payment service terminal to be analyzed.

Specifically, whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal is analyzed through each analysis submodel, and an analysis result of each analysis submodel is obtained; and determining the analysis results with large number as the judgment results that the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal. Wherein, the analysis result is two results, including: the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, and the identity information of the payment service terminal to be analyzed is not the identity information of the target payment service terminal.

In a specific embodiment, when the number of the analysis submodels is 3, analyzing whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal through each analysis submodel to obtain an analysis result of each analysis submodel, if the analysis results of the 3 analysis submodels are different, determining a large number of analysis results as a judgment result of whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, and determining a large number of analysis results as a judgment result of whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal through a global analysis judgment mode, thereby improving the accuracy of the identity information analysis model analysis of the payment service terminal to a certain extent.

Referring to fig. 3, fig. 3 is a schematic diagram illustrating a training process of the identity information analysis model of the payment service terminal in step S23.

Step S31: and acquiring an identity information sample of the target payment service terminal and an identity information sample of the negative case payment service terminal.

Firstly, an identity information sample of a target payment service terminal and an identity information sample of a negative example payment service terminal are obtained.

Step S32: dividing identity information samples of the negative example payment service terminals into preset groups, and fusing the negative example payment service terminal samples of each group with target payment service terminal samples respectively to obtain training samples of the preset groups; wherein the number of the preset groups is the same as the number of the analysis submodels.

In reality, since there are many payment service terminals with various identity information, when the identity information of the target payment service terminal is divided, the training samples have non-objectivity in the classification of a small number of samples due to the imbalance between the number of samples of the identity information of the target payment service terminal and the number of samples of the identity information of the negative example payment service terminal. Therefore, this step improves the above problems, specifically as follows: dividing identity information samples of the negative example payment service terminals into preset groups, and fusing the negative example payment service terminal samples of each group with target payment service terminal samples respectively to obtain training samples of the preset groups; wherein the number of the preset groups is the same as the number of the analysis submodels.

Step S33: respectively utilizing an even number group of training samples to train each initial analysis submodel in sequence to obtain each analysis submodel; wherein each group of training samples corresponds to one analysis submodel.

Sequentially training each initial analysis submodel by using each group of training samples to obtain each analysis submodel; wherein each group of training samples corresponds to one analysis submodel. In a specific embodiment, the initial analysis submodel includes a feature processing unit and a classification unit, and the description of the feature processing unit and the classification unit may refer to the prior art and is not described herein again.

The training process of the step is mainly used for training the feature extraction of the feature processing unit, so that the feature extraction is more accurate, and the accuracy of the classification result of the classification unit is improved.

Step S24: and if the identity information of the target payment service terminal is the identity information of the target payment service terminal, judging whether the payment service data to be analyzed comprises a service interaction prompt message of the payment service terminal to be analyzed.

And analyzing the to-be-analyzed payment service terminal of the to-be-analyzed payment service data through the identity information analysis model of the payment service terminal, and further judging whether the to-be-analyzed payment service data comprises a service interaction prompt message of the to-be-analyzed payment service terminal when the identity information of the to-be-analyzed payment service terminal is analyzed to be the identity information of the target payment service terminal.

In a specific embodiment, whether the payment service data to be analyzed includes a service interaction prompt message or a service verification prompt message of the payment service terminal to be analyzed may be analyzed through the service verification interaction prompt classification model.

In a specific embodiment, when it is detected that the payment service data to be analyzed includes a service interaction prompt message of the payment service terminal to be analyzed, since it has been analyzed that the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, a message for prompting a service interaction state is generally set when the service interaction prompt of the payment service terminal to be analyzed is performed, and therefore, the payment service terminal to be analyzed performs a second service interaction state analysis. When the payment service data to be analyzed is detected to be service verification prompt payment data, the service verification prompt of the payment service terminal to be analyzed does not need to set a relevant indication for prompting the service interaction state, generally, only the service interaction state for performing service interaction in the payment service terminal is set, and therefore, the analysis result of the first service interaction state can be used as the final service interaction state analysis result of the payment service terminal to be analyzed.

Step S25: and if the payment service data to be analyzed comprises the service interaction prompt message of the payment service terminal to be analyzed, performing operation thermalization processing on the payment service data to be analyzed to obtain the operation thermalized payment service data corresponding to the payment service data to be analyzed.

