CN117421702A

CN117421702A - User data processing method and device

Info

Publication number: CN117421702A
Application number: CN202311317168.0A
Authority: CN
Inventors: 王凯; 何慧梅
Original assignee: Advanced New Technologies Co Ltd
Current assignee: Advanced New Technologies Co Ltd
Priority date: 2019-05-14
Filing date: 2019-05-14
Publication date: 2024-01-19
Also published as: CN110264037B; CN110264037A

Abstract

The application discloses a method and a device for processing user data, wherein the method comprises the following steps: acquiring behavior data of multiple classes of users, wherein the multiple classes of users are divided based on service labels, and the behavior data of one user comprises a plurality of behavior events and time sequence identifiers of the plurality of behavior events; determining a time sequence association rule corresponding to the multiple types of users based on the behavior events and time sequence identifiers of the behavior events included in the behavior data of the multiple types of users; determining a time sequence behavior rule of the target class user based on a first time sequence association rule corresponding to the target class user and a second time sequence association rule corresponding to a non-target class user in the multiple classes of users; and determining whether the user to be identified belongs to the target class user or not based on the time sequence behavior rule of the target class user.

Description

User data processing method and device

This document is a divisional application of patent application with application number "201910400073.2", application date "2019, 05, 14 days", and application name "a method and apparatus for processing user data".

Technical Field

The present invention relates to the field of data processing, and in particular, to a method and apparatus for processing user data.

Background

In the prior art, in many application scenarios, behavioral data of a user needs to be analyzed to perform decision processing. For example, in the context of risk identification, behavior data of a high risk user and behavior data of a user to be identified may be analyzed to determine whether the user to be identified belongs to the high risk user based on the analysis result.

In general, in analyzing behavior data of a user, it is possible to analyze whether or not a user has performed certain behaviors, or the number of times certain behaviors have occurred. However, in practical applications, such a data analysis method is generally relatively simple, and cannot fully mine the behavior data of the user, so that an effective decision process cannot be performed.

Disclosure of Invention

The embodiment of the application provides a processing method and device for user data, which are used for solving the problem that in the prior art, decision processing cannot be effectively performed due to the fact that behavior data of a user cannot be fully mined.

In order to solve the technical problems, the embodiment of the application is realized as follows:

in a first aspect, a method for processing user data is provided, including:

acquiring behavior data of multiple classes of users, wherein the multiple classes of users are divided based on service labels, and the behavior data of one user comprises a plurality of behavior events and time sequence identifiers of the plurality of behavior events;

Determining a time sequence association rule corresponding to the multiple types of users based on the behavior events and time sequence identifiers of the behavior events included in the behavior data of the multiple types of users;

determining a time sequence behavior rule of the target class user based on a first time sequence association rule corresponding to the target class user and a second time sequence association rule corresponding to a non-target class user in the multiple classes of users;

and determining whether the user to be identified belongs to the target class user or not based on the time sequence behavior rule of the target class user.

In a second aspect, a processing apparatus for user data is provided, including:

the system comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit acquires behavior data of multiple types of users, the multiple types of users are divided based on service labels, and the behavior data of one user comprises a plurality of behavior events and time sequence identifiers of the plurality of behavior events;

the processing unit is used for determining time sequence association rules corresponding to the multiple types of users based on the behavior events and time sequence identifiers of the behavior events included in the behavior data of the multiple types of users;

the determining unit is used for determining the time sequence behavior rule of the target class user based on a first time sequence association rule corresponding to the target class user and a second time sequence association rule corresponding to the non-target class user in the multiple classes of users;

And the identification unit is used for determining whether the user to be identified belongs to the target class user or not based on the time sequence behavior rule of the target class user.

In a third aspect, an electronic device is presented, the electronic device comprising:

a processor; and

a memory arranged to store computer executable instructions that, when executed, cause the processor to:

In a fourth aspect, a computer-readable storage medium storing one or more programs that, when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the method of:

In a fifth aspect, a method for processing user data is provided, including:

Determining a time sequence behavior rule of the multiple types of users based on the time sequence association rule corresponding to the multiple types of users;

generating a plurality of time sequence behavior rule features based on the time sequence behavior rules of the multiple types of users;

and determining time sequence behavior characteristics of the user to be processed based on the time sequence behavior rule characteristics so as to generate sample data comprising the time sequence behavior characteristics for model training.

A sixth aspect provides a processing apparatus for user data, including:

a first determining unit, configured to determine a timing behavior rule of the multiple types of users based on the timing association rule corresponding to the multiple types of users;

a generation unit that generates a plurality of timing behavior rule features based on the timing behavior rules of the plurality of types of users;

and a second determining unit for determining time sequence behavior characteristics of the user to be processed based on the time sequence behavior rule characteristics so as to generate sample data comprising the time sequence behavior characteristics for model training.

In a seventh aspect, an electronic device is provided, the electronic device comprising:

a processor; and

In an eighth aspect, a computer-readable storage medium is provided, the computer-readable storage medium storing one or more programs, which when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the method of:

The above-mentioned at least one technical scheme that this application embodiment adopted can reach following beneficial effect:

according to the technical scheme provided by the embodiment of the application, when the behavior data of the user is analyzed, the time sequence identification of the behavior event included in the behavior data is fully considered, so that the behavior data of the user can be fully mined; when the behavior data of the multiple types of users are analyzed, the behavior data of the multiple types of users can be fully mined to obtain the time sequence association rule capable of reflecting the time sequence of the behaviors of the multiple types of users, and further the time sequence behavior rule of the target type of users is obtained, so that when decision processing is carried out based on the time sequence behavior rule of the target type of users, effective decision processing can be carried out.

