CN112700285A - Method, device and equipment for predicting user attribute label - Google Patents
Method, device and equipment for predicting user attribute label Download PDFInfo
- Publication number
- CN112700285A CN112700285A CN202110021986.0A CN202110021986A CN112700285A CN 112700285 A CN112700285 A CN 112700285A CN 202110021986 A CN202110021986 A CN 202110021986A CN 112700285 A CN112700285 A CN 112700285A
- Authority
- CN
- China
- Prior art keywords
- probability
- user
- advertisement
- data
- acquiring
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000004422 calculation algorithm Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 6
- 238000007477 logistic regression Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 101150053844 APP1 gene Proteins 0.000 description 4
- 101100189105 Homo sapiens PABPC4 gene Proteins 0.000 description 4
- 102100039424 Polyadenylate-binding protein 4 Human genes 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 239000013598 vector Substances 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 101100055496 Arabidopsis thaliana APP2 gene Proteins 0.000 description 2
- 101100016250 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) GYL1 gene Proteins 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
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
- G06Q30/0251—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
- G06Q30/0201—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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- 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
- G06Q30/0254—Targeted advertisements based on statistics
-
- 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
- G06Q30/0271—Personalized advertisement
-
- 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
- G06Q30/0277—Online advertisement
Abstract
The application relates to the technical field of attribute label prediction, and discloses a method for predicting a user attribute label, which comprises the following steps: acquiring first equipment data corresponding to a user to be detected, and acquiring first advertisement data corresponding to the user to be detected within a preset time period; acquiring a first probability of matching the first device data with the target attribute label, acquiring a second probability of matching the first advertisement data with the target attribute label, and acquiring a first mean value and a first variance of the first advertisement data in a preset unit time; and obtaining an attribute label prediction result of the user to be tested according to one or more of the first equipment data, the first advertisement data, the first probability, the second probability, the first mean value and the first variance. The method can be used for predicting the user attribute label by fusing multi-dimensional user data characteristics, and the accuracy of predicting the user attribute label is improved. The application also discloses a device and equipment for predicting the user attribute label.
Description
Technical Field
The present application relates to the technical field of attribute tag prediction, and for example, to a method, an apparatus, and a device for predicting a user attribute tag.
Background
Currently, attribute tags of users, for example: gender, age, education level, and the like are important input features used in various fields such as recommendation systems, advertisement delivery, and smart marketing. User preferences for advertisements or various goods may differ depending on their age, gender, education, income, etc. If the attribute labels of the users can be accurately obtained, the method is more accurate for various recommendation systems or advertisement delivery.
In the process of implementing the embodiments of the present disclosure, it is found that at least the following problems exist in the related art: in the prior art, the user attribute tags are predicted by the brand, industry and the like of the advertisement clicked by the user, and the accuracy of the prediction of the user attribute tags is low.
Disclosure of Invention
The following presents a simplified summary in order to provide a basic understanding of some aspects of the disclosed embodiments. This summary is not an extensive overview nor is intended to identify key/critical elements or to delineate the scope of such embodiments but rather as a prelude to the more detailed description that is presented later.
The embodiment of the disclosure provides a method, a device and equipment for predicting user attribute tags, so as to improve the accuracy of predicting the user attribute tags.
In some embodiments, the method for predicting user attribute tags comprises:
acquiring first equipment data corresponding to a user to be detected, and acquiring first advertisement data corresponding to the user to be detected within a preset time period; the first advertisement data comprise the times of clicking the advertisement by the user to be tested, the times of clicking the application program corresponding to the advertisement clicked by the user to be tested, and the times of clicking the media corresponding to the advertisement clicked by the user to be tested;
acquiring a first probability of matching the first device data with a target attribute label, acquiring a second probability of matching the first advertisement data with the target attribute label, and acquiring a first mean value and a first variance of the first advertisement data in a preset unit time;
and obtaining an attribute label prediction result of the user to be tested according to one or more of the first equipment data, the first advertisement data, the first probability, the second probability, the first mean value and the first variance.
In some embodiments, the means for predicting a user attribute tag comprises: a processor and a memory storing program instructions, the processor being configured to, upon execution of the program instructions, perform the method for predicting user attribute tags described above.
In some embodiments, the apparatus comprises the above-described means for predicting user attribute tags.
The method, the device and the equipment for predicting the user attribute label provided by the embodiment of the disclosure can realize the following technical effects: the method comprises the steps of obtaining first equipment data corresponding to a user to be detected and first advertisement data corresponding to the user to be detected in a preset time period, and obtaining an attribute tag prediction result of the user to be detected through one or more of a first probability that the first equipment data is matched with a target attribute tag, a second probability that the first advertisement data is matched with the target attribute tag, a first mean value of the first advertisement data in a preset unit time and a first variance of the first advertisement data in the preset unit time. On the basis of user data characteristics such as first equipment data and first advertisement data corresponding to a user to be detected, a first probability of matching the first equipment data with a target attribute label, a second probability of matching the first advertisement data with the target attribute label, and a first mean value and a first variance of the first advertisement data in a preset unit time are increased, and multidimensional user data characteristics are fused to predict a user attribute label, so that the accuracy of predicting the user attribute label can be improved.
The foregoing general description and the following description are exemplary and explanatory only and are not restrictive of the application.
Drawings
One or more embodiments are illustrated by way of example in the accompanying drawings, which correspond to the accompanying drawings and not in limitation thereof, in which elements having the same reference numeral designations are shown as like elements and not in limitation thereof, and wherein:
FIG. 1 is a schematic diagram of a method for predicting user attribute tags according to an embodiment of the present disclosure;
fig. 2 is a schematic diagram of an apparatus for predicting a user attribute tag according to an embodiment of the present disclosure.
Detailed Description
So that the manner in which the features and elements of the disclosed embodiments can be understood in detail, a more particular description of the disclosed embodiments, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. In the following description of the technology, for purposes of explanation, numerous details are set forth in order to provide a thorough understanding of the disclosed embodiments. However, one or more embodiments may be practiced without these details. In other instances, well-known structures and devices may be shown in simplified form in order to simplify the drawing.
The terms "first," "second," and the like in the description and in the claims, and the above-described drawings of embodiments of the present disclosure, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It should be understood that the data so used may be interchanged under appropriate circumstances such that embodiments of the present disclosure described herein may be made. Furthermore, the terms "comprising" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions.
The term "plurality" means two or more unless otherwise specified.
In the embodiment of the present disclosure, the character "/" indicates that the preceding and following objects are in an or relationship. For example, A/B represents: a or B.
The term "and/or" is an associative relationship that describes objects, meaning that three relationships may exist. For example, a and/or B, represents: a or B, or A and B.
As shown in fig. 1, an embodiment of the present disclosure provides a method for predicting a user attribute tag, including:
step S101, acquiring first equipment data corresponding to a user to be detected, and acquiring first advertisement data corresponding to the user to be detected within a preset time period; the first advertisement data comprises the times of clicking the advertisement by the user to be tested, the times of clicking the application program corresponding to the advertisement clicked by the user to be tested, and the times of clicking the media corresponding to the advertisement clicked by the user to be tested;
step S102, acquiring a first probability of matching the first device data with the target attribute label, acquiring a second probability of matching the first advertisement data with the target attribute label, and acquiring a first mean value and a first variance of the first advertisement data in a preset unit time;
step S103, obtaining an attribute label prediction result of the user to be tested according to one or more of the first device data, the first advertisement data, the first probability, the second probability, the first mean value and the first variance.
By adopting the method for predicting the user attribute label provided by the embodiment of the disclosure, the first device data corresponding to the user to be detected and the first advertisement data corresponding to the user to be detected in the preset time period can be obtained, and the attribute label prediction result of the user to be detected can be obtained through one or more of the first probability of matching the first device data with the target attribute label, the second probability of matching the first advertisement data with the target attribute label, the first mean value of the first advertisement data in the preset unit time and the first variance of the first advertisement data in the preset unit time. On the basis of user data characteristics such as first equipment data and first advertisement data corresponding to a user to be detected, a first probability of matching the first equipment data with a target attribute label, a second probability of matching the first advertisement data with the target attribute label, and a first mean value and a first variance of the first advertisement data in a preset unit time are increased, and multidimensional user data characteristics are fused to predict a user attribute label, so that the accuracy of predicting the user attribute label can be improved.
Optionally, the user attribute tags include a gender attribute tag, an age attribute tag, an education level attribute tag, and the like. In some embodiments, in the case that the gender attribute tag of the user to be tested is predicted, then the target attribute tag is the gender attribute tag. Optionally, the gender attribute tags include "male" and "female".
Optionally, the first device data comprises: and the equipment brand and the equipment model corresponding to the user to be tested. For example: "Hua is glorious 10 youth edition".
Optionally, advertisement site location information of the advertisement clicked by the user to be tested is obtained. Optionally, an Application program (APP, Application) corresponding to the advertisement is obtained through the advertisement spot location, and the number of times of clicking the Application program corresponding to the advertisement clicked by the user to be tested is recorded. Optionally, obtaining an application program corresponding to an advertisement through an advertisement spot includes: and matching the application programs corresponding to the advertisement site locations in a preset first data table, wherein the first data table stores the corresponding relation between the advertisement site locations and the application programs.
As shown in table 1, table 1 is an example table of the first data table.
| Advertisement position | Application program |
| Spid1 | APP1 |
| Spid2 | APP1 |
| Spid3 | APP2 |
| Spid4 | APP2 |
In some embodiments, as shown in table 1, the application programs corresponding to the advertisement spots Spid1 and Spid2 are APP1, and when the user to be tested clicks the advertisement Spid1 twice and clicks the Spid2 three times, the number of clicks corresponding to the APP1 is five.
Optionally, media corresponding to the advertisement is obtained through the advertisement point location, and the media click times corresponding to the user to be tested clicking the advertisement are recorded. Optionally, obtaining media corresponding to the advertisement through the advertisement spot includes: and matching the media corresponding to the advertisement site locations in a preset second data table, wherein the second data table stores the corresponding relation between the advertisement site locations and the media. As shown in table 2, table 2 is an example table of the second data table.
| Advertisement position | Media |
| Spid1 | Medium 1 |
| Spid2 | Medium 1 |
| Spid3 | Medium 1 |
| Spid4 | Medium 2 |
In some embodiments, as shown in table 2, the media corresponding to the advertisement spots Spid1, Spid2, and Spid3 are all media 1, and when the user to be tested clicks the advertisement Spid1 twice, clicks the Spid2 three times, and clicks the Spid1 four times, the number of clicks corresponding to the media 1 is nine times.
Optionally, the obtaining a first mean and a first variance of the first advertisement data in a preset unit time includes: the method comprises the steps of obtaining a first mean value and a first variance of the times of clicking the advertisement by a user to be detected in a preset unit time. Optionally, the unit time is the time of day the advertisement was clicked. Optionally, the preset time period is 1 month. In some embodiments, in 1 month, the advertisement is clicked by the user to be tested for only two days, the advertisement clicking time on the first day is 10, the advertisement clicking time on the second day is 5, and then the average value of the advertisement clicking times on a single day of the user to be tested is 7.5, and the variance is 6.25.
Optionally, the obtaining a first mean and a first variance of the first advertisement data in a preset unit time includes: the method comprises the steps of obtaining a first mean value and a first variance of the number of times of clicking of an application program corresponding to the advertisement clicked by a user to be detected in a preset unit time. In some embodiments, in 1 month, the user to be tested clicks the advertisement for only two days, the number of clicks of the APP corresponding to the advertisement on the first day is 8, and the number of clicks of the APP corresponding to the advertisement on the second day is 4, so that the average of the number of clicks of the APP corresponding to the advertisement clicked by the user to be tested for one day is 6, and the variance is 4.
Optionally, the obtaining a first mean and a first variance of the first advertisement data in a preset unit time includes: the method comprises the steps of obtaining a first mean value and a first variance of media click times corresponding to the times that a user to be detected clicks the advertisement in a preset unit time. In some embodiments, in 1 month, the user to be tested clicks the advertisement for only two days, the media click number corresponding to the advertisement on the first day is 8, the media click number corresponding to the advertisement on the second day is 4, the average value of the media click numbers corresponding to the advertisement clicked by the user to be tested on a single day is 6, and the variance is 4.
In this way, the attribute label prediction result of the user to be detected is obtained according to the first mean value and the first variance of the first advertisement data in the preset unit time, so that the activity and the stability of the user to be detected can be embodied.
Optionally, obtaining a first probability that the first device data matches the target attribute tag includes: searching a first probability of matching the first equipment data with the target attribute tag from a preset tenth data table; the tenth data table stores a first probability of matching the first device data with the target attribute tag.
Optionally, obtaining a second probability that the first advertisement data matches the target attribute tag includes: acquiring an eighth probability that the advertisement clicked by the user to be detected is matched with the target attribute label; and calculating by using the eighth probability according to a fourth preset algorithm to obtain a second probability.
Optionally, obtaining an eighth probability that the advertisement clicked by the user to be tested matches the target attribute tag includes: acquiring first attribute information of a user to be tested clicking an advertisement; searching an eleventh probability that the first attribute information matches the target attribute tag in a preset third data table; taking the average value of the eleventh probabilities of all the first attribute information matching the target attribute labels as the eighth probability that the advertisements clicked by the user to be tested are matched with the target attribute labels; and the third data table stores the eleventh probability of matching the first attribute information with the target attribute label. Optionally, the first attribute information includes an advertiser, a brand, a commodity, and an industry of the advertisement.
Optionally, the obtaining the second probability by performing a calculation according to a fourth preset algorithm by using an eighth probability includes: by calculation of
Obtaining a second probability; wherein, P2In order to be the second probability that,
the number of times of the nth advertisement clicked by the user to be tested,
and matching the nth advertisement clicked by the user to be detected with the eighth probability of the target attribute label, wherein n and m are positive integers.
Optionally, obtaining a second probability that the first advertisement data matches the target attribute tag includes: acquiring a ninth probability that an application program corresponding to the advertisement clicked by the user to be detected matches the target attribute label; and calculating by utilizing the ninth probability according to a fifth preset algorithm to obtain a second probability.
Optionally, obtaining a ninth probability that the application program corresponding to the advertisement clicked by the user to be tested matches the target attribute tag includes: finding out a ninth probability that the application program matches the target attribute label in a preset fourth data table; and the ninth probability of matching the application program and the target attribute label is stored in the fourth data table.
Optionally, the calculating according to a fifth preset algorithm by using the ninth probability to obtain the second probability includes: by calculation of
Obtaining a second probability; wherein, P2In order to be the second probability that,
the number of clicks of the o-th application program corresponding to the user to be tested,
and matching the ninth probability of the target attribute label for the o-th application program corresponding to the user to be tested, wherein o and q are positive integers.
Optionally, obtaining a second probability that the first advertisement data matches the target attribute tag includes: acquiring a tenth probability that the user to be tested clicks the media matching target attribute tag corresponding to the advertisement; and calculating by utilizing the tenth probability according to a sixth preset algorithm to obtain a second probability.
Optionally, obtaining a tenth probability that the media corresponding to the advertisement clicked by the user to be tested matches the target attribute tag includes: finding out a tenth probability of the media matching with the target attribute tag in a preset fifth data table; and the fifth data table stores the tenth probability of matching the media with the target attribute label.
Optionally, the obtaining the second probability by performing a calculation according to a sixth preset algorithm by using the tenth probability includes: by calculation of
Obtaining a second probability; wherein, P2In order to be the second probability that,
the number of clicks of the v-th media corresponding to the user to be tested,
and matching the tenth probability of the target attribute label for the v-th media corresponding to the user to be tested, wherein v and w are positive integers.
Optionally, obtaining an attribute tag prediction result of the user to be tested according to one or more of the first device data, the first advertisement data, the first probability, the second probability, the first mean, and the first variance, includes: inputting the first equipment data, the first advertisement data, the first probability, the second probability, the first mean value and the first variance into a preset attribute label prediction model to obtain an attribute label prediction result of a user to be detected; the attribute label prediction model is obtained according to second equipment data corresponding to the sample user and second advertisement data corresponding to the sample user; the second advertisement data comprises the times of clicking the advertisement by the sample user, the times of clicking the application program corresponding to the advertisement by the sample user, and the times of clicking the media corresponding to the advertisement by the sample user. The prior knowledge of the user attribute label is obtained according to the first probability of matching the first device data with the target attribute label and the second probability of matching the first advertisement data with the target attribute label, and the obtained prior knowledge of the user attribute label is input into the attribute label prediction model, so that the prediction effect of the attribute label prediction model can be improved. Due to the introduction of the prior knowledge, the complexity of the attribute tag prediction model can be reduced, so that the efficiency of predicting the user attribute tag through the attribute tag prediction model is improved.
Optionally, the attribute tag prediction result of the user to be tested includes an Identity Document (ID) of the user to be tested, the attribute tag, and a probability value of the attribute tag.
Optionally, inputting the first device data, the first advertisement data, the first probability, the second probability, the first mean value, and the first variance into a preset attribute label prediction model, including: vectorizing the first device data, the first advertisement data, the first probability, the second probability, the first mean value and the first variance to obtain feature vectors corresponding to the first device data, the first advertisement data, the first probability, the second probability, the first mean value and the first variance, and inputting the feature vectors into a preset attribute label prediction model.
Optionally, advertisement site location information of a sample user clicking an advertisement is obtained. Optionally, the application program corresponding to the advertisement is obtained through the advertisement spot location, and the number of times that the sample user clicks the application program corresponding to the advertisement is recorded. Optionally, media corresponding to the advertisement is obtained through the advertisement point location, and the number of times of clicking of the media corresponding to the sample user clicking the advertisement is recorded.
Optionally, obtaining an attribute tag prediction model according to the second device data corresponding to the sample user and the second advertisement data corresponding to the sample user includes: acquiring a third probability of matching the second device data with the target attribute label, acquiring a fourth probability of matching the second advertisement data with the target attribute label, and acquiring a second mean value and a second variance of the second advertisement data in a preset unit time; and training the logistic regression model by using the second equipment data, the second advertisement data, the third probability, the fourth probability, the second mean value and the second variance to obtain an attribute label prediction model. The priori knowledge of the population attribute labels is obtained by the third probability of matching the second equipment data with the target attribute labels and the fourth probability of matching the second advertisement data with the target attribute labels, and the obtained priori knowledge of the user attribute labels is input into the logistic regression model, so that the prediction effect of the attribute label prediction model can be improved. Due to the introduction of the prior knowledge, the complexity of the attribute label prediction model can be reduced, and therefore the efficiency of the training of the attribute label prediction model is improved.
Optionally, real attribute labels of a plurality of training sample users are obtained.
Optionally, the training the logistic regression model with second device data, second advertisement data, a third probability, a fourth probability, a second mean, and a second variance to obtain an attribute label prediction model, including: vectorizing the second device data, the second advertisement data, the third probability, the fourth probability, the second mean value and the second variance to obtain feature vectors corresponding to the second device data, the second advertisement data, the third probability, the fourth probability, the second mean value and the second variance respectively, training the logistic regression model by taking the feature vectors as input features and the real attribute labels of the sample users as output features to obtain an attribute label prediction model.
Optionally, the obtaining a second mean and a second variance of the second advertisement data in a preset unit time includes: and acquiring a second mean value and a second variance of the times of clicking the advertisement by the sample users in a preset unit time.
Optionally, the obtaining a second mean and a second variance of the second advertisement data in a preset unit time includes: and acquiring a second mean value and a second variance of the application program clicking times corresponding to the advertisement clicked by the sample user in a preset unit time.
Optionally, the obtaining a second mean and a second variance of the second advertisement data in a preset unit time includes: and acquiring a second mean value and a second variance of the media click times corresponding to the advertisement clicked by the sample user in a preset unit time.
In this way, the attribute label prediction result of the sample user is obtained according to the first sample mean value and the second sample variance of the first sample advertisement data in the preset unit time, so that the activity and stability of the sample user can be embodied.
Optionally, the second device data comprises: and sampling the equipment brand and the equipment model corresponding to the user.
Optionally, obtaining a third probability that the second device data matches the target attribute tag includes: finding out a third probability of matching the second equipment data with the target attribute tag from a preset sixth data table; and the sixth data table stores a third probability of matching the second device data with the target attribute tag.
Optionally, obtaining a third probability that the second device data matches the target attribute tag includes: by calculation of
Obtaining a third probability P3Wherein P is3And x is the second equipment number corresponding to the sample user with the real attribute label as the target attribute label, and z is the total number of the second equipment corresponding to all the sample users.
Optionally, obtaining a fourth probability that the second advertisement data matches the target attribute tag includes: acquiring a fifth probability that the advertisement clicked by the sample user is matched with the target attribute label; and calculating by utilizing the fifth probability according to a first preset algorithm to obtain a fourth probability.
Optionally, obtaining a fifth probability that the advertisement clicked by the sample user matches the target attribute tag includes: acquiring second attribute information of the advertisement clicked by the sample user; finding out a twelfth probability that the second attribute information matches the target attribute label in a preset seventh data table; taking the average value of the twelfth probabilities that the second attribute information matches the target attribute label as a fifth probability that the advertisement clicked by the sample user matches the target attribute label; and the seventh data table stores the twelfth probability of matching the second attribute information with the target attribute label. Optionally, the second attribute information includes an advertiser, brand, commodity, industry of the advertisement.
Optionally, the obtaining the fourth probability by performing a calculation according to the first preset algorithm by using the fifth probability includes: by calculation of
Obtaining a fourth probability; wherein, P4In order to be the fourth probability,
for the sample number of times the user clicked on the nth' advertisement,
a fifth probability that the nth ' advertisement clicked by the sample user matches the target attribute label, wherein n ' and m ' are both positive integers.
Optionally, obtaining a fourth probability that the second advertisement data matches the target attribute tag includes: acquiring a sixth probability that the application program corresponding to the advertisement clicked by the sample user matches the target attribute label; and calculating by utilizing the sixth probability according to a second preset algorithm to obtain a fourth probability.
Optionally, obtaining a sixth probability that the application program corresponding to the sample user clicking the advertisement matches the target attribute tag includes: searching a sixth probability that the application program matches the target attribute label in a preset eighth data table; and the eighth data table stores a sixth probability of matching the application program with the target attribute tag.
Optionally, obtaining a sixth probability that the application program corresponding to the sample user clicking the advertisement matches the target attribute tag includes: by calculation of
Obtaining a sixth probability, wherein P6And x 'is the number of sample users corresponding to the application program, the real attribute label is the target attribute label, and z' is the number of sample users corresponding to the application program.
Optionally, the calculating according to the second preset algorithm by using the sixth probability to obtain the fourth probability includes: by calculation of
Obtaining a fourth probability; wherein, P4In order to be the fourth probability,
for the number of clicks of the o' th application corresponding to the sample user,
and a sixth probability of matching the target attribute label for the o ' th application corresponding to the sample user, wherein o ' and q ' are positive integers.
Optionally, obtaining a fourth probability that the second advertisement data matches the target attribute tag includes: acquiring a seventh probability that a sample user clicks a media matching target attribute label corresponding to the advertisement; and calculating by using the seventh probability according to a third preset algorithm to obtain a fourth probability.
Optionally, obtaining a seventh probability that the media corresponding to the sample user clicking the advertisement matches the target attribute tag includes: the method comprises the following steps: searching a seventh probability that the media matches the target attribute tag in a preset ninth data table; and the ninth data table stores a seventh probability of matching the media with the target attribute tag.
Optionally, the media matching target attribute corresponding to the sample user click advertisement is obtainedA seventh probability of a tag comprising: by calculation of
Obtaining a seventh probability, wherein P7For the seventh probability, x "is the number of sample users corresponding to the media, and the real attribute label is the target attribute label, and z" is the number of sample users corresponding to the media.
Optionally, the obtaining of the fourth probability by performing a calculation according to a third preset algorithm by using a seventh probability includes: by calculation of
Obtaining a fourth probability; wherein, P4In order to be the fourth probability,
for the number of clicks of the v' th media corresponding to the sample user,
and matching the nth media corresponding to the sample user with the seventh probability of the target attribute label, wherein v 'and w' are positive integers.
Optionally, after obtaining the attribute tag prediction result of the user to be tested, the method further includes: and pushing the attribute label prediction result to the user to be tested.
Optionally, the attribute tag prediction result is pushed to the user to be tested through the mobile terminal.
Therefore, the attribute label prediction result of the user to be detected is obtained through one or more of the first equipment data, the first advertisement data, the first probability, the second probability, the first mean value and the first variance, the attribute label of the user can be obtained more accurately, various recommendation systems or advertisement putting can be more accurate, benefit maximization can be achieved for a commodity manufacturer or an advertiser, meanwhile, the commodity manufacturer can analyze the attribute information of a target user conveniently, and optimization and operation are facilitated.
As shown in fig. 2, an apparatus for predicting a user attribute tag according to an embodiment of the present disclosure includes a processor (processor)100 and a memory (memory) 101. Optionally, the apparatus may also include a Communication Interface (Communication Interface)102 and a bus 103. The processor 100, the communication interface 102, and the memory 101 may communicate with each other via a bus 103. The communication interface 102 may be used for information transfer. The processor 100 may invoke logic instructions in the memory 101 to perform the method for predicting user attribute tags of the above embodiments.
In addition, the logic instructions in the memory 101 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products.
The memory 101, which is a computer-readable storage medium, may be used for storing software programs, computer-executable programs, such as program instructions/modules corresponding to the methods in the embodiments of the present disclosure. The processor 100 executes functional applications and data processing, i.e. implements the method for predicting user attribute tags in the above embodiments, by executing program instructions/modules stored in the memory 101.
The memory 101 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal device, and the like. In addition, the memory 101 may include a high-speed random access memory, and may also include a nonvolatile memory.
By adopting the device for predicting the user attribute tag provided by the embodiment of the disclosure, the first device data corresponding to the user to be detected and the first advertisement data corresponding to the user to be detected in the preset time period can be obtained, and the attribute tag prediction result of the user to be detected can be obtained through one or more of the first probability of matching the first device data with the target attribute tag, the second probability of matching the first advertisement data with the target attribute tag, the first mean value of the first advertisement data in the preset unit time and the first variance of the first advertisement data in the preset unit time. On the basis of user data characteristics such as first equipment data and first advertisement data corresponding to a user to be detected, a first probability of matching the first equipment data with a target attribute label, a second probability of matching the first advertisement data with the target attribute label, and a first mean value and a first variance of the first advertisement data in a preset unit time are increased, and multidimensional user data characteristics are fused to predict a user attribute label, so that the accuracy of predicting the user attribute label can be improved.
The embodiment of the present disclosure provides an apparatus, which includes the above apparatus for predicting a user attribute tag. The device can obtain first device data corresponding to a user to be detected and first advertisement data corresponding to the user to be detected in a preset time period, and obtains an attribute tag prediction result of the user to be detected through one or more of a first probability of matching the first device data with a target attribute tag, a second probability of matching the first advertisement data with the target attribute tag, a first mean value of the first advertisement data in a preset unit time and a first variance of the first advertisement data in the preset unit time. On the basis of user data characteristics such as first equipment data and first advertisement data corresponding to a user to be detected, a first probability of matching the first equipment data with a target attribute label, a second probability of matching the first advertisement data with the target attribute label, and a first mean value and a first variance of the first advertisement data in a preset unit time are increased, and multidimensional user data characteristics are fused to predict a user attribute label, so that the accuracy of predicting the user attribute label can be improved.
Optionally, the device comprises a computer, a smartphone, or the like. Optionally, the device comprises a server or the like.
Optionally, when the device is a server, first device data and first advertisement data corresponding to the user to be tested are obtained through the intelligent terminal. Optionally, in the case that the device is a server, pushing the attribute tag prediction result to the user to be tested through the intelligent terminal.
Optionally, the smart terminal comprises a computer, a smart phone, a tablet, and the like.
Embodiments of the present disclosure provide a computer-readable storage medium storing computer-executable instructions configured to perform the above-described method for predicting user attribute tags.
Embodiments of the present disclosure provide a computer program product comprising a computer program stored on a computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the above-described method for predicting user attribute tags.
The computer-readable storage medium described above may be a transitory computer-readable storage medium or a non-transitory computer-readable storage medium.
The technical solution of the embodiments of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium and includes one or more instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method of the embodiments of the present disclosure. And the aforementioned storage medium may be a non-transitory storage medium comprising: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes, and may also be a transient storage medium.
The above description and drawings sufficiently illustrate embodiments of the disclosure to enable those skilled in the art to practice them. Other embodiments may incorporate structural, logical, electrical, process, and other changes. The examples merely typify possible variations. Individual components and functions are optional unless explicitly required, and the sequence of operations may vary. Portions and features of some embodiments may be included in or substituted for those of others. Furthermore, the words used in the specification are words of description only and are not intended to limit the claims. As used in the description of the embodiments and the claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. Similarly, the term "and/or" as used in this application is meant to encompass any and all possible combinations of one or more of the associated listed. Furthermore, the terms "comprises" and/or "comprising," when used in this application, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. Without further limitation, an element defined by the phrase "comprising an …" does not exclude the presence of other like elements in a process, method or apparatus that comprises the element. In this document, each embodiment may be described with emphasis on differences from other embodiments, and the same and similar parts between the respective embodiments may be referred to each other. For methods, products, etc. of the embodiment disclosures, reference may be made to the description of the method section for relevance if it corresponds to the method section of the embodiment disclosure.
Those of skill in the art would appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software may depend upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosed embodiments. It can be clearly understood by the skilled person that, for convenience and brevity of description, the specific working processes of the system, the apparatus and the unit described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the embodiments disclosed herein, the disclosed methods, products (including but not limited to devices, apparatuses, etc.) may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units may be merely a logical division, and in actual implementation, there may be another division, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form. The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to implement the present embodiment. In addition, functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. In the description corresponding to the flowcharts and block diagrams in the figures, operations or steps corresponding to different blocks may also occur in different orders than disclosed in the description, and sometimes there is no specific order between the different operations or steps. For example, two sequential operations or steps may in fact be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. Each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Claims (10)
1. A method for predicting user attribute tags, comprising:
acquiring first equipment data corresponding to a user to be detected, and acquiring first advertisement data corresponding to the user to be detected within a preset time period; the first advertisement data comprise the times of clicking the advertisement by the user to be tested, the times of clicking the application program corresponding to the advertisement clicked by the user to be tested, and the times of clicking the media corresponding to the advertisement clicked by the user to be tested;
acquiring a first probability of matching the first device data with a target attribute label, acquiring a second probability of matching the first advertisement data with the target attribute label, and acquiring a first mean value and a first variance of the first advertisement data in a preset unit time;
and obtaining an attribute label prediction result of the user to be tested according to one or more of the first equipment data, the first advertisement data, the first probability, the second probability, the first mean value and the first variance.
2. The method of claim 1, wherein obtaining the attribute tag prediction result of the user to be tested according to one or more of the first device data, the first advertisement data, the first probability, the second probability, the first mean, and the first variance comprises:
inputting the first device data, the first advertisement data, the first probability, the second probability, the first mean value and the first variance into a preset attribute label prediction model to obtain an attribute label prediction result of the user to be detected;
the attribute label prediction model is obtained according to second equipment data corresponding to a sample user and second advertisement data corresponding to the sample user; the second advertisement data comprises the times of clicking the advertisement by the sample user, the times of clicking the application program corresponding to the advertisement clicked by the sample user, and the times of clicking the media corresponding to the advertisement clicked by the sample user.
3. The method of claim 2, wherein obtaining the attribute tag prediction model according to the second device data corresponding to the sample user and the second advertisement data corresponding to the sample user comprises:
acquiring a third probability of matching the second device data with the target attribute tag, acquiring a fourth probability of matching the second advertisement data with the target attribute tag, and acquiring a second mean value and a second variance of the second advertisement data in a preset unit time;
and training a logistic regression model by using the second equipment data, the second advertisement data, the third probability, the fourth probability, a second mean value and a second variance to obtain the attribute label prediction model.
4. The method of claim 2, wherein the first device data comprises: the equipment brand and the equipment model corresponding to the user to be tested;
the second device data includes: and the equipment brand and the equipment model corresponding to the sample user.
5. The method of claim 3, wherein obtaining a fourth probability that the second advertisement data matches the target attribute tag comprises:
acquiring a fifth probability that the advertisement clicked by the sample user is matched with the target attribute label;
and calculating by utilizing the fifth probability according to a first preset algorithm to obtain the fourth probability.
6. The method of claim 3, wherein obtaining a fourth probability that the second advertisement data matches the target attribute tag comprises:
acquiring a sixth probability that the application program corresponding to the advertisement clicked by the sample user matches the target attribute label;
and calculating by utilizing the sixth probability according to a second preset algorithm to obtain the fourth probability.
7. The method of claim 3, wherein obtaining a fourth probability that the second advertisement data matches the target attribute tag comprises:
acquiring a seventh probability that the media corresponding to the advertisement clicked by the sample user matches the target attribute label;
and calculating by using the seventh probability according to a third preset algorithm to obtain the fourth probability.
8. The method according to any one of claims 1 to 7, wherein after obtaining the attribute tag prediction result of the user to be tested, the method further comprises:
and pushing the attribute label prediction result to the user to be tested.
9. An apparatus for predicting user attribute tags, comprising a processor and a memory having stored thereon program instructions, wherein the processor is configured to execute the method for predicting user attribute tags of any one of claims 1 to 8 when executing the program instructions.
10. An apparatus, comprising means for predicting user attribute tags as recited in claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110021986.0A CN112700285A (en) | 2021-01-08 | 2021-01-08 | Method, device and equipment for predicting user attribute label |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110021986.0A CN112700285A (en) | 2021-01-08 | 2021-01-08 | Method, device and equipment for predicting user attribute label |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112700285A true CN112700285A (en) | 2021-04-23 |
Family
ID=75513283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110021986.0A Pending CN112700285A (en) | 2021-01-08 | 2021-01-08 | Method, device and equipment for predicting user attribute label |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112700285A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115086692A (en) * | 2021-11-25 | 2022-09-20 | 北京达佳互联信息技术有限公司 | User data display method, device, apparatus, storage medium and program product |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106126597A (en) * | 2016-06-20 | 2016-11-16 | 乐视控股（北京）有限公司 | User property Forecasting Methodology and device |
| CN106803190A (en) * | 2017-01-03 | 2017-06-06 | 北京掌阔移动传媒科技有限公司 | A kind of ad personalization supplying system and method |
| CN106850314A (en) * | 2016-12-20 | 2017-06-13 | 上海掌门科技有限公司 | A kind of method and apparatus for determining user property model and customer attribute information |
| CN111915378A (en) * | 2020-08-17 | 2020-11-10 | 深圳墨世科技有限公司 | User attribute prediction method, device, computer equipment and storage medium |
| CN112132209A (en) * | 2020-09-19 | 2020-12-25 | 北京智能工场科技有限公司 | Attribute prediction method based on bias characteristics |
-
2021
- 2021-01-08 CN CN202110021986.0A patent/CN112700285A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106126597A (en) * | 2016-06-20 | 2016-11-16 | 乐视控股（北京）有限公司 | User property Forecasting Methodology and device |
| CN106850314A (en) * | 2016-12-20 | 2017-06-13 | 上海掌门科技有限公司 | A kind of method and apparatus for determining user property model and customer attribute information |
| CN106803190A (en) * | 2017-01-03 | 2017-06-06 | 北京掌阔移动传媒科技有限公司 | A kind of ad personalization supplying system and method |
| CN111915378A (en) * | 2020-08-17 | 2020-11-10 | 深圳墨世科技有限公司 | User attribute prediction method, device, computer equipment and storage medium |
| CN112132209A (en) * | 2020-09-19 | 2020-12-25 | 北京智能工场科技有限公司 | Attribute prediction method based on bias characteristics |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115086692A (en) * | 2021-11-25 | 2022-09-20 | 北京达佳互联信息技术有限公司 | User data display method, device, apparatus, storage medium and program product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110532451B (en) | Method and device for searching policy text, storage medium and electronic device | |
| CN108805598B (en) | Similarity information determination method, server and computer-readable storage medium | |
| Rahman et al. | Smartphone-based hierarchical crowdsourcing for weed identification | |
| CN109819015B (en) | Information pushing method, device and equipment based on user portrait and storage medium | |
| CN109063108B (en) | Search ranking method and device, computer equipment and storage medium | |
| US10345112B2 (en) | Computer implemented interactive system and method for locating products and services | |
| WO2020192013A1 (en) | Directional advertisement delivery method and apparatus, and device and storage medium | |
| CN112613917A (en) | Information pushing method, device and equipment based on user portrait and storage medium | |
| US9619457B1 (en) | Techniques for automatically identifying salient entities in documents | |
| US20190080352A1 (en) | Segment Extension Based on Lookalike Selection | |
| CN108921587B (en) | Data processing method and device and server | |
| CN111914159A (en) | Information recommendation method and terminal | |
| CN112925973A (en) | Data processing method and device | |
| Ching et al. | A higher-order Markov model for the Newsboy's problem | |
| CN111552767A (en) | Search method, search device and computer equipment | |
| CN112818230A (en) | Content recommendation method and device, electronic equipment and storage medium | |
| US20220114499A1 (en) | System and method for efficiently training intelligible models | |
| CN111311328A (en) | Method and device for determining advertisement click rate of product under advertisement channel | |
| CN114049174A (en) | Method and device for commodity recommendation, electronic equipment and storage medium | |
| CN114528413A (en) | Knowledge graph updating method, system and readable storage medium supported by crowdsourced marking | |
| CN112700285A (en) | Method, device and equipment for predicting user attribute label | |
| CN116127188A (en) | Target feedback value determining method and device, electronic equipment and storage medium | |
| CN110751501A (en) | Commodity shopping guide method, device, equipment and storage medium in new retail mode | |
| CN115730217A (en) | Model training method, material recalling method and device | |
| CN115455151A (en) | AI emotion visual identification method and system and cloud platform |
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 |