CN110012060B - Information pushing method and device of mobile terminal, storage medium and server - Google Patents
Information pushing method and device of mobile terminal, storage medium and server Download PDFInfo
- Publication number
- CN110012060B CN110012060B CN201910112939.XA CN201910112939A CN110012060B CN 110012060 B CN110012060 B CN 110012060B CN 201910112939 A CN201910112939 A CN 201910112939A CN 110012060 B CN110012060 B CN 110012060B
- Authority
- CN
- China
- Prior art keywords
- age
- gender
- prediction model
- data
- mobile terminal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
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
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/20—Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel
- H04W4/23—Services signaling; Auxiliary data signalling, i.e. transmitting data via a non-traffic channel for mobile advertising
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Signal Processing (AREA)
- Finance (AREA)
- Computer Networks & Wireless Communication (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an information pushing method, an information pushing device, a storage medium and a server of a mobile terminal, wherein the information pushing method of the mobile terminal comprises the following steps: acquiring the use data of a target user mobile terminal; inputting the use data into an age and gender prediction model to obtain a gender and age interval of the target user; and pushing information to the target user according to the gender and age interval. According to the method and the device, through predicting the gender age interval of the target user, information can be pushed to the target user in a targeted mode, the information pushing efficiency can be improved, the target user can be prevented from being disturbed, and the user experience is improved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to an information pushing method and device of a mobile terminal, a storage medium and a server.
Background
With the continuous development of networks and intelligent mobile terminals, APP application software of a plurality of mobile terminals emerges. Therefore, in many APP applications, it is very important how to improve the user experience of the APP applications, so that the existing user can continuously use the APP applications, or improve the frequency of using the APP applications by the user, or recommend information such as other applications, news, advertisements, and the like to the user in a targeted manner through the APP applications, so as to reduce the interference of the information that is not of interest to the user. Product improvement, marketing activities (targeted special effect recommendation, advertisement putting and the like) and the like of the APP are important means for improving the number and the use frequency of APP users, but due to the fact that user information is not known, the activities are difficult to develop in a targeted mode. Therefore, it is necessary to provide a method for determining user characteristic information based on the user's usage behavior characteristics of APP, so as to push other information to the user in a targeted manner according to the user's characteristic information.
In the aspect of building a user portrait algorithm, some automatic building methods of user portraits based on clustering are frequently proposed, and the traditional K-Means algorithm is mostly adopted, but the algorithm has the following problems in the clustering process: on one hand, the similarity quality index in the grouped clusters is difficult to ensure, on the other hand, the cluster number and the initial mass center need to be selected manually, and certain randomness exists, but the number of the image clusters cannot be predicted before clustering, so that the whole cluster is unpredictable and unstable; moreover, the k-Means algorithm performs clustering by calculating distances among a plurality of points, and the calculation amount is large and time-consuming.
Disclosure of Invention
The invention provides an information pushing method, an information pushing device, a storage medium and a server of a mobile terminal aiming at the defects of the prior art, and aims to solve the problems of low information pushing efficiency and poor user experience in the prior art.
The information pushing method of the mobile terminal provided by the invention comprises the following steps:
acquiring the use data of a target user mobile terminal;
inputting the use data into an age and gender prediction model to obtain a gender and age interval of the target user;
and pushing information to the target user according to the gender and age interval.
Preferably, the age and sex prediction model is established by the following method:
obtaining the use sample data of a sample user mobile terminal;
labeling the gender age interval corresponding to the use sample data to obtain a sample labeling data set;
generating a distinguishing feature set according to the sample labeling data set;
and (4) inputting the distinguishing features in the distinguishing feature set as a lightgbm algorithm, and training the gender and age prediction model by using the corresponding gender and age interval as an output.
Preferably, the sex age intervals comprise a male first age stage interval, a female first age stage interval, a male second age stage interval, a female second age stage interval, a male third age stage interval, a female third age stage interval, a male fourth age stage interval, a female fourth age stage interval; wherein the ages in the first, second, third and fourth age bracket do not intersect.
Preferably, the distinguishing features in the distinguishing feature set include one or more of a total number of applications per mobile terminal, a usage time of each application, a number of usage times of each application, an average usage time period of each application, a make and model of the mobile terminal, and a twenty-four hour system time feature.
Preferably, the time characteristic comprises a weighted average of total usage time and total usage times of the mobile terminal within twenty-four hours, and the brand and model of the mobile terminal are the brands and models of the first mobile terminals ranked from most to least in number of occurrences.
Preferably, after the pushing of the information to the target user, the method further includes:
acquiring the percentage of the target user viewing the information;
judging whether the percentage is lower than a preset threshold value or not;
if yes, continuing the step of obtaining the use data of the mobile terminal of the target user.
