CN111681035A - Liveness estimation method and device for active advertisements - Google Patents

Liveness estimation method and device for active advertisements Download PDF

Info

Publication number
CN111681035A
CN111681035A CN202010361126.7A CN202010361126A CN111681035A CN 111681035 A CN111681035 A CN 111681035A CN 202010361126 A CN202010361126 A CN 202010361126A CN 111681035 A CN111681035 A CN 111681035A
Authority
CN
China
Prior art keywords
active
advertisement
liveness
advertisements
delivery
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.)
Granted
Application number
CN202010361126.7A
Other languages
Chinese (zh)
Other versions
CN111681035B (en
Inventor
彭欣宇
李百川
田家赫
陈第
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.)
Youmi Technology Co ltd
Original Assignee
Youmi 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 Youmi Technology Co ltd filed Critical Youmi Technology Co ltd
Priority to CN202010361126.7A priority Critical patent/CN111681035B/en
Publication of CN111681035A publication Critical patent/CN111681035A/en
Application granted granted Critical
Publication of CN111681035B publication Critical patent/CN111681035B/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
    • G06Q30/0241—Advertisements
    • G06Q30/0242—Determining effectiveness of advertisements
    • G06Q30/0245—Surveys
    • 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
    • G06Q30/0272—Period of advertisement exposure

Landscapes

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

Abstract

The invention discloses a method and a device for estimating liveness of a live advertisement, which comprise the following steps: after detecting the liveness calculation instruction, determining all active advertisements needing to calculate the liveness; acquiring liveness calculation models matched with all active advertisements and calculating the liveness of each active advertisement according to the liveness calculation models; and determining all active advertisements needing to calculate the liveness specifically to filter the advertisements delivered within the target time from all the counted advertisements or to select the advertisements delivered within the target time and delivered within a preset time period before the target time as the active advertisements. Therefore, the method and the device can provide a determination mode of the active advertisement, and after the active advertisement is determined, the activity of the active advertisement is individually estimated according to the activity calculation model matched with the active advertisement, so that the method and the device are favorable for more accurately analyzing the advertisement putting effect of the active advertisement and providing more accurate reference basis for adjusting the advertisement putting strategy.

