CN110163683B - Value user key index determination method, advertisement delivery method and device - Google Patents

Value user key index determination method, advertisement delivery method and device Download PDF

Info

Publication number
CN110163683B
CN110163683B CN201910441219.8A CN201910441219A CN110163683B CN 110163683 B CN110163683 B CN 110163683B CN 201910441219 A CN201910441219 A CN 201910441219A CN 110163683 B CN110163683 B CN 110163683B
Authority
CN
China
Prior art keywords
user
behavior event
user behavior
client
value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910441219.8A
Other languages
Chinese (zh)
Other versions
CN110163683A (en
Inventor
焦淑尧
高原
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Lexin Shengwen Technology Co ltd
Original Assignee
Beijing Lexin Shengwen Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Lexin Shengwen Technology Co ltd filed Critical Beijing Lexin Shengwen Technology Co ltd
Priority to CN201910441219.8A priority Critical patent/CN110163683B/en
Publication of CN110163683A publication Critical patent/CN110163683A/en
Application granted granted Critical
Publication of CN110163683B publication Critical patent/CN110163683B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0269—Targeted advertisements based on user profile or attribute

Landscapes

  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

The invention provides a method for determining key indexes of value users, a method and a device for putting advertisements, wherein the method for determining the key indexes of the value users comprises the following steps: receiving a user behavior event from a client, and storing the user behavior event; determining a category to which the stored user behavior event belongs, wherein the category is a valuable user or a non-valuable user; performing regression analysis by using the user behavior event after the category to which the user behavior event belongs is determined to obtain the probability that the user of the client with the relevant information of the user behavior event is the value user; and determining whether the relevant information of the user behavior event is a key index of the value user according to the probability. By the scheme, the key indexes of the potential value users can be automatically determined, so that the workload of analyzing data is reduced and the error rate is reduced under the condition of improving the advertisement putting accuracy.