Drawings

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

FIG. 1 is a flow chart of a method of processing user data according to one embodiment of the present application;

FIG. 2 is a flow chart of a method of processing user data according to one embodiment of the present application;

FIG. 3 is a schematic structural diagram of an electronic device according to one embodiment of the present application;

FIG. 4 is a schematic diagram of a user data processing device according to an embodiment of the present application;

FIG. 5 is a schematic structural diagram of an electronic device according to one embodiment of the present application;

fig. 6 is a schematic structural diagram of a processing device for user data according to an embodiment of the present application.

Detailed Description

In order to better understand the technical solutions in the present application, the following description will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some embodiments of the present application, not all embodiments. 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, shall fall within the scope of the present application.

The following describes in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.

Fig. 1 is a flow chart of a method for processing user data according to an embodiment of the present application. The processing method of the user data is as follows.

S102: behavior data of multiple types of users are obtained.

In S102, multiple classes of users may be classified based on service tags, which may be determined based on specific application scenarios, for example, in risk identification scenarios, the service tags may include high risk and low risk, or may include high risk, medium risk and low risk; in a marketing scenario, business labels may include targeted users and non-targeted users.

The behavior data of the multiple types of users may be behavior data of the multiple types of users within a set period of time, which may be determined according to actual needs, and is not particularly limited herein.

The method for obtaining behavior data of multiple types of users in the embodiment can include the following steps:

first, behavior data of a plurality of users having different service tags is acquired.

Taking a user as an example, when the behavior data of the user is obtained, the behavior data of the user may include a plurality of behavior events and time sequence identifiers of the plurality of behavior events, where the time sequence identifiers of the behavior events may represent the occurrence sequence of the behavior events.

For example, the behavior events of the user a in a certain period of time include event 1, event 2, event 3, event 4 and event 5, where the time sequence identifiers of the events 1 to 5 are 2,4,1,3,5 in sequence, and then the time sequence of the events 1 to 5 is: event 3, event 1, event 4, event 2, event 5.

It should be noted that, in the acquired behavior data of multiple users, behavior events included in the behavior data of different users may be the same or different, and timing identifiers of the same behavior event in the behavior data of different users may be the same or different, where the number of times the same behavior event occurs in the behavior data of one user may be one or multiple times, and the method is not specifically limited herein.

Secondly, grouping behavior data of a plurality of users according to the service labels to obtain behavior data of a plurality of types of users.

When grouping, the behavior data of users with the same service labels can be divided into a group to obtain the behavior data of one type of users, so that the behavior data of multiple types of users can be obtained based on multiple different service labels.

For example, in the risk identification scenario, the behavior data of the high-risk user may be divided into a group according to the high risk and the low risk of the service label, and the behavior data of the low-risk user may be divided into a group, so as to obtain the behavior data of the two types of users.

After the behavior data of the multiple types of users is acquired, S104 may be performed.

S104: and determining a time sequence association rule corresponding to the multiple types of users based on the behavior events and the time sequence identifications of the behavior events included in the behavior data of the multiple types of users.

In S104, when determining the timing association rule corresponding to the multiple types of users, taking one of the types of users as an example, the method may include the following steps:

first, a frequent item set is determined based on behavioral events included in behavioral data of a class of users.

In this embodiment, the number of frequent item sets may be plural, and when determining the frequent item sets, the following principle may be followed:

one behavioral event occurs one or more times in the behavioral data of one user, each counted as one, and one frequent item set includes at least one behavioral event that occurs at least in the behavioral data of one user.

And secondly, determining a time sequence association rule corresponding to one type of user based on the frequent item set and time sequence identification of the behavior event included in the frequent item set.

Specifically, taking one frequent item set as an example, when determining the time sequence association rule, determining a user corresponding to the frequent item set, wherein the user corresponding to the frequent item set can be understood as a user including a behavior event in the frequent item set in behavior data; after the user corresponding to the frequent item set is obtained, the time sequence association rule corresponding to the frequent item set can be obtained according to the time sequence identification of the behavior event in the frequent item set in the behavior data of the user corresponding to the frequent item set.

After the time sequence association rule corresponding to one frequent item set is obtained, the time sequence association rules corresponding to a plurality of frequent item sets can be obtained based on the same method, and the time sequence association rules corresponding to the plurality of frequent item sets are the time sequence association rules corresponding to a class of users.

For ease of understanding, the following description will be given by taking 3 users included in one class of users as examples.

Assuming that 3 users A, B, C are included in one class of users, the behavior events and the timing identifications of the behavior events included in the behavior data of the users A, B and C are shown in table 1.

TABLE 1

User' s	Behavioral events	Time sequence identification of behavior events
			A	1	1
A	2	2
			A	3	3
A	4	4
			B	1	1
B	2	2
			B	3	3
B	4	4
			C	1	1
C	3	2
			C	5	3

In determining the timing association rule corresponding to the user A, B, C, first, based on the behavior event of the user A, B, C in table 1, the following frequent item sets may be obtained:

(1)，(2)，(3)，(4)，(5)，(1，2)，(1，3)，(1，4)，(1，5)，(2，3)，(2，4)，(3，4)，(3，5)，(1，2，3)，(1，2，4)，(1，3，4)，(1，3，5)，(2，3，4)，(1，2，3，4)。

secondly, taking one of the frequent item sets (1, 2) as an example, the user corresponding to the frequent item set can be determined to be the user A and the user B, and based on the time sequence identification of the event 1 and the event 2 in the behavior data of the user A and the user B, the time sequence association rule 1- & gt2 (& gtcan also be expressed by other modes) corresponding to the frequent item set (1, 2) can be obtained.