Description

Liveness estimation method and device for active advertisements
Technical Field
The invention relates to the technical field of internet, in particular to a liveness estimation method and device for a liveness advertisement.
Background
With the rapid development of the internet, more and more internet users are provided. In order to expand the audience range of advertisements and improve the influence of advertisements, the advertisements are gradually developed towards internet advertisements from traditional paper advertisements, printed wall advertisements, television advertisements and the like.
In practical application, advertisements put in every day are various, for an advertiser, different advertisements can be put in an advertisement putting platform aiming at different commodities, different advertisements can be put in the advertisement putting platform aiming at the same commodity, and for the advertiser, the liveness of the advertisements is an important basis for the advertiser to know the advertisement putting effect or change the advertisement putting strategy. Therefore, how to realize the estimation of the advertisement liveness is very important.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a method and a device for pre-estimating the liveness of an advertisement, which can pre-estimate the liveness of the active advertisement.
In order to solve the above technical problem, a first aspect of the embodiments of the present invention discloses a liveness prediction method for an active advertisement, where the method includes:
after detecting liveness calculation instructions for the active advertisements, determining all active advertisements for which liveness needs to be calculated;
acquiring liveness calculation models matched with all the active advertisements, and calculating the liveness of each active advertisement according to the liveness calculation models;
wherein the determining all active advertisements for which liveness needs to be calculated comprises:
determining a target time indicated by the liveness calculation instruction;
screening advertisements delivered within the target time from all the counted advertisements to serve as active advertisements; or screening the advertisements delivered within the target time and delivered within a preset time period before the target time from all the counted advertisements as active advertisements.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, before the calculating the liveness of each of the active advertisements according to the liveness calculation model, the method further includes:
determining the delivery attribute of each active advertisement, wherein the delivery attribute of the active advertisement is one of a continuous delivery attribute and a discontinuous delivery attribute;
calculating the delivery parameter change condition of each active advertisement according to the delivery attribute of each active advertisement, wherein the delivery parameter change condition of the active advertisement comprises the delivery frequency change rate and the interaction data change rate of the active advertisement;
wherein said calculating liveness of each of said active advertisements according to said liveness calculation model comprises:
and calculating the liveness of each active advertisement according to the delivery parameter change condition of each active advertisement and a liveness calculation model matched with the active advertisement.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the calculating, according to the delivery attribute of each active advertisement, a delivery parameter variation of each active advertisement includes:
when the delivery attribute of the active advertisement is the continuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement active characteristic of the active advertisement in the target time and the historical advertisement active characteristic of the active advertisement before the target time;
when the delivery attribute of the active advertisement is the discontinuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement activity characteristic of the active advertisement in the target time and the historical average advertisement activity characteristic of all advertisements before the target time.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the target advertisement activity characteristics of the active advertisement in the target time include the number of times of advertisement impressions of the active advertisement in the target time and advertisement interaction data;
and the advertisement interaction data of the active advertisement in the target time comprises one or more combinations of the number of praise, the number of forwarding, the number of comments, the number of browsing, the number of collection, the number of barrage and the amount of appreciation.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the calculating the liveness of each active advertisement according to the delivery parameter variation of each active advertisement and a liveness calculation model matched with the active advertisement includes:
determining a first weight value corresponding to the change rate of the releasing times and a second weight value corresponding to the change rate of the interaction data;
and substituting the first weight value, the second weight value, the delivery times change rate of each active advertisement and the interaction data change rate of each active advertisement into an activity calculation formula matched with the active advertisements to obtain the activity of each active advertisement.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the activity calculation formula is:
S＝W1*X1+W2*X2(ii) a Alternatively, the first and second electrodes may be,
S＝W3*(W1*X1+W2*X2)；
wherein S is the liveness of any of the active advertisements, W1Is the first weight value, W2Is the second weight value, X1The rate of change of the number of impressions, X, for the active advertisement2The interaction data change rate, W, for the active advertisement3The determined expansion factor corresponding to the active advertisement.
As an alternative implementation, in the first aspect of the embodiments of the present invention, X1And X2The calculation formulas of (A) and (B) are respectively as follows:
X1＝tanh((Td-Ty)/Ty)；
X2＝∑iwi*tanh((Rdi-Ryi)/Ryi)；
wherein, TdWhen the delivery attribute of the active advertisement is the continuous delivery attribute, T is the advertisement delivery times of the active advertisement in the target timeyWhen the delivery attribute of the active advertisement is the discontinuous delivery attribute, T is the advertisement delivery times of the active advertisement in a preset time period before the target timeyAveraging the historical advertisement delivery times of each advertisement in all advertisements;
i for representing data class, wiIs the weight value of the data class, RdiThe data volume of the data category in the advertisement interaction data of the active advertisement in the target time; when the delivery attribute of the active advertisement is the continuous delivery attribute, RyiWhen the delivery attribute of the active advertisement is the non-continuous delivery attribute, R is the data volume of the data category in the advertisement interaction data of the active advertisement in the preset time period before the target timeyiThe historical data amount of the data category in the historical advertisement interaction data of all advertisements is averaged.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the expansion factor corresponding to each active advertisement is determined according to all delivery areas of the active advertisement within the target time, and when all delivery areas of the active advertisement within the target time are not unique, the expansion factor corresponding to the active advertisement is specifically determined according to a region weight value corresponding to each delivery area of all delivery areas of the active advertisement within the target time.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, after the calculating the liveness of each of the active advertisements according to the liveness calculation model, the method further includes:
and acquiring the historical liveness of each active advertisement, and correcting the calculated liveness of the active advertisement according to the historical liveness of each active advertisement to obtain the corrected liveness.
The second aspect of the embodiment of the present invention discloses an activity prediction apparatus for a live advertisement, the apparatus comprising:
the determining module is used for determining all active advertisements needing to calculate the liveness after detecting the liveness calculating instruction aiming at the active advertisements;
the obtaining module is used for obtaining an activity degree calculation model matched with all the active advertisements;
the calculating module is used for calculating the liveness of each active advertisement according to the liveness calculating model;
the specific way of determining all active advertisements needing to calculate liveness by the determining module is as follows:
determining a target time indicated by the liveness calculation instruction;
screening advertisements delivered within the target time from all the counted advertisements to serve as active advertisements; or screening the advertisements delivered within the target time and delivered within a preset time period before the target time from all the counted advertisements as active advertisements.
As an optional implementation manner, in the second aspect of the embodiment of the present invention, the determining module is further configured to determine a placement attribute of each active advertisement before the calculating module calculates the liveness of each active advertisement according to the liveness calculation model, where the placement attribute of each active advertisement is one of a continuous placement attribute and a discontinuous placement attribute;
the calculation module is further configured to calculate a delivery parameter change condition of each active advertisement according to a delivery attribute of each active advertisement, where the delivery parameter change condition of each active advertisement includes a delivery frequency change rate and an interaction data change rate of each active advertisement;
the specific way for the calculation module to calculate the liveness of each active advertisement according to the liveness calculation model is as follows:
and calculating the liveness of each active advertisement according to the delivery parameter change condition of each active advertisement and a liveness calculation model matched with the active advertisement.