Description

Value user key index determination method, advertisement delivery method and device
Technical Field
The invention relates to the technical field of internet, in particular to a method for determining key indexes of value users, a method and a device for putting advertisements.
Background
Currently, for one application product, advertisements may be delivered on different platforms at the same time, e.g., Google, Facebook, etc. After the advertisement is placed for a period of time, each platform can perform data analysis respectively to obtain respective conclusions, such as retention rate or paid user rate under different factors. Since the analysis data of different platforms may be affected by different factors and the data of different platforms cannot be communicated with each other, the conclusions drawn by different platforms are different in general. Then, the conclusion of which platform is correct and which factors are the key indicators of user retention or payment cannot be automatically known. The core of the existing realization mode is that manual switching is performed to different platforms, the platforms display data to analysts, the analysts analyze the data and gather conclusions, high-income events are found, and then advertisements are selectively released by releasing personnel.
However, the analysts spend a lot of time analyzing the conclusions of each platform, and face many communication problems that require cross-functional collaboration, with the risk of errors.
Disclosure of Invention
In view of the above, the invention provides a method for determining key indexes of value users, an advertisement delivery method and an apparatus thereof, so as to automatically determine key indexes of potential value users, thereby reducing workload of analyzing data and reducing error rate under the condition of improving advertisement delivery accuracy.
According to an aspect of an embodiment of the present invention, a method for determining a key index of a value user is provided, including:
receiving a user behavior event from a client, and storing the user behavior event;
determining a category to which the stored user behavior event belongs, wherein the category is a valuable user or a non-valuable user;
performing regression analysis by using the user behavior event after the category to which the user behavior event belongs is determined to obtain the probability that the user of the client with the relevant information of the user behavior event is the value user;
and determining whether the relevant information of the user behavior event is a key index of the value user according to the probability.
According to another aspect of the embodiments of the present invention, there is provided an advertisement delivery method, including:
determining the key indexes of the value users by using the method for determining the key indexes of the value users;
and transmitting the determined value user key index to an advertisement putting platform, so that the advertisement putting platform determines a value user according to the determined value user key index, and puts the advertisement of the client to the determined value user.
According to still another aspect of the embodiments of the present invention, there is provided a value user key index determining apparatus including:
the system comprises a user behavior event acquisition unit, a user behavior event acquisition unit and a user behavior event storage unit, wherein the user behavior event acquisition unit is used for receiving a user behavior event from a client and storing the user behavior event;
the user classification unit is used for determining the category to which the stored user behavior event belongs, wherein the category is a valuable user or a non-valuable user;
the regression analysis unit is used for carrying out regression analysis by using the user behavior event after the category to which the user behavior event belongs is determined, and obtaining the probability that the user of the client with the relevant information of the user behavior event is the value user;
and the key index determining unit is used for determining whether the relevant information of the user behavior event is a value user key index according to the probability.
According to yet another aspect of embodiments of the present invention, there is provided a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the method described in the above embodiments.
According to a further aspect of the embodiments of the present invention, there is provided a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method according to the above embodiments when executing the program.
The value user key index determining method, the advertisement putting method, the value user key index determining device, the computer readable storage medium and the computer equipment can automatically and directly obtain the truest data by all the client sides of a product to obtain the accurate attribution of the value user, so that the authenticity problem of a data source can be avoided, the difference of attribution conclusions of different platforms can be avoided, various problems faced by an analyst are reduced, the workload of analyzing data is reduced under the condition of improving the advertisement putting accuracy, and the error rate is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:
FIG. 1 is a schematic flow chart diagram illustrating a method for determining key indicators of value users according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating a method for delivering advertisements according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an advertisement delivery system according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a value user key indicator determining apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention.
First, the terms of art that may be referred to in this specification are explained:
BigQuery: a quick, economical and practical full-support-pipe type cloud data warehouse with strong capacity expansion capability and oriented to data analysis;
the mobile application program: refers to an application program designed for operation on a smartphone, tablet, or other mobile device;