For other frequent item sets, corresponding time sequence association rules can be obtained based on the same method, and then the time sequence association rules corresponding to the user A, B, C are obtained:

1，2，3，4，5，1→2，1→3，1→4，1→5，2→3，2→4，3→4，3→5，1→2→3，1→2→4，1→3→4，1→3→5，2→3→4，1→2→3→4。

Optionally, in the process of determining the time sequence association rule, after obtaining the plurality of frequent item sets, the plurality of frequent item sets may be further filtered to filter the frequent item sets with a length less than or equal to a preset threshold, where the length of the frequent item sets may be understood as the number of behavior events included in the frequent item sets, and the preset threshold may be determined according to actual needs.

After the multiple frequent item sets are screened, a time sequence association rule can be obtained based on the screened frequent item sets and time sequence identifiers of the behavior events included in the screened frequent item sets.

Taking the 3 users A, B, C as an example, after obtaining a plurality of frequent item sets, the frequent item sets with the length of 1 in the plurality of frequent item sets can be filtered to obtain the frequent item sets with the length equal to or greater than 2, and based on the frequent item sets and the time sequence identifications of the behavior events included in the frequent item sets, the following time sequence association rule can be obtained:

1→2，1→3，1→4，1→5，2→3，2→4，3→4，3→5，1→2→3，1→2→4，1→3→4，1→3→5，2→3→4，1→2→3→4。

after the time sequence association rules corresponding to one class of users are obtained based on the method, the time sequence association rules corresponding to other classes of users can be obtained based on the same method.

After the timing association rule corresponding to the multiple types of users is obtained, S106 may be performed.

S106: and determining a time sequence behavior rule of the target class user based on a first time sequence association rule corresponding to the target class user and a second time sequence association rule corresponding to the non-target class user in the multiple classes of users.

In S106, the target class user may be understood as a class of users of the multiple classes of users for which the timing behavior rule is to be determined, and the non-target class user may be understood as other class of users of the multiple classes of users except the target class user. Here, in order to facilitate distinguishing between the timing association rules corresponding to different classes of users, the timing association rule corresponding to the target class of users may be represented as a first timing association rule, and the timing association rule corresponding to the non-target class of users may be represented as a second timing association rule.

In this embodiment, the timing behavior rule corresponding to the target class user may be understood as a timing rule satisfied by the behavior event of the target class user. When determining the time sequence behavior rule of the target class user, the method can comprise the following steps:

a target timing association rule of the first timing association rules is determined based on the first timing association rule and the second timing association rule.

And determining the time sequence behavior rule of the target class user based on the target time sequence association rule.

In determining the target timing association rule, at least the following three methods may be adopted:

the first method is as follows:

and taking the time sequence association rule which does not belong to the second time sequence association rule in the first time sequence association rule as a target time sequence association rule.

For example, assume that the first timing association rule is: 1- > 2, 1- > 3, 1- > 4, 1- > 5, 2- > 3, 2- > 4, 3- > 5, 1- > 2- > 3, 1- > 2- > 4, 1- > 3- > 2- > 3- > 4, 1- > 2- > 3- > 4, the second timing association rule is: 1→3,1→4,1→5,3→4,3→5,4→5,3→4→5,1→3→4→5, then 1→2,2→3,2→4,1→2→3,1→4,1→3→4,1→3→5,2→3→4,1→2→3→4 in the first time sequence association rule which does not belong to the second time sequence association rule can be regarded as the target time sequence association rule.

The second method is as follows:

determining the same time sequence association rule in the first time sequence association rule and the second time sequence association rule; determining a first credibility of the same time sequence association rule in the target class user and a second credibility of the same time sequence association rule in the non-target class user; and determining a time sequence association rule with the ratio of the first credibility to the second credibility not smaller than a set threshold value as a target time sequence association rule in the same time sequence association rules.

In this embodiment, the first reliability and the second reliability may be determined based on the same method. The first reliability may be a probability that a behavior event corresponding to the time sequence association rule occurs in behavior data of the target class user, and the probability that the behavior event occurs in the target class user may be defined as a ratio of a number of times that the behavior event occurs in the behavior data of the target class user to a total number of users of the target class user.

For example, for one time-series association rule 1→2→3 of the user A, B, C described above, the corresponding behavior events are event 1, event 2, and event 3, and it is known from table 1 that the number of times of occurrence of event 1, event 2, and event 3 in the behavior data of the user A, B, C is 2, and the total number of users is 3, so that the probability of occurrence of event 1, event 2, and event 3 is 0.67,0.67 can be regarded as the first confidence of the time-series association rule 1→2→3.

In addition, the first reliability may be a conditional probability that a behavior event corresponding to the time sequence association rule occurs in behavior data of the target class user. Taking the time sequence association rule a-b-c as an example, the conditional probability that the behavior events a, b and c corresponding to the time sequence association rule a-b-c occur in the behavior data of the target class user can be defined as the ratio of the probability that the behavior events a, b and c occur in the behavior data of the target class user to the probability that the behavior events a and b occur in the behavior data of the target class user.