As an optional implementation manner, in the second aspect of the embodiment of the present invention, the specific manner of calculating the liveness of each active advertisement by the calculation module according to the delivery parameter variation of each active advertisement and the liveness calculation model matched with the active advertisement is as follows:
when the delivery attribute of the active advertisement is the continuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement active characteristic of the active advertisement in the target time and the historical advertisement active characteristic of the active advertisement before the target time;
when the delivery attribute of the active advertisement is the discontinuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement activity characteristic of the active advertisement in the target time and the historical average advertisement activity characteristic of all advertisements before the target time.
As an optional implementation manner, in the second aspect of the embodiment of the present invention, the target advertisement activity characteristics of the active advertisement in the target time include the number of times of advertisement impressions of the active advertisement in the target time and advertisement interaction data;
and the advertisement interaction data of the active advertisement in the target time comprises one or more combinations of the number of praise, the number of forwarding, the number of comments, the number of browsing, the number of collection, the number of barrage and the amount of appreciation.
As an optional implementation manner, in the second aspect of the embodiment of the present invention, the specific manner of calculating the liveness of each active advertisement by the calculation module according to the delivery parameter variation of each active advertisement and the liveness calculation model matched with the active advertisement is as follows:
determining a first weight value corresponding to the change rate of the releasing times and a second weight value corresponding to the change rate of the interaction data;
and substituting the first weight value, the second weight value, the delivery times change rate of each active advertisement and the interaction data change rate of each active advertisement into an activity calculation formula matched with the active advertisements to obtain the activity of each active advertisement.
As an optional implementation manner, in the second aspect of the embodiment of the present invention, the activity calculation formula is:
S＝W1*X1+W2*X2(ii) a Alternatively, the first and second electrodes may be,
S＝W3*(W1*X1+W2*X2)；
wherein S is the liveness of any of the active advertisements, W1Is the first weight value, W2Is the second weight value, X1The rate of change of the number of impressions, X, for the active advertisement2The interaction data change rate, W, for the active advertisement3The determined expansion factor corresponding to the active advertisement.
As an alternative implementation, in a second aspect of an embodiment of the invention, X1And X2The calculation formulas of (A) and (B) are respectively as follows:
X1＝tanh((Td-Ty)/Ty)；
X2＝∑iwi*tanh((Rdi-Ryi)/Ryi)；
wherein, TdWhen the delivery attribute of the active advertisement is the continuous delivery attribute, T is the advertisement delivery times of the active advertisement in the target timeyWhen the delivery attribute of the active advertisement is the discontinuous delivery attribute, T is the advertisement delivery times of the active advertisement in a preset time period before the target timeyAveraging the historical advertisement delivery times of each advertisement in all advertisements;
i for representing data class, wiIs the weight value of the data class, RdiThe data volume of the data category in the advertisement interaction data of the active advertisement in the target time; when the delivery attribute of the active advertisement is the continuous delivery attribute, RyiWhen the delivery attribute of the active advertisement is the non-continuous delivery attribute, R is the data volume of the data category in the advertisement interaction data of the active advertisement in the preset time period before the target timeyiThe historical data amount of the data category in the historical advertisement interaction data of all advertisements is averaged.
As an optional implementation manner, in the second aspect of the embodiment of the present invention, the expansion factor corresponding to each active advertisement is determined according to all delivery areas of the active advertisement within the target time, and when all delivery areas of the active advertisement within the target time are not unique, the expansion factor corresponding to the active advertisement is specifically determined according to a region weight value corresponding to each delivery area of all delivery areas of the active advertisement within the target time.
As an optional implementation manner, in the second aspect of the embodiment of the present invention, the obtaining module is further configured to obtain the historical liveness of each active advertisement after the calculating module calculates the liveness of each active advertisement according to the liveness calculation model;
and, the apparatus further comprises:
and the correction module is used for correcting the calculated liveness of the active advertisement according to the historical liveness of each active advertisement to obtain the corrected liveness.
The third aspect of the embodiments of the present invention discloses another liveness estimation device for an active advertisement, the device including:
a memory storing executable program code;
a processor coupled with the memory;
the processor calls the executable program code stored in the memory to execute part or all of the steps of the liveness estimation method for the active advertisements disclosed in the first aspect of the embodiment of the invention.
A fourth aspect of the present invention discloses a computer storage medium, where the computer storage medium stores computer instructions, and the computer instructions, when called, are used to execute part or all of the steps in the liveness estimation method for active advertisements disclosed in the first aspect of the present invention.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
in the embodiment of the invention, after the liveness calculation instruction is detected, all active advertisements needing to calculate the liveness are determined; acquiring liveness calculation models matched with all active advertisements and calculating the liveness of each active advertisement according to the liveness calculation models; and determining all active advertisements needing to calculate the liveness specifically to filter the advertisements delivered within the target time from all the counted advertisements or to select the advertisements delivered within the target time and delivered within a preset time period before the target time as the active advertisements. Therefore, the method and the device can provide a determination mode of the active advertisement, and after the active advertisement is determined, the activity of the active advertisement is individually estimated according to the activity calculation model matched with the active advertisement, so that the method and the device are favorable for more accurately analyzing the advertisement putting effect of the active advertisement and providing more accurate reference basis for adjusting the advertisement putting strategy.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flowchart illustrating a method for predicting liveness of a live advertisement according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating another method for predicting liveness of an active advertisement according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of an apparatus for estimating liveness of a live advertisement according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of another liveness estimation device for active advertisements according to an embodiment of the present invention
Fig. 5 is a schematic structural diagram of an activity prediction apparatus for another active advertisement according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first," "second," and the like in the description and claims of the present invention and in the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, article, or article that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or article.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
The invention discloses a liveness estimation method and a liveness estimation device for an active advertisement, which can provide a determination mode of the active advertisement, and perform personalized estimation on the liveness of the active advertisement according to a liveness calculation model matched with the active advertisement after the active advertisement is determined, so that the method and the device are favorable for more accurately analyzing the advertisement putting effect of the active advertisement and providing more accurate reference basis for adjusting the advertisement putting strategy. The following are detailed below.
Example one
Referring to fig. 1, fig. 1 is a schematic flowchart illustrating a method for estimating liveness of a live advertisement according to an embodiment of the present invention. The method described in fig. 1 is applied to an advertisement liveness estimation device (also referred to as an estimation implementation device), such as a background server, and the embodiment of the present invention is not limited thereto. As shown in FIG. 1, the liveness estimation method for active advertisements may include the following operations:
101. after detecting liveness calculation instructions for active advertisements, the predictive device determines all active advertisements for which liveness needs to be calculated.
In an embodiment of the invention, the liveness calculation instruction is used for indicating the liveness of the active advertisement. Optionally, the liveness calculation instructions may include a specified target time, and at this time, the liveness calculation instructions are specifically used for indicating the liveness of the active advertisement within the specified target time. Further optionally, the liveness calculation instructions may further include a reference identifier for determining the active advertisement for which liveness calculation is required, where the reference identifier may include at least one of a target commodity category for determining the active advertisement delivered by the corresponding advertiser, an advertiser identifier for determining the active advertisement for which the advertised commodity belongs, and an advertisement delivery time range for determining the active advertisement delivered within the advertisement delivery time range. Therefore, the liveness calculation aiming at the required active advertisements can be realized by triggering the liveness calculation instruction in the embodiment of the invention.