marketing analysis platform: software for tracking and collecting mobile application program logs and providing better data analysis and decision support for marketers;
the following are attributed: attribution means how much effect of a certain result is caused by the common (or successive) effect of various factors, and a model established for solving the attribution problem is called an attribution model, namely attribution modeling;
SDK (Software Development Kit ): SDKs are generally a collection of development tools used by some software engineers to build application software for a particular software package, software framework, hardware platform, operating system, and the like;
interface: refers to a type of reference that defines an agreement; other types implement interfaces to ensure that they support certain operations;
putting: the method comprises the following steps that an advertisement is put on an advertisement platform, and a behavior of obtaining a user by paying is realized;
and (4) retention: the number of active days after the user installs the APP (application program) is a standard for measuring the income potential of the product;
paying: the behavior of the user purchasing in the APP.
Further, the concept of the present invention will be explained in detail:
fig. 1 is a flowchart illustrating a method for determining a key indicator of a value user according to an embodiment of the present invention. As shown in fig. 1, a flowchart of a value user key indicator determining method according to some embodiments may include:
step S110: receiving a user behavior event from a client, and storing the user behavior event;
step S120: determining a category to which the stored user behavior event belongs, wherein the category is a valuable user or a non-valuable user;
step S130: performing regression analysis by using the user behavior event after the category to which the user behavior event belongs is determined to obtain the probability that the user of the client with the relevant information of the user behavior event is the value user;
step S140: and determining whether the relevant information of the user behavior event is a key index of the value user according to the probability.
In the above step S110, the client may be an Application (APP) installed on a terminal device (e.g., a mobile phone, a tablet computer, etc.). User behavior events may be received from all clients.
The specific implementation of the step of receiving the user behavior event from the client may include: and receiving the user behavior event sent by the client through the SDK in real time. The "real-time" may refer to reporting when a user behavior event is generated by the client, and storing each received user behavior event.
The user behavior event mainly comprises a behavior event of the user of the client, and one user behavior event can comprise one behavior event. The behavioral events may include: the order of screen browsing, the buttons clicked, the time spent on a screen, whether to return the application to the background, the content of interest, etc. In addition, the user behavior event may further include a user ID (identity information), a user attribute, and the like, wherein the user attribute may include one or more related information of the user, such as gender, age, origin, occupation, and the like.
The specific implementation of the step of storing the user behavior event may include: and storing the user behavior event into a database. The received user behavior events may be sorted before being stored in the database, e.g., sorted by event occurrence time, separated by behavior events, separated by user attributes. The sorted user behavior events can be stored in a database in a certain manner. The user behavior events can be stored in a certain format by using a database, for example, each user behavior event is correspondingly formed into a piece of data containing user ID, user attributes and behavior event information, and different pieces of data are correspondingly formed by different user behavior events. The database may be a BigQuery database, or may be other databases, such as Redshift, hbase, etc. The BigQuery database can process mass data in a short time, so that the analysis speed of the user behavior event can be improved. In other embodiments, if the format of the received user behavior event is suitable for being stored in a certain format by simple processing in the database, the process of sorting the behavior event may not be needed.
In step S120, the value user may bring direct or potential value to the client, such as trading revenue, advertising revenue, etc. The value user may refer to a retention user, while the non-value user may refer to a non-retention user. A retention user is a user whose active duration is a set duration, e.g., 2-day retention, 7-day retention, 14-day retention, 28-day retention, etc., and if not a user of that active duration, may be considered a non-retention user. In this case, the user behavior event may include an installation time of the client of the user, a time when the user uses the client for the first time, a time when the user used the client for the last time, and the like. According to the time that the user uses the client for the first time and the time that the user uses the client for the last time, whether the user is a reserved user with a certain active duration can be judged. The remaining users may include all users who use the client for the first time in a certain date and remain active after the active duration is set, or may include all users who use the client for the first time in each date and have the active duration set to be active (for example, seven days).