In a specific embodiment, when the to-be-analyzed payment service data includes a service interaction prompt message of the to-be-analyzed payment service terminal, it may be determined that message identification information for prompting a service interaction state should be set in the service interaction prompt of the to-be-analyzed payment service terminal, and therefore, an elevated service interaction state of the to-be-analyzed payment service terminal, that is, a second service interaction state, needs to be analyzed.

In a specific implementation manner, the payment service data to be analyzed is screened, and since a significant part of the second service interaction state has a transaction safety problem in the payment service terminal, when the payment service data to be analyzed is split, the payment service data to be analyzed corresponding to the payment service terminal is first subjected to abnormal data screening, so as to ensure that the second service interaction state can be obtained by further analysis and processing after the second service interaction state is split based on the payment service data to be analyzed corresponding to the payment service terminal, thereby reducing the occurrence of the situation that the state data corresponding to the service interaction state is interfered by noise due to inaccurate detection of the payment service terminal.

In a specific embodiment, the payment service data to be analyzed is split based on the second preset service operation information of the screened payment service data to be analyzed, so as to obtain a second interaction state service data block of the payment service data to be analyzed.

In some possible embodiments, the difference of the operation popularity of the initial payment service data obtained by collecting data of the payment service terminal to be analyzed may be low, so that the embodiment needs to perform denoising on the payment service data to be analyzed, and achieves a better analysis effect by correcting the operation popularity of the payment service data.

Referring to fig. 4, fig. 4 is a schematic flowchart illustrating denoising of payment service data according to an embodiment of the present invention.

Step S41: and determining to obtain an operation heat change value in the operation thermalized payment service data through a highest operation heat value and a lowest operation heat value in the operation thermalized payment service data, and setting a first correction parameter.

And carrying out operation thermalization processing on the payment service data to be analyzed to obtain operation thermalization payment service data corresponding to the payment service data to be analyzed. And determining to obtain an operation heat degree change value in the operation thermalized payment service data through the highest operation heat degree value and the lowest operation heat degree value in the operation thermalized payment service data.

Step S42: and judging whether the payment service data subjected to operation thermalization needs to be subjected to denoising by using the operation heat change value and the first correction parameter.

And judging whether the payment service data subjected to operation thermalization needs to be subjected to denoising by using the operation heat change value and the first correction parameter.

Step S43: and correcting the first correction parameter to obtain a second correction parameter.

And when the payment service data with the hot operation needs to be subjected to denoising, correcting the first correction parameter to obtain a second correction parameter.

Step S44: and determining to obtain the corrected lowest operation heat value by using the second correction parameter and the operation heat change value.

And determining to obtain the corrected lowest operation heat value by using the second correction parameter and the operation heat change value.

Step S45: and correcting the operation thermalization of each service item in the operation thermalized payment service data based on the corrected lowest operation thermalization value and the second correction parameter so as to perform payment service data denoising on the operation thermalized payment service data.

And finally, correcting the operation thermalization of each service item in the operation thermalized payment service data based on the corrected lowest operation heat value, the second correction parameter and the lowest operation heat value so as to perform payment service data denoising on the operation thermalized payment service data.

And finally, the payment service data of the payment service data with the operation thermalization is de-noised through the steps. The payment service data denoising method of the embodiment can filter the payment service data which is subjected to operation thermalization and has a large operation heat change value, so that the payment service data does not need to be subjected to operation thermalization processing. But payment service data denoising is carried out on operation thermalization payment service data with small overall operation heat change value change, the difference between the change after denoising and the difference before denoising of the payment service data with large operation heat change value change is not large, therefore, time consumption can be reduced by filtration, therefore, the payment service data denoising method of the embodiment not only can carry out operation thermalization denoising on the payment service data with low operation heat change value, but also can reduce time consumption for analysis to a certain extent, and analysis efficiency is improved.

Step S26: and analyzing the second service interaction state of the payment service terminal to be analyzed to obtain the user preference information of the second service interaction state and a second preference heat value of the user preference information of the second service interaction state.

And after the payment service data subjected to operation thermalization is subjected to payment service data denoising, analyzing a second service interaction state corresponding to the payment service terminal to be analyzed of a second interaction state service data block subjected to payment service data denoising to obtain user preference information of the second service interaction state and a second preference heat value of the user preference information of the second service interaction state, wherein the second service interaction state is different from the first service interaction state.

And analyzing a second service interaction state corresponding to the payment service terminal to be analyzed based on the second interaction state service data block which is subjected to denoising or does not need to be subjected to denoising of the payment service data, so as to obtain user preference information of the second service interaction state and a second preference heat value of the user preference information of the second service interaction state.