In this embodiment of the present invention, the target time specified by the advertisement liveness calculation instruction may be a certain day (e.g., today, yesterday, etc.), a certain period of time in units of days (e.g., a certain week, a certain half month, a certain month, or a certain quarter), or a certain period of time in units of minutes or hours (e.g., No. 03 month, No. 22 am, 10:00-11:00), which is not limited in this embodiment of the present invention.
102. The pre-estimation device obtains an activity degree calculation model matched with all active advertisements.
In the embodiment of the present invention, optionally, all active advertisements may match the same liveness calculation model, or may match different liveness calculation models. Specifically, when different liveness calculation models are matched, the pre-estimation device obtains the liveness calculation model matched with all the active advertisements, and the method may include:
the pre-estimation device divides all determined active advertisements into at least one active advertisement group according to a predetermined classification rule, the active advertisements of the same active advertisement group are matched with the same liveness calculation model, and the active advertisements of different active advertisement groups are matched with different liveness calculation models;
the pre-estimation device obtains an activity degree calculation model matched with each active advertisement group.
Therefore, the active advertisements are grouped and matched with different liveness calculation models in the embodiment of the invention, which is beneficial to realizing the personalized estimation of the liveness of different types of active advertisements. For example, the estimation device may divide all the active advertisements into active advertisement groups with different activity levels according to a preset activity level, or, for example, the estimation device may divide all the active advertisements into active advertisement groups with different activity levels according to a preset delivery frequency range and a delivery frequency level matched with the preset delivery frequency range, where the active advertisement groups with different delivery frequency levels match different activity calculation models.
103. The pre-estimating device calculates the liveness of each active advertisement according to the liveness calculation model.
In an alternative embodiment, the estimation means determines all active advertisements for which liveness needs to be calculated, and may include:
the pre-estimation device determines target time indicated by the liveness calculation instruction;
screening advertisements delivered in the target time from all the counted advertisements to serve as active advertisements; alternatively, the first and second electrodes may be,
and screening the advertisements delivered within the target time and delivered within a preset time period before the target time from all the counted advertisements to serve as active advertisements.
The advertisement delivered within the target time and delivered within the preset time period before the target time is specifically an advertisement delivered within the target time and delivered within the preset time period before the target time.
Further optionally, if the liveness calculation instruction indicates the reference identifier in addition to the target time, the estimation device further screens, according to the reference identifier, the active advertisements matched with the reference identifier from the preliminarily screened active advertisements as the active advertisements needing to calculate the liveness finally after preliminarily screening the active advertisements according to the target time.
In another optional embodiment, after calculating the advertisement liveness for all active advertisements, the method may further comprise the operations of:
the pre-estimating device classifies all advertisements according to the advertiser identification of each advertisement to obtain at least one advertisement group, the advertiser identifications of the advertisements in the same advertisement group are the same, and the advertiser identifications of the advertisements in different advertisement groups are different;
the pre-estimation device arranges the activity information of all advertisements included in each advertisement group according to the predetermined content formats corresponding to different advertiser identifications;
the pre-estimation device feeds back the activeness information of all the advertisements included in each advertisement group to the advertisers matched with the advertiser identifications of all the advertisements in the advertisement group.
Wherein the liveness information of all the advertisements included in each advertisement group at least comprises the calculated liveness of each advertisement included in the advertisement group at the target time.
Therefore, after calculating the liveness of the advertisement, the optional embodiment arranges the liveness information of the advertisement delivered by the advertiser according to the advertiser identifier and the corresponding content format and feeds the information back to the corresponding advertiser in time, so that the advertiser can know the liveness of the advertisement delivered by the advertiser in time.
Therefore, the method described by the embodiment of the invention can provide a determination mode of the active advertisement, and after the active advertisement is determined, the activity of the active advertisement is individually estimated according to the activity calculation model matched with the active advertisement, so that the method is beneficial to more accurately analyzing the advertisement putting effect of the active advertisement and providing more accurate reference basis for adjusting the advertisement putting strategy.
Example two
Referring to fig. 2, fig. 2 is a flowchart illustrating another method for estimating liveness of an active advertisement according to an embodiment of the present invention. The method described in fig. 2 is applied to an advertisement liveness estimation device (also referred to as an estimation implementation device), such as a background server, and the embodiment of the present invention is not limited thereto. As shown in FIG. 2, the liveness estimation method for active advertisements may include the following operations:
201. after detecting liveness calculation instructions for active advertisements, the predictive device determines all active advertisements for which liveness needs to be calculated.
For the related description of step 201, refer to the corresponding description of step 101 in the first embodiment, which is not repeated herein.
202. The pre-estimation device determines the delivery attribute of each active advertisement, wherein the delivery attribute of each active advertisement is one of a continuous delivery attribute and a discontinuous delivery attribute.
In the embodiment of the present invention, when the delivery attribute of the active advertisement is the continuous delivery attribute, step 203 is triggered and executed; when the delivery attribute of the active advertisement is the non-continuous delivery attribute, the execution of step 204 is triggered.
Optionally, the predicting means determines the delivery attribute of each active advertisement, and may include:
the pre-estimation device collects the delivery condition of each active advertisement before the target time, and determines the delivery attribute of each active advertisement according to the delivery condition of each active advertisement before the target time.
Specifically, when the active advertisement is an advertisement that is screened by the estimation device from all the counted advertisements and is delivered within a target time, if the active advertisement is delivered within a certain time period before the target time, the estimation device may determine that the delivery attribute of the active advertisement is a continuous delivery attribute, and if the active advertisement is not delivered within a certain time period before the target time, the estimation device may determine that the delivery attribute of the active advertisement is a discontinuous delivery attribute.
203. The estimation device calculates the delivery parameter change condition of the active advertisement based on the target advertisement activity characteristic of the active advertisement in the target time and the historical advertisement activity characteristic of the active advertisement before the target time.
204. The estimation device calculates the delivery parameter change condition of the active advertisement based on the target advertisement activity characteristic of the active advertisement in the target time and the historical average advertisement activity characteristic of all advertisements before the target time.
In the embodiment of the invention, the delivery parameter change condition of the active advertisement comprises the delivery frequency change rate of the active advertisement and the interaction data change rate. Optionally, the target advertisement liveness characteristics of the active advertisement in the target time include advertisement putting times of the active advertisement in the target time and advertisement interaction data, where the advertisement interaction data includes data with user interaction attributes, and the advertisement putting times of the active advertisement in the target time may be a sum of the advertisement putting times of the active advertisement in different putting areas. Further, the targeted advertisement activity characteristics of the active advertisement at the targeted time may also include an advertisement placement area of the active advertisement at the targeted time.
Still further optionally, the advertisement interaction data of the active advertisement in the target time includes one or more combinations of the number of praise, the number of forwarding, the number of comments, the number of browsing, the number of collection, the number of barrage, and the amount of appreciation.
Therefore, the method and the device can calculate the delivery parameter change conditions of the active advertisements with different delivery attributes according to different data, realize the personalized calculation of the delivery parameter change conditions, and are beneficial to improving the accuracy of the delivery parameter change conditions of the active advertisements with different delivery attributes.
205. The pre-estimation device obtains an activity calculation formula matched with all active advertisements.
It should be noted that, no precedence order of execution exists between any one of step 202, step 203, and step 204 and step 205, and step 205 may be executed before step 202, after step 202, or simultaneously with step 202, which is not limited in the embodiment of the present invention.
206. The pre-estimation device determines a first weight value corresponding to the change rate of the delivery times and a second weight value corresponding to the change rate of the interactive data.
In the embodiment of the present invention, it should be noted that the change rates of the delivery times of all active advertisements may correspond to the same first weight value, or the change rates of the delivery times within the same delivery time change rate range correspond to the same first weight value, or the change rate of the delivery times of each active advertisement has the corresponding first weight value, which is not limited in the embodiment of the present invention. Similarly, the detailed description of the second weight value corresponding to the change rate of the interactive data may refer to the detailed description of the first weight value corresponding to the change rate of the delivery times, and is not repeated.
207. And the pre-estimation device substitutes the first weight value, the second weight value, the change rate of the delivery times of each active advertisement and the change rate of the interaction data of each active advertisement into an activity calculation formula matched with the active advertisement to obtain the activity of each active advertisement.
Optionally, the activity calculation formula is:
S＝W1*X1+W2*X2(ii) a Alternatively, the first and second electrodes may be,
S＝W3*(W1*X1+W2*X2)；
wherein S is the liveness of any active advertisement, W1A first weight value, W, corresponding to the variation rate of the impression times of the active advertisement2A second weight value, X, corresponding to the interactive data change rate of the active advertisement1The rate of change of the number of impressions, X, for the active advertisement2Interaction data change rate, W, for the active advertisement3The determined expansion factor corresponding to the active advertisement.
Further optionally, the expansion factor corresponding to the active advertisement is determined according to all the delivery areas of the active advertisement within the target time, and when all the delivery areas of the active advertisement within the target time are not unique, the expansion factor corresponding to the active advertisement is specifically determined according to a region weight value corresponding to each delivery area of all the delivery areas of the active advertisement within the target time. Namely: each region has a region weight value corresponding to the region weight value, and when all the delivery regions of the active advertisement in the target time are not unique, the expansion multiple corresponding to the active advertisement may be specifically equal to the sum of the region weight values corresponding to each delivery region of all the delivery regions of the active advertisement in the target time.
Therefore, in addition to the advertisement interaction data, the method and the device for evaluating the liveness of the active advertisements can comprehensively evaluate the liveness of the active advertisements from the dimensions of the launching areas, and are favorable for realizing the liveness estimation of the advertisements facing different launching areas (such as facing the international market).
Alternatively, the delivery area may be a county-level area, a city-level area, a provincial-level area, or a country-level area.
Further optionally, X1And X2The calculation formulas of (A) and (B) are respectively as follows:
X1＝tanh((Td-Ty)/Ty)；
X2＝∑iwi*tanh((Rdi-Ryi)/Ryi)；
wherein, TdThe number of ad impressions for the active ad in the target time,when the delivery attribute of the active advertisement is a continuous delivery attribute, TyWhen the delivery attribute of the active advertisement is the discontinuous delivery attribute, T is the advertisement delivery times of the active advertisement in a preset time period before the target timeyAveraging the historical advertisement delivery times of each advertisement in all advertisements; i for representing data class, wiIs the weight value of the data class, RdiData volume of data of the data category in advertisement interaction data for the active advertisement within a target time; when the delivery attribute of the active advertisement is a continuous delivery attribute, RyiWhen the delivery attribute of the active advertisement is the non-continuous delivery attribute, R is the data volume of the data category in the advertisement interaction data of the active advertisement in the preset time period before the target timeyiThe historical data amount of the data category in the historical advertisement interaction data of all advertisements is averaged. And the number of values of i is matched with the number of data categories corresponding to the advertisement interaction data in the target time, and one value corresponds to one data category. For example, when the advertisement interaction data of the active advertisement in the target time includes the praise number, the comment number and the forwarding number, the number of the data categories corresponding to the advertisement interaction data in the target time is 3, the value of i is 3, one of the values corresponds to the praise category, one of the remaining two values corresponds to the comment category, and the other value corresponds to the forwarding category.
In an alternative embodiment, after performing step 207, the method may further include the following operations:
208. the pre-estimation device obtains the historical liveness of each active advertisement, and corrects the calculated liveness of the active advertisement according to the historical liveness of each active advertisement to obtain the corrected liveness.
In this optional embodiment, when the active advertisement is a new active advertisement (i.e., an advertisement delivered for the first time within the target time), the historical liveness of the active advertisement is a default initial liveness, when the active advertisement is an old active advertisement (i.e., an advertisement delivered within the target time and within a preset time period before the target time), the historical liveness of the active advertisement is a modified liveness previously calculated for the active advertisement according to a liveness calculation model corresponding to the active advertisement and a modification method of the liveness, and when the active advertisement is an inactive advertisement, the historical liveness of the active advertisement is a liveness previously calculated by using a liveness calculation model corresponding to the inactive advertisement.
Optionally, the activity calculation model corresponding to the inactive advertisement is an attenuation model, and a calculation formula of the attenuation model may be:
S＝Sa*0.5(D/L)(ii) a Alternatively, the first and second electrodes may be,
S＝I*Sa*0.5(D/L)；
wherein S is the liveness of the inactive advertisement, SaThe method comprises the steps that the activity degree of the advertisement is calculated in historical time, L is a predetermined half-life period, D is interval duration between the historical time and a target time appointed when the activity degree of the non-active advertisement is calculated, the historical time is the time when the fact that the non-active advertisement is an active advertisement when the activity degree needs to be calculated is detected before the target time and the interval duration from the target time is the shortest, and I is an activity degree correction coefficient corresponding to the advertisement.
Therefore, the method described by the embodiment of the invention can provide a determination mode of the active advertisement, and after the active advertisement is determined, the activity of the active advertisement is individually estimated according to the activity calculation model matched with the active advertisement, so that the method is beneficial to more accurately analyzing the advertisement putting effect of the active advertisement and providing more accurate reference basis for adjusting the advertisement putting strategy. In addition, the method can calculate the delivery parameter change conditions of the active advertisements with different delivery attributes according to different data, realize the personalized calculation of the delivery parameter change conditions, and is favorable for improving the accuracy of the delivery parameter change conditions of the active advertisements with different delivery attributes. In addition, the initial liveness can be corrected according to the historical liveness after the initial liveness is calculated, and the corrected liveness is used as the finally estimated liveness, so that the liveness of the active advertisement in the appointed time can be corrected through the historical liveness, the condition that the estimated liveness is deviated due to sudden change of the advertisement liveness characteristics of the active advertisement in the appointed time (such as sudden increase of advertisement putting times and sudden increase of advertisement interaction data) is favorably reduced, and the accuracy of the finally estimated liveness is favorably further improved. In addition, besides the advertisement interaction data, the liveness of the active advertisements can be comprehensively evaluated from the dimensions of the delivery areas, and the liveness estimation of the advertisements in different delivery areas is facilitated.
EXAMPLE III
Referring to fig. 3, fig. 3 is a schematic structural diagram of an activity degree estimation apparatus for an activity advertisement according to an embodiment of the present invention. As shown in fig. 3, the apparatus may include:
a determining module 301, configured to determine all active advertisements for which liveness needs to be calculated after detecting liveness calculation instructions for the active advertisements.
An obtaining module 302, configured to obtain an activity calculation model matching all active advertisements.
A calculating module 303, configured to calculate the liveness of each active advertisement according to the liveness calculation model.
Optionally, the specific way for determining all active advertisements that need to calculate liveness by the determining module 301 is as follows:
determining a target time indicated by the liveness calculation instruction;
screening advertisements delivered in the target time from all the counted advertisements to serve as active advertisements; or screening the advertisements delivered within the target time and delivered within a preset time period before the target time from all the counted advertisements as active advertisements.
Therefore, the device described by implementing the figure 3 can determine the active advertisement, and after the active advertisement is determined, the activity of the active advertisement is individually estimated according to the activity calculation model matched with the active advertisement, so that the advertisement delivery effect of the active advertisement can be more accurately analyzed, and a more accurate reference basis is provided for adjusting the advertisement delivery strategy.
In an optional embodiment, the determining module 302 is further configured to determine the placement attribute of each active advertisement before the calculating module 303 calculates the liveness of each active advertisement according to the liveness calculation model, where the placement attribute of each active advertisement is one of a continuous placement attribute and a discontinuous placement attribute.
The calculating module 303 is further configured to calculate a delivery parameter change condition of each active advertisement according to the delivery attribute of each active advertisement, where the delivery parameter change condition of each active advertisement includes a delivery frequency change rate of the active advertisement and an interaction data change rate.
The specific way for the calculating module 303 to calculate the liveness of each active advertisement according to the liveness calculation model is as follows:
and calculating the liveness of each active advertisement according to the delivery parameter change condition of each active advertisement and the liveness calculation model matched with the active advertisement.
In another alternative embodiment, the calculating module 303 calculates the liveness of each active advertisement according to the delivery parameter variation of each active advertisement and the liveness calculation model matched with the active advertisement, in a specific manner:
when the delivery attribute of the active advertisement is a continuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement activity characteristic of the active advertisement in the target time and the historical advertisement activity characteristic of the active advertisement before the target time;
and when the delivery attribute of the active advertisement is the discontinuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement activity characteristic of the active advertisement in the target time and the historical average advertisement activity characteristic of all the advertisements before the target time.
Optionally, the target advertisement activity characteristics of the active advertisement in the target time include the number of times of advertisement impressions of the active advertisement in the target time and advertisement interaction data. Further optionally, the advertisement interaction data of the active advertisement in the target time includes one or more combinations of the number of praise, the number of forwarding, the number of comments, the number of browsing, the number of collection, the number of barrage, and the amount of appreciation.
In yet another alternative embodiment, the calculating module 303 calculates the liveness of each active advertisement according to the delivery parameter variation of each active advertisement and the liveness calculation model matched with the active advertisement, in a specific manner:
determining a first weight value corresponding to the change rate of the putting times and a second weight value corresponding to the change rate of the interactive data;
and substituting the first weight value, the second weight value, the change rate of the delivery times of each active advertisement and the change rate of the interaction data of each active advertisement into an activity calculation formula matched with the active advertisements to obtain the activity of each active advertisement.
In this optional embodiment, further optionally, the activity calculation formula is:
S＝W1*X1+W2*X2(ii) a Alternatively, the first and second electrodes may be,
S＝W3*(W1*X1+W2*X2)；
wherein S is the liveness of any active advertisement, W1Is a first weight value, W2Is a second weight value, X1The rate of change of the number of impressions, X, for the active advertisement2Interaction data change rate, W, for the active advertisement3The determined expansion factor corresponding to the active advertisement.
Yet further alternatively, X1And X2The calculation formulas of (A) and (B) are respectively as follows:
X1＝tanh((Td-Ty)/Ty)；
X2＝∑iwi*tanh((Rdi-Ryi)/Ryi)；
wherein, TdWhen the delivery attribute of the active advertisement is the continuous delivery attribute, T is the advertisement delivery times of the active advertisement in the target timeyWhen the active advertisement is delivered with the attribute of being delivered as the number of times of advertisement delivery in the preset time period before the target timeWhen attribute is not continuously released, TyAveraging the historical advertisement delivery times of each advertisement in all advertisements; i for representing data class, wiIs the weight value of the data class, RdiData volume of the data category in the advertisement interaction data in the target time for the active advertisement; when the delivery attribute of the active advertisement is a continuous delivery attribute, RyiWhen the delivery attribute of the active advertisement is the non-continuous delivery attribute, R is the data volume of the data category in the advertisement interaction data of the active advertisement in the preset time period before the target timeyiThe historical data amount of the data category in the historical advertisement interaction data of all advertisements is averaged.
Still further optionally, the expansion factor corresponding to each active advertisement is determined according to all placement areas of the active advertisement within the target time, and when all placement areas of the active advertisement within the target time are not unique, the expansion factor corresponding to the active advertisement is specifically determined according to a region weight value corresponding to each placement area of all placement areas of the active advertisement within the target time.
In yet another alternative embodiment, the obtaining module 302 may be further configured to obtain the historical liveness of each active advertisement after the calculating module 303 calculates the liveness of each active advertisement according to the liveness calculation model. And as shown in fig. 4, the apparatus may further include:
and the correcting module 304 is configured to correct the calculated liveness of each active advertisement according to the historical liveness of the active advertisement, so as to obtain a corrected liveness.
Therefore, the device described in fig. 4 can also calculate the delivery parameter change conditions of the active advertisements with different delivery attributes according to different data, thereby realizing the personalized calculation of the delivery parameter change conditions and being beneficial to improving the accuracy of the delivery parameter change conditions of the active advertisements with different delivery attributes. In addition, the initial liveness can be corrected according to the historical liveness after the initial liveness is calculated, and the corrected liveness is used as the finally estimated liveness, so that the liveness of the active advertisement in the appointed time can be corrected through the historical liveness, the condition that the estimated liveness is deviated due to sudden change of the advertisement liveness characteristics of the active advertisement in the appointed time (such as sudden increase of advertisement putting times and sudden increase of advertisement interaction data) is favorably reduced, and the accuracy of the finally estimated liveness is favorably further improved. In addition, besides the advertisement interaction data, the liveness of the active advertisements can be comprehensively evaluated from the dimensions of the delivery areas, and the liveness estimation of the advertisements in different delivery areas is facilitated.
Example four
Referring to fig. 5, fig. 5 is a schematic structural diagram of another liveness estimation device for active advertisements according to an embodiment of the present invention. As shown in fig. 5, the apparatus may include:
a memory 401 storing executable program code;
a processor 402 coupled with the memory 401;
the processor 402 calls the executable program code stored in the memory 401 to perform the steps of the liveness method for active advertisements disclosed in the first or second embodiment of the present invention.
EXAMPLE five
The embodiment of the invention discloses a computer storage medium, which stores computer instructions, and the computer instructions are used for executing the steps in the liveness method of the active advertisements disclosed in the first embodiment or the second embodiment of the invention when being called.
The above-described embodiments of the apparatus are merely illustrative, and the modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above detailed description of the embodiments, those skilled in the art will clearly understand that the embodiments may be implemented by software plus a necessary general hardware platform, and may also be implemented by hardware. Based on such understanding, the above technical solutions may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, wherein the storage medium includes a Read-Only Memory (ROM), a Random Access Memory (RAM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), a One-time Programmable Read-Only Memory (OTPROM), an Electrically Erasable rewritable Read-Only Memory (EEPROM), a compact disc-Read-Only Memory (CD-ROM) or other magnetic disk memories, a magnetic tape Memory, a magnetic disk, a magnetic tape Memory, a magnetic tape, and a magnetic tape, Or any other medium which can be used to carry or store data and which can be read by a computer.
Finally, it should be noted that: the method and apparatus for estimating liveness of a live advertisement disclosed in the embodiments of the present invention are only the preferred embodiments of the present invention, and are only used for illustrating the technical solutions of the present invention, not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art; the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (18)