Preferably, the age and sex prediction model includes a first prediction model and a second prediction model, the first prediction model inputs the total number of applications of each mobile terminal, the usage time of each application, the number of times of usage of each application, the average usage duration of each application, and the twenty-four hour system time characteristics in the usage data into the age and sex prediction model, and the second prediction model inputs the brand and model number of the mobile terminal in the usage data into the age and sex prediction model;
after judging whether the percentage is lower than a preset threshold value, the method further comprises the following steps:
if the percentage is lower than the preset threshold value, judging whether the age and gender prediction model is the first prediction model;
if the first prediction model is the first prediction model, obtaining a gender age interval of the target user according to the second prediction model;
and pushing information to the target user according to the gender and age interval.
The invention also provides an information pushing device of the mobile terminal, which comprises:
the acquisition module is used for acquiring the use data of the mobile terminal of the target user;
the prediction module is used for inputting the use data into an age and gender prediction model to obtain a gender and age interval of the target user;
and the pushing module is used for pushing information to the target user according to the gender and age interval.
The present invention also provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the information pushing method of the mobile terminal according to any one of the foregoing items.
The invention also proposes a server comprising:
one or more processors;
a storage device to store one or more programs,
when the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the information push method of the mobile terminal according to any one of the preceding claims.
The invention has the following beneficial effects:
according to the method and the device, the gender age interval to which the target user belongs is predicted through the lightgbm algorithm and the habit of using the mobile terminal by the target user, information such as the APP application program, product advertisements, video programs and news content can be pushed to the target user in a targeted mode, on one hand, the information pushing efficiency can be improved, the viewing rate of the pushed information can be improved, on the other hand, harassment that the target user receives inappropriate APP or product Guangzhou can be reduced, and user experience is improved.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart illustrating an information pushing method of a mobile terminal according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart illustrating an embodiment of building an age and gender prediction model according to the present invention;
fig. 3 is a flowchart illustrating an information pushing method of a mobile terminal according to another embodiment of the present invention;
fig. 4 is a schematic structural diagram of a server according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative only and should not be construed as limiting the invention.
It will be understood by those skilled in the art that, unless otherwise specified, the singular forms "a", "an", "the" and "the" may include the plural forms as well, and the "first" and "second" used herein are only used to distinguish one technical feature from another and are not intended to limit the order, number, etc. of the technical features. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, 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.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
Those skilled in the art will appreciate that a server as used herein includes, but is not limited to, a computer, a network host, a single network server, a collection of network servers, or a cloud of servers. Here, the Cloud is composed of a large number of computers or network servers based on Cloud Computing (Cloud Computing), which is a type of distributed Computing, a super virtual computer consisting of a collection of loosely coupled computers. As used herein, "terminal" and "terminal device" include both devices that are wireless signal receivers, devices that have only wireless signal receivers without transmit capability, and devices that have receive and transmit hardware, devices that have receive and transmit hardware capable of performing two-way communications over a two-way communications link. Such as a smart phone, a tablet computer, a PDA, an MID (Mobile Internet Device) and/or a smart Device with music/video playing function, a set-top box, etc.
The invention provides an information pushing method of a mobile terminal, which is used for improving the pushing efficiency of information and reducing the interference of inappropriate information to a user. The embodiment shown in fig. 1 comprises the following steps:
step S10: acquiring the use data of a target user mobile terminal;
step S20: inputting the use data into an age and gender prediction model to obtain a gender and age interval of the target user;
step S30: and pushing information to the target user according to the gender and age interval.
Wherein each step is as follows:
step S10: and acquiring the use data of the mobile terminal of the target user.
When information needs to be pushed to a target user, because reading habits and using habits of the target user are different, contents or application programs which are interested in the target user are also different. For example, gaming, financial type applications are more easily accepted by male users; social, cosmetic applications are more likely to be liked by female users; the application programs such as cartoons, mini-games and the like are easily accepted by teenagers, and the application programs such as health preserving and broadcasting are more easily accepted by middle-aged and old people. If a health maintenance application program is pushed to teenagers or a cartoon application program is pushed to middle-aged and old people, most of clients may have harassed bad experience due to the fact that the pushed content is not matched with the target client, and the pushing effect is poor. When the current partial APP application program pushes information to a user, contents which are possibly interested by the user can be pushed according to the reading habit or the using habit of the user in the APP application program; for example, in a news APP application, content related to news frequently seen by a user may be pushed to the user according to the news category or news keyword frequently seen by the user. However, for different APP applications, due to the system permission limitation or the permission limitation of the privacy information, it is generally difficult for one APP application to obtain the privacy information in another APP application, so it is difficult to push another APP application according to the reading habit or the using habit of one APP application.