For example, with respect to the above-described time-series association rule 1→2→3 of the user A, B, C, the corresponding behavior events are event 1, event 2, and event 3, and it is known from table 1 that the occurrence probabilities of event 1, event 2, and event 3 are 0.67, and the occurrence probabilities of event 1 and event 2 are 0.67, and therefore, the occurrence conditional probabilities of event 1, event 2, and event 3 are 1,1 can be regarded as the first confidence of the time-series association rule 1→2→3.

After the first reliability and the second reliability of the timing association rule are obtained based on the above-described method, the target timing association rule may be determined based on the second method. In determining the target timing association rule, for ease of understanding, it may be exemplified.

Assume that the first timing association rule and the first confidence level are: 1→2 (0.67), 1→3 (1), 1→4 (0.67), 1→5 (0.33), 2→3 (0.67), 2→4 (0.67), 3→4 (0.67), 3→5 (0.33), 1→2→3 (0.67), 1→2→4 (0.67), 1→3→4 (0.67), 1→3→5 (0.33), 2→3→4 (0.67), 1→2→3→4 (0.67), the second timing association rule and the second reliability are: 1- > 3 (0.5), 1- > 4 (0.5), 1- > 5 (0.5), 3- > 4 (0.5), 3- > 5 (0.5), 4- > 5 (1), 3- > 4- > 5 (0.5), 1- > 3- > 4- > 5 (0.5).

From this, the same time sequence association rule in the target class user and the non-target class user is 1→3,1→4,1→5,3→4,3→5, and the ratio of the first reliability and the second reliability of these time sequence association rules is 2,1.33,0.67,1.33,0.67 in turn, and if the set threshold is 1, 1→3,1→4,3→4 can be used as the target time sequence association rule.

The third method is as follows:

the combination of the target timing association rule obtained by the first method and the second method described above is used as the target timing association rule. Specific embodiments will not be described in detail herein.

It should be noted that, for the three methods described above, in practical application, any method may be selected to determine the target timing association rule, where, preferably, if there is a timing association rule that does not belong to the second timing association rule in the first timing association rule, the first method or the third method may be selected to determine the target timing association rule, otherwise, the second method may be selected to determine the target timing association rule.

After the target time sequence association rule is obtained based on the method, the time sequence behavior rule of the target class user can be determined based on the target time sequence association rule.

When determining the time sequence behavior rule of the target class user, one implementation way may be to directly use the target time sequence association rule as the time sequence behavior rule corresponding to the target class user. For example, if the target timing relationship is: 1- & gt 2, 2- & gt 3, 2- & gt 4, 1- & gt 2- & gt 3, 1- & gt 2- & gt 4, 1- & gt 3- & gt 5, 2- & gt 3- & gt 4, 1- & gt 2- & gt 3- & gt 4, the time sequence behavior rule of the target class user is as follows: 1- > 2, 2- > 3, 2- > 4, 1- > 2- > 3, 1- > 2- > 4, 1- > 3- > 5, 2- > 3- > 4, 1- > 2- > 3- > 4.

Another implementation manner may be to take a target timing association rule with a reliability not smaller than a set reliability threshold as a timing behavior rule of the target class user.

For example, if the target timing relationship is: 1- > 2, 2- > 3, 2- > 4, 1- > 2- > 3, 1- > 2- > 4, 1- > 3- > 5, 2- > 3- > 4, 1- > 2- > 3- > 4, the confidence level is 0.67,0.67,0.67,0.67,0.67,0.67,0.33,0.67,0.67 in turn, and assuming that the confidence level threshold is set to 0.5, the timing behavior rule of the target class user may be: 1- > 2, 2- > 3, 2- > 4, 1- > 2- > 3, 1- > 2- > 4, 1- > 3- > 4, 2- > 3- > 4, 1- > 2- > 3- > 4.

In this embodiment, any implementation manner may be selected to determine the timing behavior rule of the target class user, which is not specifically limited herein.

After the timing behavior rule of the target class user is obtained, S108 may be performed.

S108: and determining whether the user to be identified belongs to the target class user or not based on the time sequence behavior rule of the target class user.

The user to be identified can be understood as a user without a service tag, and when determining whether the user to be identified belongs to the target class user based on the time sequence behavior rule of the target class user, the method can comprise the following steps:

firstly, behavior data of a user to be identified is obtained.

The behavior data of the user to be identified can comprise a plurality of behavior events of the user to be identified in a set time period and time sequence identifications of the behavior events.

And secondly, determining a time sequence association rule of the user to be identified according to the behavior event and the time sequence identifier of the behavior event included in the behavior data of the user to be identified.

When determining the time sequence association rule of the user to be identified, the specific implementation manner can refer to the content of the time sequence association rule corresponding to the user to be identified, which is described above, and the description is not repeated here.

And finally, judging whether the time sequence association rule of the user to be identified hits the time sequence behavior rule of the target class user, if so, determining that the user to be identified belongs to the target class user, and if not, determining that the user to be identified belongs to the non-target class user.

For example, the timing behavior rules of the target class user are known as: 1→2,2→3,2→4,1→2→3,1→2→4,1→3→4,2→3→4,1→2→3→4, then:

assume that the timing association rule of the user to be identified is: 3- > 4, 1- > 2- > 3, 1- > 2- > 4, 1- > 3- > 4, and by comparing, it can be known that the time sequence association rules 1- > 2- > 3, 1- > 2- > 4 and 1- > 3- > 4 of the user to be identified hit the time sequence behavior rules 1- > 2- > 3, 1- > 2- > 4 and 1- > 3- > 4 of the target class user, so that the user to be identified can be determined to belong to the target class user.

Assume that the user timing association rule to be identified is: 4-5, 2-3-5, 2-4-5, 3-4-5, and comparing to determine that any time sequence association rule of the user to be identified does not hit any time sequence behavior rule of the target class user, thereby determining that the user to be identified does not belong to the target class user.

In this embodiment, after determining that the user to be identified belongs to the target class user, the user to be identified may be further processed based on the application scenario.

For example, in the risk identification scenario, if it is determined that the user a to be identified belongs to a high risk user, the user a may be directly intercepted, and the user a is not allowed to access a security page or conduct online transactions; if the user A is determined to belong to the low risk user, the user A can be directly admitted, and the user A is allowed to access a security page or conduct online transactions and the like.

In this way, in the risk identification scene, after the user to be identified is identified based on the time sequence behavior rule, the user to be identified can be intercepted or admitted based on the identification result without other judgment, so that the system pressure can be reduced, in addition, after the user to be identified is determined to be a low risk user based on the time sequence behavior rule, the user to be identified can be admitted without other judgment processing, the whole process consumes less time, and therefore the user experience can be improved.

For another example, in the marketing scenario, if it is determined that the user a to be identified belongs to the target user, commodity information of the target commodity may be pushed to the user a; if it is determined that the user a does not belong to the target user, the recommendation of the commodity information of the target commodity to the user a may be canceled. Therefore, accurate pushing can be realized, unnecessary commodity information is prevented from being recommended to a user, and user experience is improved.

According to the technical scheme provided by the embodiment of the application, when the behavior data of the user is analyzed, the time sequence identification of the behavior event included in the behavior data is fully considered, so that the behavior data of the user can be fully mined; when the behavior data of the multiple types of users are analyzed, the behavior data of the multiple types of users can be fully mined to obtain a time sequence association rule capable of reflecting the time sequence of the behaviors of the multiple types of users, and further the time sequence behavior rule of the target type of users is obtained, so that effective identification can be performed when whether the users to be identified belong to the target type of users or not is determined based on the time sequence behavior rule of the target type of users.

Fig. 2 is a flow chart of a method for processing user data according to an embodiment of the present application, where the method for processing user data includes the following steps.

S202: behavior data of multiple types of users are obtained.

The behavior data of one user can comprise a plurality of behavior events and time sequence identifiers of the behavior events.

S204: determining a time sequence association rule corresponding to the multiple types of users based on the behavior events and time sequence identifiers of the behavior events included in the behavior data of the multiple types of users; :

the specific implementation of S202 to S204 may refer to the specific implementation of the corresponding steps in the embodiment shown in fig. 1, and the description is not repeated here.

S206: and determining the time sequence behavior rule of the multi-class user based on the time sequence association rule corresponding to the multi-class user.

In S206, for one of the target class users, the timing behavior rule of the target class user may be determined based on the method described in the embodiment shown in fig. 1, and after the timing behavior rule of the target class user is obtained, the timing behavior rules of non-target class users other than the target class user in the multiple class users may be determined based on the same method.

S208: and generating a plurality of time sequence behavior rule features based on the time sequence behavior rules of the multiple types of users.

In this embodiment, one time sequence behavior rule may generate one time sequence behavior rule feature, and the number of the plurality of time sequence behavior rule features may be equal to the total number of time sequence behavior rules of multiple types of users.

For example, the timing behavior rules for multiple classes of users include: 1→2,2→3,2→4,1→2→3,1→2→4,1→3→4,1→3→5,2→3→4,1→2→3→4,4→5→5,1→3→4→5, then the timing behavior rule feature fea1 can be generated based on the timing behavior rule 1→2, the timing behavior rule features fea2, … … can be generated based on the timing behavior rule 2→3, and the like, 12 timing behavior rule features fea1 to fea12 can be obtained.

S210: and determining time sequence behavior characteristics of the user to be processed based on the time sequence behavior rule characteristics so as to generate sample data comprising the time sequence behavior characteristics for model training.

In S210, when determining the time series behavior feature of the user to be processed based on the plurality of time series behavior rule features, the method may include the steps of:

firstly, behavior data of a user to be processed is obtained.

The behavior data of the user to be processed can comprise a plurality of behavior events of the user to be processed in a set time period and time sequence identifications of the behavior events.

Secondly, determining a time sequence association rule of the user based on the behavior event and the time sequence identification of the behavior event included in the behavior data of the user to be processed.

The specific implementation manner may be referred to the specific content of determining the time sequence association rule corresponding to the user in the embodiment shown in fig. 1, and the description is not repeated here.

Finally, for one of the timing behavior rule features, it may be determined whether the timing association rule of the user to be processed hits the timing behavior rule corresponding to the timing behavior rule feature, where the timing behavior rule corresponding to the timing behavior rule feature may be understood as the timing behavior rule generating the timing behavior rule feature. For example, if the above-described fea1 is generated by the time series behavior rule 1→2, the time series behavior rule corresponding to fea1 is 1→2.

If the judgment result is yes, the user to be processed can be determined to have the time sequence behavior rule characteristic, otherwise, the user to be processed can be determined to not have the time sequence behavior rule characteristic.

After determining whether the user to be processed has one time series behavior rule feature based on the above-described method, it may be determined whether the user to be processed has other time series behavior rule features based on the same method, which will not be described in detail herein.

In this embodiment, whether the user to be processed has a plurality of time sequence behavior rule features may be represented by a numerical value, and the numerical value may be regarded as the time sequence behavior feature of the user to be processed.