1. A method for predicting liveness of a live advertisement, the method comprising:
after detecting liveness calculation instructions for the active advertisements, determining all active advertisements for which liveness needs to be calculated;
acquiring liveness calculation models matched with all the active advertisements, and calculating the liveness of each active advertisement according to the liveness calculation models;
wherein the determining all active advertisements for which liveness needs to be calculated comprises:
determining a target time indicated by the liveness calculation instruction;
screening advertisements delivered within the target time from all the counted advertisements to serve as active advertisements; or screening the advertisements delivered within the target time and delivered within a preset time period before the target time from all the counted advertisements as active advertisements.
2. The liveness prediction method of active advertisements as claimed in claim 1 wherein before said calculating the liveness of each of said active advertisements according to said liveness calculation model, said method further comprises:
determining the delivery attribute of each active advertisement, wherein the delivery attribute of the active advertisement is one of a continuous delivery attribute and a discontinuous delivery attribute;
calculating the delivery parameter change condition of each active advertisement according to the delivery attribute of each active advertisement, wherein the delivery parameter change condition of the active advertisement comprises the delivery frequency change rate and the interaction data change rate of the active advertisement;
wherein said calculating liveness of each of said active advertisements according to said liveness calculation model comprises:
and calculating the liveness of each active advertisement according to the delivery parameter change condition of each active advertisement and a liveness calculation model matched with the active advertisement.
3. The method of claim 2, wherein the calculating the variation of the delivery parameters of each active advertisement according to the delivery attributes of each active advertisement comprises:
when the delivery attribute of the active advertisement is the continuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement active characteristic of the active advertisement in the target time and the historical advertisement active characteristic of the active advertisement before the target time;
when the delivery attribute of the active advertisement is the discontinuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement activity characteristic of the active advertisement in the target time and the historical average advertisement activity characteristic of all advertisements before the target time.
4. The liveness estimation method of active advertisements according to claim 3, characterized in that the target advertisement liveness characteristics of the active advertisements in the target time comprise the advertisement putting times and advertisement interaction data of the active advertisements in the target time;
and the advertisement interaction data of the active advertisement in the target time comprises one or more combinations of the number of praise, the number of forwarding, the number of comments, the number of browsing, the number of collection, the number of barrage and the amount of appreciation.
5. The method for predicting the liveness of active advertisements according to any one of claims 2 to 4, wherein the calculating the liveness of each active advertisement according to the delivery parameter variation of each active advertisement and the liveness calculation model matched with the active advertisement comprises:
determining a first weight value corresponding to the change rate of the releasing times and a second weight value corresponding to the change rate of the interaction data;
and substituting the first weight value, the second weight value, the delivery times change rate of each active advertisement and the interaction data change rate of each active advertisement into an activity calculation formula matched with the active advertisements to obtain the activity of each active advertisement.
6. The liveness prediction method of active advertisements as claimed in claim 5 wherein the liveness calculation formula is:
S＝W1*X1+W2*X2(ii) a Alternatively, the first and second electrodes may be,
S＝W3*(W1*X1+W2*X2)；
wherein S is the liveness of any of the active advertisements, W1Is the first weight value, W2Is the second weight value, X1The rate of change of the number of impressions, X, for the active advertisement2The interaction data change rate, W, for the active advertisement3The determined expansion factor corresponding to the active advertisement.
7. The method of claim 6, wherein X is a measure of liveness of an active advertisement1And X2The calculation formulas of (A) and (B) are respectively as follows:
X1＝tanh((Td-Ty)/Ty)；
X2＝∑iwi*tanh((Rdi-Ryi)/Ryi)；
wherein, TdWhen the delivery attribute of the active advertisement is the continuous delivery attribute, T is the advertisement delivery times of the active advertisement in the target timeyWhen the delivery attribute of the active advertisement is the discontinuous delivery attribute, T is the advertisement delivery times of the active advertisement in a preset time period before the target timeyAveraging the historical advertisement delivery times of each advertisement in all advertisements;
i for representing data class, wiIs the weight value of the data class, RdiThe data volume of the data category in the advertisement interaction data of the active advertisement in the target time; when the delivery attribute of the active advertisement is the continuous delivery attribute, RyiThe data volume of the data category in the advertisement interaction data of the active advertisement in a preset time period before the target time is considered as the data volume of the data category in the active advertisementWhen the advertisement delivery attribute is the discontinuous delivery attribute, RyiThe historical data amount of the data category in the historical advertisement interaction data of all advertisements is averaged.
8. The method for predicting the liveness of active advertisements according to claim 6 or 7, characterized in that the expansion factor corresponding to each active advertisement is determined according to all the delivery areas of the active advertisement in the target time, and when all the delivery areas of the active advertisement in the target time are not unique, the expansion factor corresponding to the active advertisement is specifically determined according to the area weight value corresponding to each delivery area of all the delivery areas of the active advertisement in the target time.
9. The liveness estimation method of active advertisements according to claim 8, wherein after calculating the liveness of each of the active advertisements according to the liveness calculation model, the method further comprises:
and acquiring the historical liveness of each active advertisement, and correcting the calculated liveness of the active advertisement according to the historical liveness of each active advertisement to obtain the corrected liveness.
10. An apparatus for predicting liveness of a live advertisement, the apparatus comprising:
the determining module is used for determining all active advertisements needing to calculate the liveness after detecting the liveness calculating instruction aiming at the active advertisements;
the obtaining module is used for obtaining an activity degree calculation model matched with all the active advertisements;
the calculating module is used for calculating the liveness of each active advertisement according to the liveness calculating model;
the specific way of determining all active advertisements needing to calculate liveness by the determining module is as follows:
determining a target time indicated by the liveness calculation instruction;
screening advertisements delivered within the target time from all the counted advertisements to serve as active advertisements; or screening the advertisements delivered within the target time and delivered within a preset time period before the target time from all the counted advertisements as active advertisements.
11. The liveness prediction device of active advertisements according to claim 10, wherein the determining module is further configured to determine the placement attribute of each of the active advertisements before the calculating module calculates the liveness of each of the active advertisements according to the liveness calculation model, and the placement attribute of the active advertisements is one of a continuous placement attribute and a discontinuous placement attribute;
the calculation module is further configured to calculate a delivery parameter change condition of each active advertisement according to a delivery attribute of each active advertisement, where the delivery parameter change condition of each active advertisement includes a delivery frequency change rate and an interaction data change rate of each active advertisement;
the specific way for the calculation module to calculate the liveness of each active advertisement according to the liveness calculation model is as follows:
and calculating the liveness of each active advertisement according to the delivery parameter change condition of each active advertisement and a liveness calculation model matched with the active advertisement.
12. The apparatus for predicting liveness of active advertisements as claimed in claim 11, wherein the computing module computes the liveness of each active advertisement according to the delivery parameter variation of each active advertisement and the liveness computation model matched with the active advertisement by:
when the delivery attribute of the active advertisement is the continuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement active characteristic of the active advertisement in the target time and the historical advertisement active characteristic of the active advertisement before the target time;
when the delivery attribute of the active advertisement is the discontinuous delivery attribute, calculating the delivery parameter change condition of the active advertisement based on the target advertisement activity characteristic of the active advertisement in the target time and the historical average advertisement activity characteristic of all advertisements before the target time.
13. The liveness estimation device of active advertisements according to claim 12, wherein the target advertisement liveness characteristics of the active advertisements in the target time comprise advertisement placement times and advertisement interaction data of the active advertisements in the target time;
and the advertisement interaction data of the active advertisement in the target time comprises one or more combinations of the number of praise, the number of forwarding, the number of comments, the number of browsing, the number of collection, the number of barrage and the amount of appreciation.
14. The apparatus for predicting liveness of active advertisements as claimed in any one of claims 11-13, wherein said calculating module calculates liveness of each of said active advertisements according to variation of delivery parameters of each of said active advertisements and a liveness calculation model matched with said active advertisements by:
determining a first weight value corresponding to the change rate of the releasing times and a second weight value corresponding to the change rate of the interaction data;
and substituting the first weight value, the second weight value, the delivery times change rate of each active advertisement and the interaction data change rate of each active advertisement into an activity calculation formula matched with the active advertisements to obtain the activity of each active advertisement.
15. The liveness prediction device of an active advertisement as claimed in claim 14, wherein the liveness calculation formula is:
S＝W1*X1+W2*X2(ii) a Alternatively, the first and second electrodes may be,
S＝W3*(W1*X1+W2*X2)；
wherein S is the liveness of any of the active advertisements, W1Is the first weight value, W2Is the second weight value, X1The rate of change of the number of impressions, X, for the active advertisement2The interaction data change rate, W, for the active advertisement3The determined expansion factor corresponding to the active advertisement.
16. The liveness prediction device of an active advertisement as claimed in claim 15, wherein X is1And X2The calculation formulas of (A) and (B) are respectively as follows:
X1＝tanh((Td-Ty)/Ty)；
X2＝∑iwi*tanh((Rdi-Ryi)/Ryi)；
wherein, TdWhen the delivery attribute of the active advertisement is the continuous delivery attribute, T is the advertisement delivery times of the active advertisement in the target timeyWhen the delivery attribute of the active advertisement is the discontinuous delivery attribute, T is the advertisement delivery times of the active advertisement in a preset time period before the target timeyAveraging the historical advertisement delivery times of each advertisement in all advertisements;
i for representing data class, wiIs the weight value of the data class, RdiThe data volume of the data category in the advertisement interaction data of the active advertisement in the target time; when the delivery attribute of the active advertisement is the continuous delivery attribute, RyiWhen the delivery attribute of the active advertisement is the non-continuous delivery attribute, R is the data volume of the data category in the advertisement interaction data of the active advertisement in the preset time period before the target timeyiHistorical number of data of the data category in historical advertisement interaction data for all advertisementsData volume average.
17. The apparatus for predicting liveness of active advertisements as claimed in claim 15 or 16, wherein the expansion factor corresponding to each active advertisement is determined according to all delivery areas of the active advertisement within the target time, and when all delivery areas of the active advertisement within the target time are not unique, the expansion factor corresponding to the active advertisement is specifically determined according to a region weight value corresponding to each delivery area of all delivery areas of the active advertisement within the target time.
18. The liveness prediction device of active advertisements as claimed in claim 17, wherein the obtaining module is further configured to obtain the historical liveness of each of the active advertisements after the calculating module calculates the liveness of each of the active advertisements according to the liveness calculation model;
and, the apparatus further comprises:
and the correction module is used for correcting the calculated liveness of the active advertisement according to the historical liveness of each active advertisement to obtain the corrected liveness.
CN202010361126.7A 2020-04-30 2020-04-30 Method and device for estimating liveness of liveness advertisement Active CN111681035B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010361126.7A CN111681035B (en) 2020-04-30 2020-04-30 Method and device for estimating liveness of liveness advertisement

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010361126.7A CN111681035B (en) 2020-04-30 2020-04-30 Method and device for estimating liveness of liveness advertisement

Publications (2)

Publication Number Publication Date
CN111681035A true CN111681035A (en) 2020-09-18
CN111681035B CN111681035B (en) 2024-04-09

Family

ID=72452657

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010361126.7A Active CN111681035B (en) 2020-04-30 2020-04-30 Method and device for estimating liveness of liveness advertisement

Country Status (1)

Country Link
CN (1) CN111681035B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112116264A (en) * 2020-09-24 2020-12-22 北京易华录信息技术股份有限公司 Activity evaluation method and apparatus
CN115018551A (en) * 2022-06-29 2022-09-06 北京飞天经纬科技股份有限公司 CRM data management method and device for advertisement service and storage medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103065260A (en) * 2013-01-22 2013-04-24 分众(中国)信息技术有限公司 Interactive advertisement information management system based on cloud computing
US20160275544A1 (en) * 2015-03-16 2016-09-22 Electronics And Telecommunications Research Institute Apparatus for analyzing advertisement effect and method of analyzing advertisement effect
CN108924597A (en) * 2018-06-26 2018-11-30 北京酷云互动科技有限公司 Channel hot value appraisal procedure, hot spot acquisition methods and its system
CN109275031A (en) * 2018-09-25 2019-01-25 有米科技股份有限公司 A kind of temperature appraisal procedure, device and the electronic equipment of video
CN109801107A (en) * 2019-01-17 2019-05-24 有米科技股份有限公司 A kind of advertisement liveness predictor method, device, server and storage medium
CN109801100A (en) * 2018-12-26 2019-05-24 北京达佳互联信息技术有限公司 Advertisement placement method, device and computer readable storage medium
CN110533439A (en) * 2018-05-23 2019-12-03 北京国双科技有限公司 The method and device of value is launched in a kind of evaluation advertisement
CN110910165A (en) * 2019-10-29 2020-03-24 深圳市星河互动科技有限公司 Multi-channel attribution method and related equipment

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103065260A (en) * 2013-01-22 2013-04-24 分众(中国)信息技术有限公司 Interactive advertisement information management system based on cloud computing
US20160275544A1 (en) * 2015-03-16 2016-09-22 Electronics And Telecommunications Research Institute Apparatus for analyzing advertisement effect and method of analyzing advertisement effect
CN110533439A (en) * 2018-05-23 2019-12-03 北京国双科技有限公司 The method and device of value is launched in a kind of evaluation advertisement
CN108924597A (en) * 2018-06-26 2018-11-30 北京酷云互动科技有限公司 Channel hot value appraisal procedure, hot spot acquisition methods and its system
CN109275031A (en) * 2018-09-25 2019-01-25 有米科技股份有限公司 A kind of temperature appraisal procedure, device and the electronic equipment of video
CN109801100A (en) * 2018-12-26 2019-05-24 北京达佳互联信息技术有限公司 Advertisement placement method, device and computer readable storage medium
CN109801107A (en) * 2019-01-17 2019-05-24 有米科技股份有限公司 A kind of advertisement liveness predictor method, device, server and storage medium
CN110910165A (en) * 2019-10-29 2020-03-24 深圳市星河互动科技有限公司 Multi-channel attribution method and related equipment

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112116264A (en) * 2020-09-24 2020-12-22 北京易华录信息技术股份有限公司 Activity evaluation method and apparatus
CN115018551A (en) * 2022-06-29 2022-09-06 北京飞天经纬科技股份有限公司 CRM data management method and device for advertisement service and storage medium

Also Published As

Publication number Publication date
CN111681035B (en) 2024-04-09

Similar Documents

Publication Publication Date Title
US20240221028A1 (en) Preservation of scores of the quality of traffic to network sites across clients and over time
CN110033314B (en) Advertisement data processing method and device
CN106022834B (en) Advertisement anti-cheating method and device
US7346471B2 (en) Web data outlier detection and mitigation
CN111681035A (en) Liveness estimation method and device for active advertisements
KR101012165B1 (en) Method and system for offering advertisement adjusting bidding price per unit time according to advertising traffic
CN108009862B (en) Method and device for early warning of balance of advertising expense
US20080065464A1 (en) Predicting response rate
US8219447B1 (en) Real-time adaptive probabilistic selection of messages
US20110046991A1 (en) Systems and methods for comparing and improving sales performance over heterogeneous geographical sales regions
CN111260414B (en) Advertisement inventory prediction method and device
CN106296247B (en) Network information resource online ordering method and device
CN105590240A (en) Discrete calculating method of brand advertisement effect optimization
CN108932627B (en) Estimation method and device for consumption speed of CPA (continuous advertisement access) advertisement and readable storage medium
CN105608604A (en) Continuous calculation method of brand advertisement effectiveness optimization
CN111681036B (en) Advertisement liveness estimation implementation method and device
CN110889737A (en) Method and device for recommending rental roommates, server and storage medium
CN112001774A (en) Neural network-based tobacco input quantity research and judgment method and system
CN112235636A (en) Method and device for calculating browsing value attribute of video with goods
CN111695929A (en) Method and device for carrying out liveness estimation on non-active advertisement
CN106682930B (en) Data processing method and device
CN111985971A (en) Advertisement screening method, device, equipment and storage medium
CN115544385B (en) Platform recommendation method, electronic device and computer-readable storage medium
CN106097045B (en) Multimedia resource pushing method and device
CN111340532B (en) Data delivery method, device, server and computer readable storage medium

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