In an existing operating system of a mobile terminal, an APP application program with certain authority can monitor the running condition of each program in the mobile terminal through the operating system, that is, obtain the usage data. The usage data may include one or more of a total number of applications per mobile terminal, a usage time of each application, a number of uses of each application, an average usage duration of each application, a make and model of the mobile terminal, a twenty-four hour system time signature, and the like. When the target user is a user with an unknown gender and age interval, the use data of the mobile terminal is obtained firstly in the step, so that the user can be classified according to the use data, and the information push pertinence is improved.
Step S20: and inputting the use data into an age and gender prediction model to obtain a gender and age interval of the target user.
The specific data content included in the usage data may be different according to the age-gender prediction model. For example, when the age and sex prediction model uses the average usage time of each application program and the brand and model of the mobile terminal as the prediction basis, the usage data may only include the average usage time of each application program and the brand and model of the mobile terminal; of course, in some embodiments, additional features may be included for further reference.
In one embodiment of the present invention, as shown in fig. 2, the age and gender prediction model is built by the following method:
step S01: obtaining the use sample data of a sample user mobile terminal;
step S02: marking the gender age interval corresponding to the use sample data to obtain a sample marking data set;
step S03: generating a distinguishing feature set according to the sample labeling data set;
step S04: and (4) inputting the distinguishing features in the distinguishing feature set as a lightgbm algorithm, and training the gender and age prediction model by using the corresponding gender and age interval as an output.
Step S05: and pushing information to the target user according to the gender and age interval.
The usage sample data in this embodiment may include, but is not limited to, features in the usage data of the target user mobile terminal, so as to train different gender and age prediction models according to specific situations, thereby improving applicability and prediction accuracy of the present invention. The sample user may be a user of known gender and age. For example, when the user is a real-name user, the age and gender of the user can be accurately known from the real-name information of the user. When the gender age interval corresponding to the use sample data is labeled, each use sample data can be in one-to-one correspondence with the corresponding age and gender to obtain a sample labeling data set. And then, selecting required sample data according to the factors such as data integrity, preset distinguishing characteristics and the like in the used sample data to form a distinguishing characteristic set. The distinguishing characteristic set comprises a distinguishing characteristic as an input and a gender and age interval as an output.
In some embodiments, if the obtained usage data of the target user is missing, or cannot be obtained, or is empty, the parameter may be deleted, or a clustering algorithm may be used to calculate the similarity between the missing data and the non-missing data for padding.
And classifying the obtained age and gender of the user to obtain the gender age interval. For example, eight sex age intervals of a young male, a middle male, an old male, a young female, a middle female and an old female can be divided, and no intersection exists between each sex age interval, so that information can be pushed to the eight sex age intervals in a targeted manner.
Thus, in a further embodiment of this embodiment, the gender age intervals include a male first age interval, a female first age interval, a male second age interval, a female second age interval, a male third age interval, a female third age interval, a male fourth age interval, a female fourth age interval; and the first age stage interval, the second age stage interval, the third age stage interval and the fourth age stage interval have no intersection. In conjunction with the foregoing description, the first age group interval may correspond to teenagers, and the age group of teenagers may be set to 7-17 years; the second age range may correspond to young adults, the young adults may be set to 18-39 years old; the third age group interval may correspond to the middle age, and the age group of the middle age may be set to 40-59 years old; the fourth age group section may correspond to an old age, and the age group of the old age may be set to 60 years or more. The numbers referred to in this embodiment include the present number. According to the implementation, most users of the mobile terminal are divided into different intervals according to the age characteristics, and the targeted message pushing is facilitated according to the age characteristics of the users, so that the probability of interference to the users is reduced.
The lightgbm algorithm is a gradient lifting framework based on tree learning, supports high-efficiency parallel training, and can process large-scale data. The lightgbm algorithm mainly includes two algorithms: GOSS algorithm: most samples with small gradients can be eliminated, and information gain is calculated by only using the remaining samples, so that the number of samples is reduced; the EFB algorithm: mutually exclusive features are bundled, thereby reducing sample features. The lightgbm algorithm model can be implemented by referring to the existing open source model and setting corresponding estimator parameters, which are not described herein again.
In one embodiment of the present invention, setting the corresponding estimator parameters may be embodied as follows:
(1) The parameter estimator type is set to GBDT, the estimator number num _ boost _ rounds is 1000, early _ stopping _ u rounds is 600, the parameter target object is set to multi-class multiclass, the class number num _ class is 8 classes and the loss function metric is multi _ logs.