For example, if the user to be processed has a certain time sequence behavior rule feature, the time sequence behavior rule feature can be represented by a value 1; if the user to be processed does not have a certain time sequence behavior rule feature, the time sequence behavior rule feature can be represented by a value 0, and the values 0 and 1 can be regarded as the time sequence behavior feature of the user to be processed.

For ease of understanding, the above description will be given taking the 12 timing behavior rule features fea1 to fea12 as an example, and the timing behavior rules corresponding to the 12 timing behavior rule features fea1 to fea12 are: 1- > 2, 2- > 3, 2- > 4, 1- > 2- > 3, 1- > 2- > 4, 1- > 3- > 5, 2- > 3- > 4, 1- > 2- > 3- > 4, 4- > 5, 3- > 4- > 5, 1- > 3- > 4- > 5.

The timing association rule for the user to be processed is assumed to include: 1- > 2, 2- > 4, 3- > 4, 1- > 2- > 3, 1- > 3- > 4, 1- > 3- > 5, 1- > 2- > 3- > 4, it can be determined that the user to be processed has a timing behavior rule feature fea1, fea3, fea4, fea6, fea7, fea9, does not have a timing behavior rule feature fea2, fea5, fea8, fea10, fea11, fea12, 1 indicates that the user to be processed has a certain timing behavior rule feature, and 0 indicates that the user to be processed does not have a certain timing behavior rule feature, and the timing behavior feature of the user to be processed can be obtained as follows: 1,0,1,1,0,1,1,0,1,0,0,0.

In this embodiment, after determining to obtain the time sequence behavior feature of the user to be processed, under the condition that the service tag of the user to be processed is known, sample data may also be obtained based on the time sequence behavior feature of the user to be processed, where the sample data includes the time sequence behavior feature of the user to be processed, and the sample data may be used to perform model training.

For example, after obtaining the time series behavior characteristics of the target class user and the non-target class user, the time series behavior characteristics may be used as sample data to perform model training, so as to obtain a model for identifying whether the user is the target class user or the non-target class user. After the model is obtained, when the user to be identified is identified based on the model, the time sequence behavior characteristics of the user to be identified can be determined based on the method recorded in the embodiment, the time sequence behavior characteristics of the user to be identified are used as model input, and whether the user to be identified belongs to the target type user or the non-target type user can be determined based on the output of the model.

According to the technical scheme provided by the embodiment of the application, when the behavior data of the user is analyzed, the time sequence identification of the behavior event included in the behavior data is fully considered, so that the behavior data of the user can be fully mined; when the behavior data of the multiple types of users are analyzed, the behavior data of the multiple types of users can be fully mined to obtain time sequence association rules capable of reflecting time sequences of the behaviors of the multiple types of users, further the time sequence behavior rules of the multiple types of users are obtained, and a plurality of time sequence behavior rule features are generated based on the time sequence behavior rules of the multiple types of users, so that the time sequence behavior features of the users can be obtained more abundantly based on the time sequence behavior rule features, and after a model is carried out based on sample data comprising the time sequence behavior features, the model obtained by training can be effectively used for decision processing.

The foregoing describes specific embodiments of the present application. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present application. Referring to fig. 3, at the hardware level, the electronic device includes a processor, and optionally an internal bus, a network interface, and a memory. The Memory may include a Memory, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory (non-volatile Memory), such as at least 1 disk Memory. Of course, the electronic device may also include hardware required for other services.

The processor, network interface, and memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or EISA (Extended Industry Standard Architecture ) bus, among others. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one bi-directional arrow is shown in FIG. 3, but not only one bus or type of bus.

And the memory is used for storing programs. In particular, the program may include program code including computer-operating instructions. The memory may include memory and non-volatile storage and provide instructions and data to the processor.

The processor reads the corresponding computer program from the nonvolatile memory into the memory and then runs to form the processing device of the user data on the logic level. The processor is used for executing the programs stored in the memory and is specifically used for executing the following operations:

The method performed by the user data processing device disclosed in the embodiment shown in fig. 3 of the present application may be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the embodiments of the present application may be embodied directly in hardware, in a decoded processor, or in a combination of hardware and software modules in a decoded processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method.

The electronic device may also execute the method of fig. 1 and implement the functions of the processing device for user data in the embodiment shown in fig. 1, which are not described herein.

Of course, other implementations, such as a logic device or a combination of hardware and software, are not excluded from the electronic device of the present application, that is, the execution subject of the following processing flow is not limited to each logic unit, but may be hardware or a logic device.

The present embodiments also provide a computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a portable electronic device comprising a plurality of application programs, enable the portable electronic device to perform the method of the embodiment of fig. 1, and in particular to:

Fig. 4 is a schematic structural diagram of a processing device 40 for user data according to an embodiment of the present application. Referring to fig. 4, in a software implementation, the processing device 40 for user data may include: an acquisition unit 41, a processing unit 42, a determination unit 43, and an identification unit 44, wherein:

an obtaining unit 41, configured to obtain behavior data of multiple types of users, where the multiple types of users are divided based on service labels, and the behavior data of one user includes multiple behavior events and time sequence identifiers of the multiple behavior events;

a processing unit 42, configured to determine a time sequence association rule corresponding to the multiple types of users based on the behavior events and time sequence identifiers of the behavior events included in the behavior data of the multiple types of users;

a determining unit 43, configured to determine a time sequence behavior rule of a target class user based on a first time sequence association rule corresponding to the target class user and a second time sequence association rule corresponding to a non-target class user in the multiple classes of users;

The identifying unit 44 determines whether the user to be identified belongs to the target class user based on the time series behavior rule of the target class user.

Optionally, the determining unit 43 determines a timing behavior rule of the target class user, including:

determining a target timing association rule in the first timing association rule based on the first timing association rule and the second timing association rule;

Optionally, the determining unit 43 determines a target timing association rule in the first timing association rule, including at least one of the following:

determining a time sequence association rule which does not belong to the second time sequence association rule in the first time sequence association rule as the target time sequence association rule;

determining the same time sequence association rule in the first time sequence association rule and the second time sequence association rule; determining a first credibility of the same time sequence association rule in the target class user and a second credibility of the same time sequence association rule in the non-target class user; and determining a time sequence association rule that the ratio of the first credibility to the second credibility is not smaller than a set threshold value as the target time sequence association rule.

Optionally, the first reliability is a probability or a conditional probability that a behavior event corresponding to the time sequence association rule occurs in the behavior data of the target class user.

Optionally, the identifying unit 44 determines, based on a time sequence behavior rule of the target class user, whether the user to be identified belongs to the target class user, including:

acquiring behavior data of the user to be identified, wherein the behavior data comprises a plurality of behavior events and time sequence identifiers of the behavior events;

determining a time sequence association rule of the user to be identified based on the behavior event and the time sequence identifier of the behavior event included in the behavior data of the user to be identified;

judging whether the time sequence association rule of the user to be identified hits the time sequence behavior rule of the target class user or not;

if yes, determining that the user belongs to the target class user;

if not, determining that the user does not belong to the target class user.

Optionally, the processing unit 42 determines a timing association rule corresponding to the multiple types of users, including:

for one of the classes of users, the following operations are performed:

determining a frequent item set based on behavioral events included in the behavioral data of the class of users;

And determining a time sequence association rule corresponding to the user based on the frequent item set and the time sequence identification of the behavior event included in the frequent item set.

The processing device 40 for user data provided in the embodiment of the present application may also execute the method of fig. 1 and implement the functions of the processing device for user data in the embodiment shown in fig. 1, which is not described herein again.

Fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application. Referring to fig. 5, at the hardware level, the electronic device includes a processor, and optionally an internal bus, a network interface, and a memory. The Memory may include a Memory, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory (non-volatile Memory), such as at least 1 disk Memory. Of course, the electronic device may also include hardware required for other services.

The processor, network interface, and memory may be interconnected by an internal bus, which may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component Interconnect, peripheral component interconnect standard) bus, or EISA (Extended Industry Standard Architecture ) bus, among others. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one bi-directional arrow is shown in FIG. 5, but not only one bus or type of bus.

The method performed by the user data processing apparatus disclosed in the embodiment shown in fig. 5 of the present application may be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or by instructions in the form of software. The processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present application may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the embodiments of the present application may be embodied directly in hardware, in a decoded processor, or in a combination of hardware and software modules in a decoded processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The storage medium is located in a memory, and the processor reads the information in the memory and, in combination with its hardware, performs the steps of the above method.

The electronic device may also execute the method of fig. 2 and implement the functions of the processing device for user data in the embodiment shown in fig. 2, which are not described herein.

The present embodiments also provide a computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a portable electronic device comprising a plurality of application programs, enable the portable electronic device to perform the method of the embodiment of fig. 2, and in particular to:

Fig. 6 is a schematic structural diagram of a processing device 60 for user data according to an embodiment of the present application. Referring to fig. 6, in a software implementation, the processing device 40 for user data may include: an acquisition unit 61, a processing unit 62, a first determination unit 63, a generation unit 64, and a second determination unit 65, wherein:

an obtaining unit 61, configured to obtain behavior data of multiple types of users, where the multiple types of users are divided based on service labels, and the behavior data of one user includes multiple behavior events and time sequence identifiers of the multiple behavior events;

the processing unit 62 determines a time sequence association rule corresponding to the multiple types of users based on the behavior events and time sequence identifiers of the behavior events included in the behavior data of the multiple types of users;

a first determining unit 63 configured to determine a timing behavior rule of the multiple types of users based on the timing association rule corresponding to the multiple types of users;

A generating unit 64 that generates a plurality of timing behavior rule features based on the timing behavior rules of the plurality of types of users;

the second determining unit 65 determines time series behavior characteristics of the user to be processed based on the plurality of time series behavior rule characteristics to generate sample data including the time series behavior characteristics for model training.

Optionally, the second determining unit 65 determines, based on the plurality of timing behavior rule features, timing behavior features of the user to be processed, including:

acquiring behavior data of the user to be processed, wherein the behavior data comprises a plurality of behavior events and time sequence identifiers of the behavior events;

determining a time sequence association rule of the user to be processed based on the behavior event and the time sequence identifier of the behavior event included in the behavior data of the user to be processed;

for one of the timing behavior rule features, the following is performed:

judging whether the time sequence association rule of the user to be processed hits the time sequence behavior rule corresponding to the time sequence behavior rule characteristic or not;

if yes, determining that the user to be processed has the time sequence behavior rule characteristics;

if not, determining that the user to be processed does not have the time sequence behavior rule characteristic.

The processing device 60 for user data provided in the embodiment of the present application may also execute the method of fig. 2, and implement the functions of the processing device for user data in the embodiment shown in fig. 2, which is not described herein again.

In summary, the foregoing description is only a preferred embodiment of the present application, and is not intended to limit the scope of the present application. 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.

The system, apparatus, module or unit set forth in the above embodiments may be implemented in particular by a computer chip or entity, or by a product having a certain function. One typical implementation is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises the element.

All embodiments in the application are described in a progressive manner, and identical and similar parts of all embodiments are mutually referred, so that each embodiment mainly describes differences from other embodiments. In particular, for system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, as relevant to see a section of the description of method embodiments.

Claims

1. A method of processing user data, comprising:

acquiring behavior data of multiple types of users;

determining a time sequence association rule corresponding to the multiple types of users based on the behavior data of the multiple types of users;

Determining a target time sequence association rule in the first time sequence association rule based on a first time sequence association rule corresponding to a target user in the multiple types of users and a second time sequence association rule corresponding to a non-target user;

determining a time sequence behavior rule of the target class user according to the target time sequence association rule;

2. The method of claim 1, determining a target timing association rule of the first timing association rules, comprising at least one of:

3. The method according to claim 2,

the first credibility is the probability or the conditional probability that the behavior event corresponding to the time sequence association rule appears in the behavior data of the target class user.

4. The method of claim 1, determining a timing behavior rule for the target class of users based on the target timing association rule, comprising:

determining a target time sequence association rule with the reliability not smaller than a set reliability threshold value in the target time sequence association rule as a time sequence behavior rule of the target class user, wherein the reliability of the target time sequence association rule is the probability or conditional probability that a behavior event corresponding to the target time sequence association rule appears in behavior data of the target class user; or alternatively, the first and second heat exchangers may be,

and determining the target time sequence association rule as the time sequence behavior rule of the target class user.

5. The method of claim 1, determining whether a user to be identified belongs to the target class user based on a time-sequential behavior rule of the target class user, comprising:

if yes, determining that the user belongs to the target class user;

if not, determining that the user does not belong to the target class user.

6. The method of claim 1, the multiple classes of users being based on traffic label partitioning; obtaining behavior data of multiple types of users, including:

acquiring behavior data of a plurality of users with different service labels;

and grouping the behavior data of the plurality of users according to the service labels to obtain the behavior data of the plurality of types of users.

7. The method of claim 1, wherein the behavior data of a user includes a plurality of behavior events and a timing identification of the plurality of behavior events; based on the behavior data of the multiple types of users, determining a time sequence association rule corresponding to the multiple types of users comprises the following steps:

and determining a time sequence association rule corresponding to the multiple types of users based on the behavior events and the time sequence identifications of the behavior events included in the behavior data of the multiple types of users.

8. The method of claim 7, determining a timing association rule for the plurality of classes of users, comprising:

for one of the classes of users, the following operations are performed:

9. The method of claim 1, further comprising at least one of:

under the condition that the target class user is a high-risk user and the user to be identified belongs to the target class user, intercepting the user to be identified;

performing admission processing on the user to be identified under the condition that the target user is a low-risk user and the user to be identified belongs to the target user;

pushing commodity information of a target commodity to the user to be identified under the condition that the target user is a target user and the user to be identified belongs to the target user;

and under the condition that the target class user is a target user and the user to be identified does not belong to the target class user, pushing commodity information of target commodities to the user to be identified is canceled.

10. A method of processing user data, comprising:

acquiring behavior data of multiple types of users;

determining time sequence behavior characteristics of a user to be processed based on time sequence behavior rules of the multiple types of users so as to generate sample data comprising the time sequence behavior characteristics for model training;

wherein determining the timing behavior rules of the multiple classes of users comprises:

determining target time sequence association rules in first time sequence association rules based on first time sequence association rules corresponding to target users and second time sequence association rules corresponding to non-target users aiming at any target users in multiple types of users; and determining the time sequence behavior rule of the target class user according to the target time sequence association rule.

11. The method of claim 10, determining a temporal behavior characteristic of a user to be processed based on a temporal behavior rule of the plurality of classes of users, comprising:

And determining the time sequence behavior characteristics of the user to be processed based on the time sequence behavior rule characteristics.

12. The method of claim 11, determining a temporal behavior feature of a user to be processed based on the plurality of temporal behavior rule features, comprising:

for one of the timing behavior rule features, the following is performed:

13. A processing apparatus for user data, comprising:

the acquisition unit acquires behavior data of multiple types of users;

the processing unit is used for determining time sequence association rules corresponding to the multiple types of users based on the behavior data of the multiple types of users;

The determining unit is used for determining target time sequence association rules in the first time sequence association rules based on first time sequence association rules corresponding to target users in the multiple types of users and second time sequence association rules corresponding to non-target users; determining a time sequence behavior rule of the target class user according to the target time sequence association rule;

14. A processing apparatus for user data, comprising:

the acquisition unit acquires behavior data of multiple types of users;

a second determining unit for determining time sequence behavior characteristics of the users to be processed based on the time sequence behavior rules of the multiple types of users so as to generate sample data comprising the time sequence behavior characteristics for model training;

the first determining unit determines a time sequence behavior rule of the multiple types of users, and includes:

15. An electronic device, comprising:

a processor; and

acquiring behavior data of multiple types of users;

16. A computer readable storage medium storing one or more programs, which when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the method of:

acquiring behavior data of multiple types of users;

17. An electronic device, comprising:

a processor; and

acquiring behavior data of multiple types of users;

18. A computer readable storage medium storing one or more programs, which when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform the method of:

acquiring behavior data of multiple types of users;