In other embodiments, the value user may be a paying user, while the non-value user may be a non-paying user. The client may provide some service items for the user to purchase, and if the user purchases a certain service item within the set active time period, the user may be considered as a user who pays for the set active time period, for example, no payment for 7 days, no payment for 14 days, no payment for 28 days, etc. Therefore, the paid user may refer to a user who pays in the set active time period, and if the user does not pay in the set active time period, the user may be considered as a non-paid user, or a long-term non-paid user. In this case, the user behavior event may include a time when the user first uses the client, whether the user pays for the client, and the like. And calculating whether the user is a non-payment user with set active duration according to the time when the user uses the client for the first time and the reporting time of the user behavior event.
And reading the stored user behavior event, wherein the user behavior event can comprise user identity information, behavior event information, user attribute information and the like. Determining the category to which the stored user behavior event belongs, namely, dividing the user behavior event into two categories: value users and non-value users. After determining the category to which the stored user behavior event belongs, the resulting user behavior event may contain a tag identifying what category is, e.g., whether it is a value user or a non-value user. At this time, the user behavior event after determining the category to which the user behavior event belongs may include user identity information, behavior event information, user attribute information, and a category. Wherein, the user identity information can be the ID of the user; behavioral event information may include one or more of the order of screen browsing, buttons clicked, time spent on a screen, whether to return the application to the background, content of interest, etc.; the user attribute information may include one or more of gender, age, source, occupation, and the like.
In step S130, the information related to the user behavior event may be the behavior event information and/or the user attribute information, specifically, may be some behavior event information, such as sharing action, or may be some user attribute information, such as occupation or age; alternatively, there may be some sort of behavior event information, such as click content and dwell time, or there may be some sort of user attribute information, such as occupation and age, or a combination of user attribute information and behavior event information, such as age and click content.
One piece of information including the user identity information, the behavior event information, the user attribute information, and the category may be used as one sample, or the user identity information, the behavior event information, and the category may be used as one sample, or the user identity information, the user attribute information, and the category may be used as one sample. The user attribute information and/or behavioral event information in each sample may be referred to as a factor or as an indicator. The user identity information in each sample may be one type, and the behavior event information may be one type. Regression analysis, such as binary logistic regression, linear regression, and the like, may be performed on the obtained large number of samples to obtain the probability that the user corresponding to a certain user identity information or a combination of several user identity information is a valuable user, or obtain the probability that the user corresponding to a certain user attribute information or a combination of several user attribute information and a certain user identity information or a combination of several user identity information is a valuable user. The obtained probability can be used for evaluating the contribution size of the information to a user to become a valuable user, and may be a fraction or a ratio smaller than or equal to one and larger than or equal to zero, or the probability may be presented in the form of a weighted value, at this time, the sum of the probabilities corresponding to all the factors is one or other fixed values.
In some embodiments, the step S130 of performing regression analysis by using the user behavior event after determining the category to which the user behavior event belongs to obtain a probability that the user of the client having the relevant information of the user behavior event is the valuable user may include: and performing binary logistic regression analysis by using the user behavior event after the category to which the user behavior event belongs is determined, so as to obtain the probability that the user of the client with the relevant information of the user behavior event is the value user.
In step S140, the related information (factors) of each user behavior event may correspond to a respective probability. It is possible to determine which information or information (factors) to use by comparing the related information (different factors) of different user behavior events as a key indicator for evaluating whether the user of the client is a valuable user. For example, information corresponding to probabilities of first several names in size is used as a key index, or information corresponding to probabilities of being larger than a set value is used as a key index. The method and the device can learn what users are more likely to bring potential value through the obtained key indexes, so that the advertisements can be put into corresponding clients according to the key indexes, and the accuracy of advertisement positioning is improved.
In the embodiment of the invention, the user behavior events are obtained from the client, the user behavior events are classified, and regression analysis is carried out by utilizing the classified user behavior events, so that the most real data can be automatically and directly obtained by utilizing all the clients of a product to obtain the accurate attribution of a value user, the authenticity problem of a data source can be avoided, the difference of attribution conclusions of different platforms can be avoided, various problems faced by an analyst are reduced, the workload of analyzing data is reduced under the condition of improving the advertisement putting accuracy, and the error rate is reduced.
The embodiment of the invention also provides an advertisement putting method, and the advertisement putting method of some embodiments can comprise the following steps:
s201: determining a key index of the value user by using the method for determining the key index of the value user in each embodiment;
s202: and transmitting the determined value user key index to an advertisement putting platform, so that the advertisement putting platform determines a value user according to the determined value user key index, and puts the advertisement of the client to the determined value user.
In step S201, for example, information corresponding to probabilities of first several names may be used as a key index, or information corresponding to probabilities of being larger than a set value may be used as a key index.
In the step S202, the determined value user key indicator may be transmitted to an advertisement platform (which may be an existing advertisement platform) through an interface of the advertisement platform, and then the existing program of the advertisement platform may be used to find a corresponding program to deliver an advertisement to the client; or, the program code corresponding to the method for determining the value user key index according to the embodiments may be implanted into an advertisement platform, and after obtaining the value user key index, the corresponding program may be directly found to deliver the advertisement to the client. The advertisement platform can record a large amount of and various information of the users, so that the target users can be found conveniently.
When the input data during regression analysis is insufficient, whether the obtained key indexes are correct or not can be verified manually according to experience of the user. The user behavior events from the clients can be continuously received in real time, after the key indexes are calculated once, the subsequent user behavior events are increased, the key indexes can be periodically analyzed by utilizing the increased user behavior events, and if the key indexes are changed, a new key index value advertisement platform can be input to carry out more accurate advertisement putting.
In the embodiment, the value user key indexes determined by the method can be used for accurately putting advertisements for client products.
In order that those skilled in the art will better understand the present invention, embodiments of the present invention will be described below with reference to specific examples.
Fig. 2 is a flowchart illustrating a method for delivering an advertisement according to an embodiment of the present invention. Fig. 3 is a schematic structural diagram of an advertisement delivery system according to an embodiment of the present invention. Referring to fig. 2 and 3, a method for delivering advertisements according to an embodiment may include the steps of:
1. the mobile application client sends the behavior event of the user to a BigQuery database for storage through the SDK;
among them, it can be Redshift, hbase database. The data reported by the client can be sorted by the data sorting server and then stored in the database.
2. Dividing users into reserved users and non-reserved users, or paid users and long-term non-paid users;
the event is automatically sorted through a program, and the classification principle is that the calculation of the APP is opened for the next day of user installation and the APP is saved for the user. The payment and non-payment are defined by specific statistical events.
3. And (4) importing the classified output in the step (2) into an analysis system, and performing binary logistic regression through the existing user behaviors to find out the key behaviors triggered to be stored or paid by the user.
The analysis data may include the id and attributes (gender, age, source, province, etc.) of the user, behavior events, the sequence of browsing the screen, how long the button is clicked, on which screen the application stays, whether to return the application to the background, and what to pay attention to. The model can be built by inputting existing data, the probability of leaving by a user who outputs what behavior occurs is higher, and then the weight that the user will retain is output by inputting the behavior of the newly installed user on the day. The analysis system refers to a machine learning system, part of the functions of which may need training by inputting user behavior events that have been collected and behavior characteristics that are retained or paid for later. In addition to binary logistic regression, analysis can be performed by other models, such as linear regression. The model can be trained by performing logistic regression calculation on mass data, and the obtained machine learning classification model can be used for receiving relevant information of a new user, so that the contribution condition of the new user can be estimated.
4. And outputting the result to an analyst. The analyst may verify the results manually if he wishes to do so. If no manual analysis is needed, the delivery decision can be entered directly.
Wherein, the payment behavior event can be that the client uses the product on the 2 nd day, 7 th day, 14 th day, 28 th day and the like after installation.
5. And automatically or one-click issuing the decision to the releasing platform, so that the releasing platform can utilize the analysis decision to purposefully release the decision to a specific potential user which has the most value on the mobile program.
The advertisement delivery platform can have a targeted delivery function, for example, by analyzing that a 28-35 year old male and a financial practitioner would prefer a client product, the characteristics can be informed to the advertisement platform and delivered to such internet users by the advertisement platform.