Wherein, boosting _ type is the algorithm or estimator adopted, in lightgbm algorithm, the algorithm or estimator can also include random forest algorithm, random forest algorithm is not sensitive to abnormal value, but GBDT is very sensitive to abnormal value, in the invention, GBDT is more suitable to adopt. num _ boost _ round is an estimator iteration number, a too small value of the number may cause under-fitting of the trained gender and age prediction model, and a too large value of the number may cause resource waste, and may be determined according to the data amount in the distinguishing feature set, and is usually set to be more than 100. The early _ stopping _ rounds is: if one measurement of the once verification data is not improved in the latest early _ stopping _ round, the model stops training to realize accelerated analysis and reduce excessive iteration, and the model can be set according to the data volume in the distinguishing feature set and the iteration times of the estimator; too small of an early stopping rounds may result in under-fitting of the trained gender age prediction model, and too large of a number may result in waste of resources. The parameter target object can be set to multi-classification, and the further refined user classification number num _ class can be 8 classes, so that the prediction result is more accurate. The loss function metric may be selected as a multi-class log-loss function multi _ loglos to make the non-linear multi-class result more accurate, thereby improving the accuracy of the gender age prediction model.
(2) Setting a parameter, category _ feature, to be a category feature, such as a mobile phone brand and model, specified automatically or according to a sample user's feature set; the lightgbm algorithm can be directly trained by the class features to obtain a result of classification according to the class features.
(3) Parameters num _ leaves, min _ data _ in _ leaf and max _ depth are adjusted through a grid search model GridSearchCV to train a lightgbm model.
Wherein, max _ depth: is the maximum depth of the tree, and may be considered to first decrease max _ depth when the gender age prediction model is overfit). num _ leaves is the maximum number of leaves of the tree, each tree is a base classifier in the lightgbm algorithm, and the value of the base classifier is less than or equal to 2 max_depth (ii) a When lightgbm uses the leaf-wise algorithm, the complexity of the tuning tree uses num _ leaves instead of max _ depth, which would result in overfitting if the maximum number of leaves exceeded this value. The value of min _ data _ in _ leaf depends on the number of samples in the sample annotation dataset and num _ leaves, and setting it to a larger value can avoid generating an excessively deep tree; in practical applications, for large data sets, it is typically set to be hundreds or thousands. num _ leaves, min _ data _ in _ leaf and max _ depth are parameters encapsulated in the lightgbm algorithm, and the purpose of parameter adjustment in the step is to optimize the lightgbm algorithm model so as to obtain a gender and age prediction model with a more accurate prediction result. Therefore, the invention also provides the following embodiments:
the training of the gender and age prediction model by using the distinguishing features in the distinguishing feature set as lightgbm algorithm input and the corresponding gender and age interval as output comprises:
establishing a lightgbm algorithm framework, and setting estimator parameters;
and taking the distinguishing features in the distinguishing feature set as input, taking the corresponding gender age interval as output, and training the gender age prediction model by adjusting model parameters of a lightgbm algorithm through grid search GridSearchCV.
A better gender and age prediction model can be obtained through adjusting parameters of GridSearchCV.
The embodiment shown in fig. 2 provides a method for establishing an age and gender prediction model, and a distinguishing feature set is generated by the sample labeling data set, so that inaccurate or incomplete data in the used sample data can be filtered out, and distinguishing features more matched with the prediction result can be obtained; the lightgbm algorithm is more suitable for a big data algorithm, and a more accurate gender and age prediction model can be trained by setting various parameters in the algorithm and adjusting the parameters in the algorithm, so that the rationality and the accuracy of the gender and age prediction model can be improved.
In one embodiment of the present invention for building a gender age prediction model, the distinguishing characteristics in the distinguishing characteristic set may include one or more of a total number of applications per mobile terminal, a usage time of each application, a number of uses of each application, an average usage duration of each application, a brand and model number of the mobile terminal, and a twenty-four hour system time characteristic. In some embodiments, the brand and model of the mobile terminal may be the brands and models of the first mobile terminals ranked from many to few in number of occurrences, and the other signals are set as other, without knowing the specific model of the mobile terminal used by the user less, so as to reduce sample characteristics and improve training efficiency.
Further, the time characteristic includes a total usage time of the mobile terminal within twenty-four hours and a weighted average of the total usage times, and is a comprehensive attribute of the access times and the access times of the device in each hour of the day.
In another embodiment of the present invention, the distinguishing feature set includes category features, the category features including a brand and a model of the mobile terminal; the training of the gender and age prediction model by using the distinguishing features in the distinguishing feature set as lightgbm algorithm input and the corresponding gender and age interval as output comprises:
and training the gender age prediction model by taking the distinguishing category characteristics specified in the distinguishing characteristic set as input and the corresponding gender age interval as output.
The embodiment can directly train the gender and age prediction model according to the brand and the model of the mobile terminal, so that the gender and age interval to which the user belongs can be predicted according to the brand and the model of the mobile terminal used by the user, the training complexity of the gender and age prediction model can be greatly simplified, and the training efficiency is improved.
According to the scheme, a model for predicting the age and sex of the target user can be established through a lightgbm algorithm, targeted APP pushing and information such as advertisements, videos and news pushing are facilitated, the pushing efficiency of enterprises can be improved, disturbance of unsuitable APPs or advertisements on the user is reduced, and user experience is improved.