In a specific implementation manner, the second interaction state service data block may be detected through the service interaction state detection model, so as to obtain third payment channel information of the second service interaction state in the second interaction state service data block, service interaction state identity information of the second service interaction state, and a preference heat value of the second service interaction state. And analyzing each payment service information of the second service interaction state through an analysis network based on the second interaction state service data block and third payment channel information of the second service interaction state in the second interaction state service data block to obtain second user preference information and a preference heat value of each payment service information corresponding to the second user preference information.

In this step, after obtaining the third payment channel information of the second service interaction state in the second interaction state service data block, the service interaction state identity information of the second service interaction state, the preference heat value of the second service interaction state, and the preference heat value of each kind of payment service information corresponding to the second user preference information and the second user preference information, the analysis of the second service interaction state is completed.

In a specific embodiment, when the second service interaction state analysis fails to analyze the service interaction state information of the second service interaction state (for example, when the service interaction state is switched or the service interaction state is not switched), the service interaction state information of the first service interaction state is used as the service interaction state information of the payment service terminal to be analyzed.

Step S27: and determining the user preference information of the payment service terminal to be analyzed based on the first user preference information and the second user preference information or/and the first preference heat value and the second preference heat value.

In a specific embodiment, before comparing the first service interaction state information with the second service interaction state information, it is detected whether the service interaction state information of the first service interaction state is analyzed in step S22, and if not, the service interaction state information of the second service interaction state is used as the service interaction state information of the payment service terminal to be analyzed; and if so, comparing the first service interaction state information with the second service interaction state information.

Sequentially comparing whether each payment service information corresponding to the first user preference information is the same as the payment service information of the second user preference information corresponding to the service operation information; if the payment service information is the same as the service operation information, determining the payment service information as the interactive state payment information of the service operation information corresponding to the user preference information of the payment service terminal to be analyzed; if not, comparing the preference heat value of the payment service information of the first user preference information with the preference heat value of the payment service information of the service operation information corresponding to the second user preference information; and determining the payment service information with high preference heat value as the interactive state payment information of the service operation information corresponding to the user preference information of the payment service terminal to be analyzed.

In a specific embodiment, after the user preference information of the payment service terminal to be analyzed is obtained, whether the payment service terminal to be analyzed has a hidden payment safety hazard or not can be judged according to the identity information of the payment service terminal to be analyzed, and if the hidden payment safety hazard exists, the user preference information of the payment service terminal to be analyzed can be reported, so that related personnel can perform safety analysis on the user information of the payment service terminal to be analyzed according to the user preference information.

Through the method, the information processing method of the embodiment analyzes the first service interaction state of the payment service terminal to be analyzed to obtain the service interaction state information of the first service interaction state, analyzes the second service interaction state of the payment service terminal to be analyzed after judging that the payment service data to be analyzed comprises the service interaction prompt message of the payment service terminal to be analyzed to obtain the service interaction state information of the second service interaction state, and finally determines the user preference information of the payment service terminal to be analyzed based on the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state, thereby comprehensively determining the user preference information of the payment service terminal to be analyzed by comparing the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state, the method and the device improve the accuracy of the business interaction state analysis and reduce the deviation of the business interaction state analysis, thereby accurately mining the preference information of the user and further realizing accurate business service recommendation. For example, for service recommendation of a certain user, before the service recommendation, 10 different types of service services may need to be selected for pushing, now only 3 different types of service services may need to be selected for pushing, and if the intended service of the user is service1, network resources occupied by pushing the 3 different types of service services are obviously less than network resources occupied by pushing the 10 different types of service services, so that the pushing efficiency may be improved, and network resource consumption may be reduced. For another example, in some more precise pushing processes, the service1 may be pushed directly, which may further reduce the occupation of network resources.

In an alternative embodiment, the method further comprises: and determining a service push strategy aiming at the payment service terminal to be analyzed according to the user preference information of the payment service terminal to be analyzed.

Further, the service push strategy for the payment service terminal to be analyzed is determined according to the user preference information of the payment service terminal to be analyzed, and the method specifically may further include the contents described in steps a to G.