In this embodiment, for the well-regulated data, the behavior characteristics of the retention user or the payment user are calculated by using binary logistic regression, and an analysis result favorable for retention is output for manual verification. And if the decision is confirmed, the decision is sent to the releasing platform. Through the data collection SDK and the data analysis platform, manual auxiliary analysis verification and application scheme summarization can be realized. The machine learning is utilized to analyze the collected data, the analyzed decision suggestions or the attribution model of the specific result is directly displayed to the analysts, and the marketers only need to return to verify the analysis result and directly feed back the analysis result to the putting platform, so that the putting platform is helped to make the setting putting on the best users suitable for the APP. The system can gather all data to be analyzed, autonomously analyze the key behaviors of the user, and directly issue the release only by manually checking the analysis result. The workload of the analysts and the input personnel is greatly reduced, and the communication problem of cross-functional cooperation is solved.
Based on the same inventive concept as the method for determining the key index of the value user shown in fig. 1, the embodiment of the present application further provides a device for determining the key index of the value user, as described in the following embodiments. Because the principle of solving the problem of the device for determining the key index of the value user is similar to that of the method for determining the key index of the value user, the implementation of the device for determining the key index of the value user can refer to the implementation of the method for determining the key index of the value user, and repeated parts are not repeated.
Fig. 4 is a schematic structural diagram of a value user key indicator determining apparatus according to an embodiment of the present invention. As shown in FIG. 4, the value user key indicator determining apparatus of some embodiments may include:
a user behavior event obtaining unit 210, configured to receive a user behavior event from a client, and store the user behavior event;
the user classifying unit 220 is configured to determine a category to which the stored user behavior event belongs, where the category is a valuable user or a non-valuable user;
a regression analysis unit 230, configured to perform regression analysis using the user behavior event after determining the category to which the user behavior event belongs, to obtain a probability that the user of the client having the relevant information of the user behavior event is the valuable user;
and a key index determining unit 240, configured to determine whether the information related to the user behavior event is a value user key index according to the probability.
Embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method described in the above embodiments.
The embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and when the processor executes the computer program, the steps of the method described in the above embodiment are implemented.
In some embodiments, the user behavior event obtaining unit may include: and the user behavior event acquisition module is used for receiving the user behavior event sent by the client through the SDK in real time.
In some embodiments, the user behavior event obtaining unit 210 may include: the user behavior event acquisition module is used for storing the user behavior event into a database; the database comprises a BigQuery database.
In some embodiments, the user behavior event after determining the category to which the user behavior event belongs includes user identity information, behavior event information, user attribute information, and category; and the related information of the user behavior event is the behavior event information or the user attribute information.
In some embodiments, the value user is a retention user and the non-value user is a non-retention user; or, the value user is a paid user, and the non-value user is a non-paid user.
In some embodiments, the regression analysis unit 230 may include: and the regression analysis module is used for performing binary logistic regression analysis by using the user behavior event after the category to which the user behavior event belongs is determined, so as to obtain the probability that the user of the client with the relevant information of the user behavior event is the value user.
An embodiment of the present invention further provides an advertisement delivery system, where the advertisement delivery system may include:
a value user key index determining device, configured to determine a value user key index by using the value user key index determining method according to each of the embodiments;
and the advertisement putting device is used for transmitting the determined value user key index to an advertisement putting platform so that the advertisement putting platform determines a value user according to the determined value user key index and puts the advertisement of the client to the determined value user.
In summary, according to the value user key index determining method, the advertisement delivery method, the value user key index determining device, the computer readable storage medium and the computer device of the embodiments of the present invention, the user behavior events are obtained from the clients, the user behavior events are classified, and regression analysis is performed using the classified user behavior events, so that accurate attribution of the value user can be automatically and directly obtained by using all clients of a product to obtain truest data, thereby avoiding the authenticity problem of data sources, avoiding the difference of attribution conclusions of different platforms, and reducing various problems faced by analysts, thereby reducing the workload of analyzing data and reducing the error rate under the condition of improving the advertisement delivery accuracy.
In the description herein, reference to the description of the terms "one embodiment," "a particular embodiment," "some embodiments," "for example," "an example," "a particular example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. The sequence of steps involved in the various embodiments is provided to schematically illustrate the practice of the invention, and the sequence of steps is not limited and can be suitably adjusted as desired.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (8)