Step S30: and pushing information to the target user according to the gender and age interval.
When the present invention obtains the gender age zone to which the target user belongs according to step S20, information matching the gender age zone can be pushed to the user according to the gender age zone. For example, a princess game, a child apparel accessory product, and the like are pushed to a first age group of women, a competition game, a ball game program, and the like are pushed to a second age group of men, a parent application program, a beauty makeup product, and the like are pushed to women in a third age group of women, and a military program, a health care application program, and the like are pushed to a fourth age group of men.
According to the method and the device, the gender age interval to which the target user belongs is predicted through the lightgbm algorithm and the habit of using the mobile terminal by the target user, information such as the APP application program, product advertisements, video programs and news content can be pushed to the target user in a targeted mode, on one hand, the information pushing efficiency can be improved, the viewing rate of the pushed information can be improved, on the other hand, harassment that the target user receives inappropriate APP or product Guangzhou can be reduced, and user experience is improved.
The usage data of the target user mobile terminal and the usage sample data of the sample user mobile terminal may be obtained in a backend embedded manner, for example, device data of the mobile terminal, an application installation list on each device, an opening/closing timestamp of each application on each device, model data (i.e., a brand and a model of each device), and APP application data (i.e., category information of each application, etc.) are obtained.
In one embodiment of the present invention, gender may be represented by two numbers, 1 and 2, representing male and female, respectively; the age groups adopt four numbers from 0 to 3, which respectively represent different age groups, and the larger the number is, the larger the corresponding age is. For example, one may define in sequence: 10 is a young male, 11 is a young male, 12 is a middle-aged male, 13 is an old male, 20 is a young female, 21 is a young female, 22 is a middle-aged female, and 23 is an old female. In the present invention, one mobile terminal corresponds to one target user or sample user and to one gender age zone.
In an embodiment of the information push method of another mobile terminal according to the present invention, as shown in fig. 3, after pushing information to the target user, the method further includes:
step S40: acquiring the percentage of the target user viewing the information;
step S50: judging whether the percentage is lower than a preset threshold value or not;
step S60: if yes, continuing the step of obtaining the use data of the mobile terminal of the target user.
The embodiment can acquire the feedback information of the user after pushing the information to the target user so as to verify the matching of information pushing; and when the viewing rate of the pushed information is not improved, the use data of the mobile terminal of the target user is obtained again, so that the gender and age interval to which the user belongs is predicted again according to the latest use data, the purpose of correcting the gender and age interval to which the user belongs is achieved, and the matching property of information pushing is improved.
When the gender and age interval to which the user belongs is predicted again to be the same as the previously predicted result, if information matched with the gender and age interval to which the user belongs is continuously pushed to the user, the viewing rate of the user may still not be increased. Therefore, the invention also provides another information pushing method:
the age and gender prediction model comprises a first prediction model and a second prediction model, the first prediction model inputs the total number of applications of each mobile terminal in the use data, the use time of each application, the use times of each application, the average use duration of each application and the twenty-four hour system time characteristics into the age and gender prediction model, and the second prediction model inputs the brand and model of the mobile terminal in the use data into the age and gender prediction model;
after the step of judging whether the percentage is lower than a preset threshold value, the method further comprises the following steps:
if the percentage is lower than the preset threshold value, judging whether the age and gender prediction model is the first prediction model;
if the first prediction model is the first prediction model, obtaining a gender and age interval of the target user according to the second prediction model;
and pushing information to the target user according to the gender and age interval.
The input characteristics of the first prediction model and the second prediction model are different, and the gender age interval of the target user can be obtained from another angle; according to the embodiment, when the prediction result of the first prediction model is not ideal, the gender age interval of the target user can be obtained again by adopting the second prediction model; when the age and the gender of the target user predicted by the second prediction model are in another gender and age interval, information matched with the another gender and age interval can be pushed to the user, so that the viewing rate of the pushed information is improved, and the pushing efficiency is improved.
The invention also provides an information pushing device of the mobile terminal, which comprises:
the acquisition module is used for acquiring the use data of the mobile terminal of the target user;
the prediction module is used for inputting the use data into an age and gender prediction model to obtain a gender and age interval of the target user;
and the pushing module is used for pushing information to the target user according to the gender and age interval.
An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the information pushing method of the mobile terminal described in any one of the above. The storage medium includes, but is not limited to, any type of disk including floppy disks, hard disks, optical disks, CD-ROMs, and magneto-optical disks, ROMs (Read-Only memories), RAMs (Random AcceSS memories), EPROMs (EraSable Programmable Read-Only memories), EEPROMs (Electrically EraSable Programmable Read-Only memories), flash memories, magnetic cards, or optical cards. That is, a storage medium includes any medium that stores or transmits information in a form readable by a device (e.g., a computer). Which may be a read-only memory, magnetic or optical disk, or the like.