And step A, acquiring each piece of first push item information containing first target service content according to the interest tag corresponding to the user preference information. In a particular embodiment, the interest tag may be swimming, mountain climbing, badminton, fishing, tasting tea, calligraphy, musical instrument, reading, photography, outing, baking, movie, painting, and the like

And step B, aiming at each piece of first push item information, determining a first matching value of second push item information matched with the first push item information in a push item information set and first item content description information corresponding to the second push item information. In a specific embodiment, the second push item information refers to an item matching the first push item information. The first matching value refers to a matching value of the item content in the pushed item information set and the item content of the second pushed item information.

And C, selecting a first amount of third push item information from each second push item information. Further, this step may be implemented by: numbering each piece of second pushed item information according to a first matching value corresponding to each piece of second pushed item information, and selecting a first number of pieces of third pushed item information according to the descending order of the first matching values.

And step D, aiming at each first item content description information corresponding to each third item information, determining an interest tendency matching value corresponding to the first item content description information according to a first matching value corresponding to each third item information of the first item content description information. Further, in some possible embodiments, the determining the interest tendency matching value corresponding to the first item content description information according to the first matching value corresponding to each third push item information of the first item content description information in step D may further include: determining target matching degree records corresponding to each first matching value according to the first matching value corresponding to each third push item information of the first item content description information and each preset matching degree record; and determining fourth matching values corresponding to the first matching values respectively according to the corresponding relation between each matching degree record and the loading matching value, and determining interest tendency matching values corresponding to the first item content description information according to the fourth matching values.

And E, determining a second quantity of third pushed item information corresponding to second item content description information with the highest interest tendency matching value, and determining a second matching value corresponding to the second item content description information according to the first quantity, the second quantity and a preset first item matching mode.

And step F, determining the highest third matching value in the third push item information of the second item content description information, and determining the first target matching value of the first target service content according to the third matching value and the second matching value.

And G, determining a service pushing strategy aiming at the payment service terminal to be analyzed according to the first target matching value and a predetermined matching degree threshold value. In practical implementation, the process of determining the threshold of the matching degree in advance described in step G includes: acquiring a first push item search index value of a third amount of fourth push item information in a preset time period; if the search index value of the first push item is not smaller than a preset first search index threshold value, determining a matching degree threshold value corresponding to the search index value of the first push item according to a predetermined first matching degree endpoint threshold value; if the search index value of the first push item is not larger than the preset second search index threshold value, determining a matching degree threshold value corresponding to the search index value of the first push item according to a second predetermined matching degree endpoint threshold value; if the first push item search index value is larger than a preset second search index threshold and smaller than a preset first search index threshold, determining a matching degree threshold corresponding to the first push item search index value according to a predetermined first matching degree endpoint threshold, the second matching degree endpoint threshold, the first push item search index value, the first search index threshold and the second search index threshold; wherein the first search index threshold is greater than the second search index threshold.

In the above embodiment, the process of predetermining the first matching degree endpoint threshold and the second matching degree endpoint threshold includes: determining a target item matching deviation rate record corresponding to the target item matching deviation rate according to the determined target item matching deviation rate and each preset item matching deviation rate record; and determining a first matching degree end point threshold value and a second matching degree end point threshold value corresponding to the target item matching deviation rate record according to the corresponding relation between each item matching deviation rate record and a third matching degree end point threshold value and a fourth matching degree end point threshold value. Further, the process of predetermining the corresponding relationship between each item matching deviation rate record and the third matching degree endpoint threshold and the fourth matching degree endpoint threshold includes: respectively acquiring fifth push item information of a fourth quantity of preset second push item search index values and sixth push item information of a fifth quantity of preset third push item search index values, wherein the second push item search index values are larger than the third push item search index values; for each item matching deviation rate record, determining a third matching degree endpoint threshold corresponding to the item matching deviation rate record based on the fourth quantity of fifth pushed item information; and determining a fourth matching degree endpoint threshold corresponding to the item matching deviation rate record based on the fifth quantity of sixth pushed item information.

For example, the service push policy may include in what time period and in what manner, the corresponding service is pushed to the payment service terminal, so that interference of a push action on other ongoing services of the service terminal can be avoided, and meanwhile, the push can be selected when network resources (such as bandwidth) are sufficient, thereby reducing occurrence of other abnormal conditions such as time consumed by push, and improving efficiency of service push.

Through the content described in the steps A to G, firstly, according to the interest tag corresponding to the user preference information, obtaining each piece of first pushed item information of the first target service content, determining a first matching value of second pushed item information matched with the first pushed item information in a pushed item information set and first item content description information corresponding to the second pushed item information aiming at each piece of first pushed item information, secondly, according to first item content description information corresponding to third pushed item information selected from each piece of second pushed item information, further analyzing an interest tendency matching value corresponding to the first item content description information according to the content, and further determining a first target matching value of the first target service content according to the interest tendency matching value. And then determining a service push strategy aiming at the payment service terminal to be analyzed according to the first target matching value and a predetermined matching degree threshold value. Therefore, the first target service content is deeply analyzed according to the interest tag corresponding to the user preference information so as to analyze a target matching value matched with the interest preference, the actual demand and the behavior pattern of the user, then a service push strategy of the payment service terminal to be analyzed is determined according to the target matching value, and then service push is carried out on the user.

On the basis, please refer to fig. 5, the invention further provides a block diagram of an information processing apparatus 50 for cloud service and user portrait mining, which includes the following functional modules.

The service data obtaining module 51 is configured to obtain payment service data to be analyzed of the payment service terminal to be analyzed.

The interaction state analysis module 52 is configured to analyze a first service interaction state of the payment service terminal to be analyzed, so as to obtain service interaction state information of the first service interaction state.

The interaction prompt judging module 53 is configured to judge whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal; if the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, judging whether the payment service data to be analyzed comprises a service interaction prompt message of the payment service terminal to be analyzed; and if the payment service data to be analyzed comprises the service interaction prompt message of the payment service terminal to be analyzed, analyzing a second service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the second service interaction state.

A preference information determining module 54, configured to determine, based on the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state, user preference information of the payment service terminal to be analyzed.

Based on the same inventive concept, the embodiment of the invention also provides an information processing system for cloud service and user portrait mining, which is specifically described below.

An information processing system aiming at cloud service and user portrait mining comprises a payment service terminal to be analyzed and a cloud server which are communicated with each other;

the service payment terminal to be analyzed uploads the service payment data to be analyzed to the cloud server;

the cloud server acquires payment service data to be analyzed of the payment service terminal to be analyzed; analyzing a first service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the first service interaction state; judging whether the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal; if the identity information of the payment service terminal to be analyzed is the identity information of the target payment service terminal, judging whether the payment service data to be analyzed comprises a service interaction prompt message of the payment service terminal to be analyzed; if the payment service data to be analyzed comprises the service interaction prompt message of the payment service terminal to be analyzed, analyzing a second service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the second service interaction state; and determining the user preference information of the payment service terminal to be analyzed based on the service interaction state information of the first service interaction state and the service interaction state information of the second service interaction state.

It is understood that the description of the above system may refer to the description of the corresponding method embodiment.

On the basis, please refer to fig. 6 in combination, which provides a cloud server 110, including a processor 110a, a memory 110b connected to the processor 110a, and a bus 110 c; wherein, the processor 110a and the memory 110b complete the communication with each other through the bus 110 c; the processor 110a is used to call the program instructions in the memory 110b to perform the above-mentioned method.

Further, a readable storage medium is provided, on which a program is stored, which when executed by a processor implements the method described above.

It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.

Claims

1. An information processing method for cloud service and user portrait mining is characterized by comprising the following steps:

2. The information processing method according to claim 1, wherein the analyzing the first service interaction state of the payment service terminal to be analyzed to obtain the service interaction state information of the first service interaction state includes:

3. The information processing method according to claim 1, wherein the determining whether the identity information of the payment service terminal to be analyzed is the identity information of a target payment service terminal includes:

4. The information processing method according to claim 1, wherein before analyzing the second service interaction state of the payment service terminal to be analyzed to obtain the service interaction state information of the second service interaction state, the method comprises:

5. The information processing method of claim 4, wherein the denoising the payment service data heated by the operation comprises:

6. The information processing method according to any one of claims 1 to 5, wherein the acquiring payment service data to be analyzed of the payment service terminal to be analyzed includes:

acquiring initial payment service data;

7. The information processing method according to claim 1, wherein the analyzing a first service interaction state of the payment service terminal to be analyzed to obtain service interaction state information of the first service interaction state, further comprises:

8. The information processing method according to any one of claims 1 to 7, characterized by further comprising:

9. The information processing method according to claim 8, wherein determining a service push policy for the payment service terminal to be analyzed according to the user preference information of the payment service terminal to be analyzed comprises:

10. A cloud server comprising a processor, a memory, and a bus; the processor and the memory are communicatively connected through the bus, and the processor is used for reading the computer program from the memory and executing the computer program to realize the method of any one of the above claims 1 to 9.