1. A method for determining key indexes of value users is characterized by comprising the following steps:
receiving a user behavior event from a client, and storing the user behavior event;
determining a category to which the stored user behavior event belongs, wherein the category is a valuable user or a non-valuable user; the value user is a retention user and the non-value user is a non-retention user, or the value user is a payment user and the non-value user is a non-payment user;
performing regression analysis by using the user behavior event after the category to which the user behavior event belongs is determined to obtain the probability that the user of the client with the relevant information of the user behavior event is the value user; determining the user behavior event of the category to which the user behavior event belongs, wherein the user behavior event comprises user identity information, behavior event information, user attribute information and the category;
determining whether the relevant information of the user behavior event is a key index of a value user according to the probability; the related information of the user behavior event is the behavior event information or the user attribute information;
wherein:
receiving a user behavior event from a client and storing the user behavior event, wherein the user behavior event comprises:
receiving user behavior events from a client, sequencing the user behavior events according to event occurrence time, separating behavior events in the user behavior events, separating user attributes in the user behavior events, correspondingly forming data containing user identity information, user attribute information and behavior event information for each user behavior event, and storing the data into a database;
determining a category to which the stored user behavior event belongs, including:
judging whether the user is a reserved user with set active duration or not according to the time of using the client by the user for the first time in the user behavior event and the time of using the client by the user for the last time, or judging whether the user is a user paying for the set active duration or not according to the time of using the client by the user for the first time in the user behavior event, the reporting time of the user behavior event and the information of whether the user pays for the client so as to determine whether the category of the stored user behavior event belongs to a valuable user or not;
performing regression analysis by using the user behavior event after determining the category to which the user behavior event belongs to obtain the probability that the user of the client with the relevant information of the user behavior event is the value user, wherein the probability comprises the following steps:
and generating a sample according to the information in the user behavior event after the category to which the user behavior event belongs is determined, taking behavior event information and/or user attribute information in the sample as an index, and performing regression analysis by using the generated sample to obtain the probability that the user of the client with the behavior event information and/or the user attribute information in the user behavior event is a valuable user.
2. The value user key indicator determination method of claim 1, wherein receiving a user behavior event from a client comprises:
and receiving the user behavior event sent by the client through the SDK in real time.
3. The value user key indicator determination method of claim 1, wherein the database comprises a BigQuery database.
4. The method of claim 1, wherein performing regression analysis using the user behavior event after determining the category to which the user behavior event belongs to obtain a probability that the user of the client having the information related to the user behavior event is the value user comprises:
and performing binary logistic regression analysis by using the user behavior event after the category to which the user behavior event belongs is determined, so as to obtain the probability that the user of the client with the relevant information of the user behavior event is the value user.
5. An advertisement delivery method, comprising:
determining a value user key indicator using a value user key indicator determination method according to any one of claims 1 to 4;
and transmitting the determined value user key index to an advertisement putting platform, so that the advertisement putting platform determines a value user according to the determined value user key index, and puts the advertisement of the client to the determined value user.
6. A value user key indicator determination apparatus, comprising:
the system comprises a user behavior event acquisition unit, a user behavior event acquisition unit and a user behavior event storage unit, wherein the user behavior event acquisition unit is used for receiving a user behavior event from a client and storing the user behavior event;
the user classification unit is used for determining the category to which the stored user behavior event belongs, wherein the category is a valuable user or a non-valuable user; the value user is a retention user and the non-value user is a non-retention user, or the value user is a payment user and the non-value user is a non-payment user;
the regression analysis unit is used for carrying out regression analysis by using the user behavior event after the category to which the user behavior event belongs is determined, and obtaining the probability that the user of the client with the relevant information of the user behavior event is the value user; determining the user behavior event of the category to which the user behavior event belongs, wherein the user behavior event comprises user identity information, behavior event information, user attribute information and the category;
a key index determining unit, configured to determine whether the relevant information of the user behavior event is a key index of a valuable user according to the probability; the related information of the user behavior event is the behavior event information or the user attribute information;
wherein:
receiving a user behavior event from a client and storing the user behavior event, wherein the user behavior event comprises:
receiving user behavior events from a client, sequencing the user behavior events according to event occurrence time, separating behavior events in the user behavior events, separating user attributes in the user behavior events, correspondingly forming data containing user identity information, user attribute information and behavior event information for each user behavior event, and storing the data into a database;
determining a category to which the stored user behavior event belongs, including:
judging whether the user is a reserved user with set active duration or not according to the time of using the client by the user for the first time in the user behavior event and the time of using the client by the user for the last time, or judging whether the user is a user paying for the set active duration or not according to the time of using the client by the user for the first time in the user behavior event, the reporting time of the user behavior event and the information of whether the user pays for the client so as to determine whether the category of the stored user behavior event belongs to a valuable user or not;
performing regression analysis by using the user behavior event after determining the category to which the user behavior event belongs to obtain the probability that the user of the client with the relevant information of the user behavior event is the value user, wherein the probability comprises the following steps:
and generating a sample according to the information in the user behavior event after the category to which the user behavior event belongs is determined, taking behavior event information and/or user attribute information in the sample as an index, and performing regression analysis by using the generated sample to obtain the probability that the user of the client with the behavior event information and/or the user attribute information in the user behavior event is a valuable user.
7. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 5.
8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1 to 5 are implemented when the program is executed by the processor.
CN201910441219.8A 2019-05-24 2019-05-24 Value user key index determination method, advertisement delivery method and device Active CN110163683B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910441219.8A CN110163683B (en) 2019-05-24 2019-05-24 Value user key index determination method, advertisement delivery method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910441219.8A CN110163683B (en) 2019-05-24 2019-05-24 Value user key index determination method, advertisement delivery method and device