An embodiment of the present invention further provides a server, where the server includes:
one or more processors;
a storage device for storing one or more programs,
when the one or more programs are executed by the one or more processors, the one or more processors implement the information push method of the mobile terminal.
Fig. 4 is a schematic structural diagram of the server of the present invention, which includes a processor 320, a storage device 330, an input unit 340, a display unit 350, and the like. Those skilled in the art will appreciate that the structural elements shown in fig. 4 do not constitute a limitation to all servers, and may include more or fewer components than those shown, or some of the components may be combined. The storage 330 may be used to store the application 310 and various functional modules, and the processor 320 executes the application 310 stored in the storage 330, thereby performing various functional applications of the device and data processing. The storage 330 may be an internal memory or an external memory, or include both internal and external memories. The internal memory may include read-only memory, programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), flash memory, or random access memory. The external memory may include a hard disk, a floppy disk, a ZIP disk, a usb disk, a magnetic tape, etc. The disclosed memory devices include, but are not limited to, these types of memory devices. The disclosed storage device 330 is provided by way of example only and not by way of limitation.
The input unit 340 is used to receive input of a signal, and to receive user attribute information of a target user on a first statistical date and access information to a specified target. The input unit 340 may include a touch panel and other input devices. The touch panel can collect touch operations of a user on or near the touch panel (for example, operations of the user on or near the touch panel by using any suitable object or accessory such as a finger, a stylus and the like) and drive the corresponding connecting device according to a preset program; other input devices may include, but are not limited to, one or more of a physical keyboard, function keys (e.g., play control keys, switch keys, etc.), a trackball, a mouse, a joystick, and the like. The display unit 350 may be used to display information input by a user or information provided to the user and various menus of the computer device. The display unit 350 may take the form of a liquid crystal display, an organic light emitting diode, or the like. The processor 320 is a control center of the computer device, connects various parts of the entire computer using various interfaces and lines, and performs various functions and processes data by operating or executing software programs and/or modules stored in the storage device 330 and calling data stored in the storage device.
In one embodiment, the server includes one or more processors 320, and one or more storage devices 330, one or more applications 310, wherein the one or more applications 310 are stored in the storage device 330 and configured to be executed by the one or more processors 320, and the one or more applications 310 are configured to perform the attrition user-alerting method described in the above embodiments.
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and may be performed in other orders unless explicitly stated herein. Moreover, at least a portion of the steps in the flow chart of the figure may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
It should be understood that each functional unit in the embodiments of the present invention may be integrated into one processing module, each unit may exist alone physically, or two or more units may be integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode.
The foregoing is only a partial embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (9)
1. An information push method of a mobile terminal is characterized by comprising the following steps:
acquiring the use data of a target user mobile terminal;
inputting the use data into an age and gender prediction model to obtain a gender and age interval of a target user, wherein the age and gender prediction model is constructed by a distinguishing feature set determined by a sample labeling data set, and the sample labeling data set is determined by labeling each use sample data with a corresponding gender and age interval; the gender age intervals comprise a plurality of age stage intervals respectively corresponding to different genders; the age and gender prediction model comprises a first prediction model and a second prediction model; the first predictive model makes a first prediction of a gender age interval for a first data of the usage data; the second prediction model makes a second prediction of a gender age interval for a second data of the usage data;
according to the gender and age interval, information is pushed to a target user, and the method comprises the following steps: if the percentage of the target user viewing the information is lower than a preset threshold value after the information is pushed to the target user based on the first predicted result, pushing the information to the target user based on the second predicted result;
the age and gender prediction model is established by the following method: obtaining the use sample data of a sample user mobile terminal; marking the gender age interval corresponding to the use sample data to obtain a sample marking data set; generating a distinguishing feature set according to the sample labeling data set; and (4) inputting the distinguishing features in the distinguishing feature set as a lightgbm algorithm, and training the age and gender prediction model by taking the corresponding gender and age interval as output.
2. The method of claim 1, wherein the gender age intervals comprise a male first age interval, a female first age interval, a male second age interval, a female second age interval, a male third age interval, a female third age interval, a male fourth age interval, a female fourth age interval; wherein the ages in the first, second, third and fourth age bracket do not intersect.
3. The method of claim 1, wherein the distinguishing characteristics in the distinguishing characteristic set comprise one or more of a total number of applications per mobile terminal, a usage time of each application, a number of uses of each application, an average usage duration of each application, a make and model of the mobile terminal, and a twenty-four hour system time characteristic.
4. The method of claim 3, wherein the time characteristic comprises a weighted average of total usage time and total usage times of the mobile terminal within twenty-four hours, and wherein the brand and model of the mobile terminal is the brand and model of the first mobile terminals ranked from many to few in occurrence number.
5. The method of claim 1, wherein after pushing information to the target user, further comprising:
acquiring the percentage of the target user viewing the information;
judging whether the percentage is lower than a preset threshold value or not;
if yes, continuing the step of obtaining the use data of the mobile terminal of the target user.
6. The method according to claim 5, wherein the first prediction model inputs the total number of applications, the usage time of each application, the number of times of usage of each application, the average usage duration of each application, and the twenty-four hour system time characteristics of each mobile terminal in the usage data into the age-gender prediction model, and the second prediction model inputs the brand and model of the mobile terminal in the usage data into the age-gender prediction model;
after judging whether the percentage is lower than a preset threshold value, the method further comprises the following steps:
if the percentage is lower than the preset threshold value, judging whether the age and gender prediction model is the first prediction model;
if the first prediction model is the first prediction model, obtaining a gender age interval of the target user according to the second prediction model;
and pushing information to the target user according to the gender and age interval.
7. An information pushing device of a mobile terminal, comprising:
the acquisition module is used for acquiring the use data of the mobile terminal of the target user;
the prediction module is used for inputting the use data into an age and gender prediction model to obtain a gender and age interval of the target user, wherein the age and gender prediction model is constructed by a distinguishing feature set determined by a sample marking data set, and the sample marking data set is determined by marking each use sample data with the corresponding gender and age interval; the gender age intervals comprise a plurality of age stage intervals respectively corresponding to different genders; the age and gender prediction model comprises a first prediction model and a second prediction model; the first predictive model makes a first prediction of a gender age interval for a first data of the usage data; the second predictive model makes a second prediction of gender-age intervals for a second data of the usage data;
the pushing module is used for pushing information to the target user according to the gender and age interval, and comprises: if the percentage of the target user viewing the information is lower than a preset threshold value after the information is pushed to the target user based on the first predicted result, pushing the information to the target user based on the second predicted result;
the age and gender prediction model is established by the following method: obtaining the use sample data of a sample user mobile terminal; labeling the gender age interval corresponding to the use sample data to obtain a sample labeling data set; generating a distinguishing feature set according to the sample labeling data set; and (4) inputting the distinguishing features in the distinguishing feature set as a lightgbm algorithm, and training the age and gender prediction model by taking the corresponding gender and age interval as output.
8. A computer-readable storage medium, on which a computer program is stored, wherein the program, when executed by a processor, implements an information pushing method for a mobile terminal according to any one of claims 1 to 6.
9. A server, characterized in that the server comprises:
one or more processors;
storage means for storing one or more programs, which when executed by the one or more processors, cause the one or more processors to implement the information push method of the mobile terminal according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910112939.XA CN110012060B (en) | 2019-02-13 | 2019-02-13 | Information pushing method and device of mobile terminal, storage medium and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910112939.XA CN110012060B (en) | 2019-02-13 | 2019-02-13 | Information pushing method and device of mobile terminal, storage medium and server |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110012060A CN110012060A (en) | 2019-07-12 |
| CN110012060B true CN110012060B (en) | 2023-04-18 |
Family
ID=67165747
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910112939.XA Active CN110012060B (en) | 2019-02-13 | 2019-02-13 | Information pushing method and device of mobile terminal, storage medium and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110012060B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112488742A (en) * | 2019-09-12 | 2021-03-12 | 北京三星通信技术研究有限公司 | User attribute information prediction method and device, electronic equipment and storage medium |
| CN112702375B (en) * | 2019-10-23 | 2022-06-24 | 腾讯科技（深圳）有限公司 | Information pushing method and device, computer equipment and storage medium |
| CN111242723B (en) * | 2020-01-02 | 2020-09-15 | 平安科技（深圳）有限公司 | User child and child condition judgment method, server and computer readable storage medium |
| CN112306517A (en) * | 2020-03-30 | 2021-02-02 | 尼尔森网联媒介数据服务有限公司 | Method, device, storage medium and electronic equipment for processing data of application program |
| CN111601169B (en) * | 2020-05-21 | 2022-01-14 | 广州欢网科技有限责任公司 | Method, device, storage medium and server for determining age stage of family member |
| CN112084402A (en) * | 2020-08-24 | 2020-12-15 | 浙江云合数据科技有限责任公司 | Method for predicting user attribute by analyzing application program use data |
| CN112765479B (en) * | 2021-04-06 | 2021-09-14 | 北京神州泰岳智能数据技术有限公司 | Information recommendation method and device, electronic equipment and readable storage medium |
| CN113158944B (en) * | 2021-04-29 | 2023-07-18 | 重庆锐云科技有限公司 | Physical examination queuing method, system, equipment and storage medium based on face recognition |
| CN113435922B (en) * | 2021-06-10 | 2024-04-26 | 郑州阿帕斯数云信息科技有限公司 | Advertisement data pushing method, device, equipment and storage medium |
| CN115689626B (en) * | 2022-10-31 | 2024-03-01 | 荣耀终端有限公司 | User attribute determining method of terminal equipment and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102541590A (en) * | 2011-12-14 | 2012-07-04 | 奇智软件（北京）有限公司 | Software recommending method and system |
| CN104573048A (en) * | 2015-01-20 | 2015-04-29 | 电子科技大学 | User basic attribute predicting method based on flow data of smart phone |
| CN106251174A (en) * | 2016-07-26 | 2016-12-21 | 北京小米移动软件有限公司 | Information recommendation method 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 |
| CN107463704A (en) * | 2017-08-16 | 2017-12-12 | 北京百度网讯科技有限公司 | Searching method and device based on artificial intelligence |
| CN107798567A (en) * | 2017-11-21 | 2018-03-13 | 成都高德唯斯科技股份有限公司 | Brand message method for pushing, device and electronic equipment |
| CN109255651A (en) * | 2018-08-22 | 2019-01-22 | 重庆邮电大学 | A kind of search advertisements conversion intelligent Forecasting based on big data |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8582807B2 (en) * | 2010-03-15 | 2013-11-12 | Nec Laboratories America, Inc. | Systems and methods for determining personal characteristics |
| US10389828B2 (en) * | 2017-04-28 | 2019-08-20 | Sap Se | Enhanced data collection and analysis facility |
-
2019
- 2019-02-13 CN CN201910112939.XA patent/CN110012060B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102541590A (en) * | 2011-12-14 | 2012-07-04 | 奇智软件（北京）有限公司 | Software recommending method and system |
| CN104573048A (en) * | 2015-01-20 | 2015-04-29 | 电子科技大学 | User basic attribute predicting method based on flow data of smart phone |
| CN106251174A (en) * | 2016-07-26 | 2016-12-21 | 北京小米移动软件有限公司 | Information recommendation method 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 |
| CN107463704A (en) * | 2017-08-16 | 2017-12-12 | 北京百度网讯科技有限公司 | Searching method and device based on artificial intelligence |
| CN107798567A (en) * | 2017-11-21 | 2018-03-13 | 成都高德唯斯科技股份有限公司 | Brand message method for pushing, device and electronic equipment |
| CN109255651A (en) * | 2018-08-22 | 2019-01-22 | 重庆邮电大学 | A kind of search advertisements conversion intelligent Forecasting based on big data |
Non-Patent Citations (2)
| Title |
|---|
| Predicting Age & Gender of Mobile Users at Scale - A Distributed Machine Learning Approach;Kajanan Sangaralingam 等;《IEEE》;20181231;全文 * |
| 基于Android日志的APP用户行为分析研究与实现;李冰;《中国优秀硕士学位论文全文数据库 信息科技辑》;20181015;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110012060A (en) | 2019-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110012060B (en) | Information pushing method and device of mobile terminal, storage medium and server | |
| US10958748B2 (en) | Resource push method and apparatus | |
| CN110543598B (en) | Information recommendation method and device and terminal | |
| CN110909205B (en) | Video cover determination method and device, electronic equipment and readable storage medium | |
| US10152479B1 (en) | Selecting representative media items based on match information | |
| WO2021135562A1 (en) | Feature validity evaluation method and apparatus, and electronic device and storage medium | |
| CN109033408B (en) | Information pushing method and device, computer readable storage medium and electronic equipment | |
| CN107798027B (en) | Information popularity prediction method, information recommendation method and device | |
| CN110019794B (en) | Text resource classification method and device, storage medium and electronic device | |
| CN107613022A (en) | Content delivery method, device and computer equipment | |
| JP6547070B2 (en) | Method, device and computer storage medium for push information coarse selection sorting | |
| WO2023109059A1 (en) | Method for determining fusion parameter, information recommendation method, and model training method | |
| KR20160124087A (en) | Determining an active persona of a user device | |
| US20190294259A1 (en) | Sticker recommendation method and apparatus | |
| US10511681B2 (en) | Establishing and utilizing behavioral data thresholds for deep learning and other models to identify users across digital space | |
| CN110413867B (en) | Method and system for content recommendation | |
| CN108304432B (en) | Information push processing method, information push processing device and storage medium | |
| WO2016196526A1 (en) | Viewport-based implicit feedback | |
| CN110347781B (en) | Article reverse arrangement method, article recommendation method, device, equipment and storage medium | |
| WO2020258773A1 (en) | Method, apparatus, and device for determining pushing user group, and storage medium | |
| CN112749330B (en) | Information pushing method, device, computer equipment and storage medium | |
| JP2014215685A (en) | Recommendation server and recommendation content determination method | |
| CN112052399A (en) | Data processing method and device and computer readable storage medium | |
| CN115239356A (en) | Recommended content management method and related device | |
| CN111506617B (en) | Method and device for generating label based on user data |
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 |