Publications (2)

Publication Number Publication Date
CN110163683A CN110163683A (en) 2019-08-23
CN110163683B true CN110163683B (en) 2020-04-14

Family

ID=67632770

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910441219.8A Active CN110163683B (en) 2019-05-24 2019-05-24 Value user key index determination method, advertisement delivery method and device

Country Status (1)

Country Link
CN (1) CN110163683B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111274907B (en) * 2020-01-16 2023-04-25 支付宝（中国）网络技术有限公司 Method and apparatus for determining category labels of users using category recognition model
CN112184046A (en) * 2020-10-12 2021-01-05 上海移卓网络科技有限公司 Advertisement service user value evaluation method, device, equipment and storage medium
CN112801693B (en) * 2021-01-18 2024-06-28 百果园技术(新加坡)有限公司 Advertisement feature analysis method and system based on high-value user
CN113918740A (en) * 2021-11-24 2022-01-11 北京达佳互联信息技术有限公司 Event information processing method and device, electronic equipment and storage medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105574147A (en) * 2015-12-15 2016-05-11 腾讯科技（深圳）有限公司 Information processing method and server
CN106203836A (en) * 2016-07-12 2016-12-07 中国石油化工股份有限公司 A kind of appraisal procedure of oil refining enterprise equipment dependability performance management
CN106682938A (en) * 2016-12-23 2017-05-17 广州帷策智能科技有限公司 Big-data precision marketing model establishing method and device
CN109558530A (en) * 2018-10-23 2019-04-02 深圳壹账通智能科技有限公司 User's portrait automatic generation method and system based on data processing

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080065395A1 (en) * 2006-08-25 2008-03-13 Ferguson Eric J Intelligent marketing system and method
CN106295382B (en) * 2015-05-20 2019-06-14 阿里巴巴集团控股有限公司 A kind of Information Risk preventing control method and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105574147A (en) * 2015-12-15 2016-05-11 腾讯科技（深圳）有限公司 Information processing method and server
CN106203836A (en) * 2016-07-12 2016-12-07 中国石油化工股份有限公司 A kind of appraisal procedure of oil refining enterprise equipment dependability performance management
CN106682938A (en) * 2016-12-23 2017-05-17 广州帷策智能科技有限公司 Big-data precision marketing model establishing method and device
CN109558530A (en) * 2018-10-23 2019-04-02 深圳壹账通智能科技有限公司 User's portrait automatic generation method and system based on data processing

Also Published As

Publication number Publication date
CN110163683A (en) 2019-08-23

Similar Documents

Publication Publication Date Title
CN110163683B (en) Value user key index determination method, advertisement delivery method and device
CN110209764A (en) The generation method and device of corpus labeling collection, electronic equipment, storage medium
US11270068B2 (en) System and interface for electronic communication generation and tracking
US20160285672A1 (en) Method and system for processing network media information
CN109325640A (en) User's Value Prediction Methods, device, storage medium and equipment
EP3764303A1 (en) Information processing device, etc. for calculating prediction data
CN106919625A (en) A kind of internet customer attribute recognition methods and device
CN110322093B (en) Information processing method, information display method, information processing device and computing equipment
CN110728422A (en) Building information model, method, device and settlement system for construction project
CN111882420A (en) Generation method of response rate, marketing method, model training method and device
CN110659961A (en) Method and device for identifying off-line commercial tenant
CN102402717A (en) Data analysis facility and method
CN107944905A (en) A kind of method and system of construction enterprises' material purchases price analysis
CN113326434B (en) Information recommendation method, equipment and storage medium
CN109727125A (en) Borrowing balance prediction technique, device, server, storage medium
WO2021129368A1 (en) Method and apparatus for determining client type
CN113837855A (en) Cloud network-based cross-border value-added tax batch intelligent processing method, device and equipment
CN111325572B (en) Data processing method and device
CN113822749A (en) Merchant settlement and payment processing method, device, equipment and medium based on RPA and AI
CN117094764A (en) Bank integral processing method and device
CN108460049A (en) A kind of method and system of determining information category
CN111415138A (en) Creative processing method and system, client and server
CN115511536A (en) Method and device for analyzing life-long value of customer
CN113469558A (en) Material management method and system
CN111241799A (en) Iterative calculation method for massive formulas of online server

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant