WO2017124919A1 - Pushing method and apparatus for application program, and server - Google Patents

Abstract

Disclosed are a pushing method and apparatus for an application program, and a server. The method comprises: receiving a wireless network list sent by a terminal device, wherein the wireless network list records a plurality of wireless network identifiers scanned by the terminal device; determining a real-time geographical location of the terminal device according to the plurality of wireless network identifiers; querying and obtaining at least one application program corresponding to the real-time geographical location; and pushing the at least one application program to the terminal device. According to the embodiments of the present invention, the mode of recommending an application program can be improved.

Description

Application push method, device and server Technical field

The present invention relates to the field of computers, and in particular to a method, an apparatus, and a server for pushing an application.

Background technique

Merchants tend to push apps to their mobile phones. There are three main ways to recommend a common application.

(1) Based on the recommendation of the leaderboard. In this way, based on the download amount of the APP, all the APPs are sorted, and then the top ranked APP is recommended to all users.

(2) Population-based recommendations. In this way, the entire user community is first divided into multiple human bodies according to certain criteria. For each group, recommend an app that fits the group. In the crowd-based APP recommendation, the app unique to the group is recommended for each type of group.

(3) Based on personalized recommendations. In this way, starting from the online behavior data of the user, analyzing the personalized needs (interests) of the user, and then recommending the APP that meets the personalized needs of the current user according to the user's interests and preferences.

It should be noted that the recommendation schemes of the above three applications are not comprehensive enough to pay attention to the dimension of the user requirements, and the success rate of the application recommendation is low.

In the related art, only the user's interests are concerned when pushing the application to the user, which results in a low success rate of the application recommendation. For this problem, no effective solution has been proposed yet.

In addition, with the development of technology, the function of the terminal is increasingly powerful. In order to implement various functions of the terminal, it is usually necessary to install various types of applications (APP, application) on the terminal, such as office applications, casual applications, and professional tool applications.

Currently, users typically use a dedicated software download site to download applications. This type of website categorizes the application and counts the downloads of each application in each category. Next, the website sorts the applications according to the amount of downloads and recommends the top ranked applications to the user. That is to say, in the prior art, an application is recommended for the user based on the download amount.

However, the inventors found in the research process of the present application that the prior art solution for recommending an application based on the download ranking list would cause each user to be recommended for the same application. However, because different users have different needs, this often results in users not needing the recommended application. Therefore, existing application recommendation methods often fail to meet the needs of users. In this case, when the user obtains the application required by himself, he often needs to enter the software download website to search, which takes a lot of time and effort, thereby increasing the burden on the user and resulting in poor user experience.

Summary of the invention

An object of an embodiment of the present invention is to provide a method, an apparatus, and a server for pushing an application.

According to one embodiment of the invention, the manner in which an application is pushed to a user can be improved.

According to an aspect of the embodiments of the present invention, a method for pushing an application program includes: receiving a wireless network list sent by a terminal device, where the wireless network list records a plurality of wireless network identifiers scanned by the terminal device; The wireless network identifier determines a real-time geographic location of the terminal device; the query obtains at least one application corresponding to the real-time geographic location; and pushes at least one application to the terminal device.

According to another aspect of the present invention, a push device for an application is provided, including: a first receiving unit, configured to receive a wireless network list sent by the terminal device, where the wireless network list records the terminal device scanning a plurality of wireless network identifiers; a determining unit, configured to determine a real-time geographic location of the terminal device according to the plurality of wireless network identifiers; a query unit configured to query at least one application program corresponding to the real-time geographic location; and a pushing unit configured to The terminal device pushes at least one application.

According to another aspect of the present invention, a server is provided, including: a receiver, configured to receive a wireless network list sent by the terminal device, where the wireless network list records multiple wireless network identifiers scanned by the terminal device a processor, configured to determine a real-time geographic location of the terminal device according to the plurality of wireless network identifiers, and query the at least one application corresponding to the real-time geographic location; and the transmitter, configured to push the at least one application to the terminal device.

In the embodiment of the present invention, the wireless network list sent by the receiving terminal device is used, wherein the wireless network list records the plurality of wireless network identifiers scanned by the terminal device; and determines the real-time geographic location of the terminal device according to the plurality of wireless network identifiers; Obtaining at least one application corresponding to the real-time geographic location; pushing at least one application to the terminal device.

According to one embodiment of the invention, the manner in which an application is pushed to a user can be improved.

In addition, in order to overcome the problems in the related art, the present disclosure also provides an application recommendation method and corresponding apparatus.

According to another aspect of an embodiment of the present disclosure, an application recommendation method is provided, including: monitoring a target event for each application; generating a monitoring record according to the monitoring result, wherein the monitoring record includes: a user identifier, each a target event occurrence time of the type and a name of the application targeted by the target event; transmitting the monitoring record to the server; receiving an application recommendation list generated by the server according to the monitoring record, wherein the server receives After the monitoring record, the user needs prediction values for the respective applications are calculated according to the information included in the monitoring records, and the application recommendation list is generated according to the user's demand prediction values for the respective applications.

Preferably, the transmitting the monitoring record to the server comprises: detecting whether the terminal is connected to the wireless network; and if the terminal is connected to the wireless network, transmitting the monitoring record to the server by using the wireless network.

According to another aspect of the embodiments of the present disclosure, an application recommendation method is provided, including: acquiring a monitoring record transmitted by each terminal, where the monitoring record includes: a user identifier, each type of target event occurrence time, and a target event. The name of the application; the predicted value of the user's demand for each application is calculated according to the information included in the monitoring record; and the application recommendation list is generated according to the predicted value of the user's demand for each application, and The application recommendation list is transmitted to the target terminal.

Preferably, the calculating, according to each piece of information included in the monitoring record, a predicted value of the user's demand for each application, including: according to the target event occurrence time and the target event for the application included in the monitoring record The name is used to count the number of occurrences of the target event corresponding to each application in the preset time period corresponding to the target terminal corresponding to each user identifier; and calculate the user to each application according to the number of occurrences of the target event corresponding to each application. Demand forecast.

Preferably, the calculating the predicted value of the demand for each application by the user according to the number of occurrences of the target event corresponding to the respective applications, including: calculating the user according to the number of occurrences of the target event corresponding to the respective applications a demand value for each application; a user requirement matrix is created according to the user's demand value for each application, wherein the user requirement matrix includes a demand value of each user for each application; and the user demand matrix Performing decomposition to obtain a user feature matrix for characterizing a degree of preference of the user for features included in the application, and an application feature matrix, wherein the application feature matrix is used to represent each application and the feature The degree of similarity; calculating the predicted demand value of the user for each application according to the user feature matrix and the application feature matrix.

Preferably, if the target event comprises: a startup event, and/or a download event, and/or a search event, and/or a browsing event, the user pairs are calculated according to the number of occurrences of the target event corresponding to the respective applications. When applying the demand value of the application, the following formula is used:

R _ij =t ₁ ×start _ij +t ₂ ×down _ij +t ₃ ×search _ij +t ₄ ×view _ij ;

Where R _ij represents the demand value of user i for application j, start _ij represents the number of times user i starts application j, down _ij represents the number of times user i downloads application j, and search _ij represents the number of times user i searches for application j , view _ij represents the number of times user i browses the application j, and t ₁ , t ₂ , t _{3 ,} and t ₄ are preset constants.

Preferably, when calculating the predicted demand value of the user for each application according to the user feature matrix and the application feature matrix, the following formula is adopted:

Where P _ij represents the predicted value of the demand of the user i for the application j, U _ik is the element in the user feature matrix, represents the degree of preference of the user i for the feature of the k category, and V _kj is the element in the application feature matrix, indicating the application The degree of similarity between the program j and the features of the k category, and N indicates the number of categories of features.

According to another aspect of the embodiments of the present disclosure, an application recommendation apparatus is provided, including: a listening module, configured to monitor a target event for each application; and a record generating module, configured to generate a monitoring record according to the monitoring result, wherein The monitoring record includes: a user identifier, a target event occurrence time of each type, and a name of an application for which the target event is targeted; a transmission module, configured to transmit the monitoring record to the server; and a receiving module, configured to receive the server Returning an application recommendation list generated according to the monitoring record, wherein, after receiving the monitoring record, the server calculates a predicted value of the user's demand for each application according to each piece of information included in the monitoring record, and The application recommendation list is generated according to the user's demand forecast value for each application.

Preferably, the transmission module includes: a detecting unit, configured to detect whether the terminal is connected to the wireless network, and a wireless transmission unit, configured to transmit the monitoring record to the server by using the wireless network if the terminal is connected to the wireless network.

According to another aspect of the embodiments of the present disclosure, an application recommendation apparatus is provided, including: an acquisition module, configured to acquire a monitoring record transmitted by each terminal, where the monitoring record includes: a user identifier, and various types of target events. The time of occurrence and the name of the application to which the target event is directed; a calculation module, configured to calculate a predicted value of the user's demand for each application according to each piece of information included in the monitoring record; and a recommendation module for using the user pair The demand prediction value of each application generates an application recommendation list and transmits the application recommendation list to the target terminal.

Preferably, the calculation module includes: a statistical sub-module, configured to count, according to the target event occurrence time and the name of the application for the target event, the target terminal corresponding to each user identifier is in a preset time period. The number of times the target event is generated by each application; the calculation sub-module is configured to calculate a predicted value of the demand of each user for each application according to the number of occurrences of the target event corresponding to the respective applications.

Preferably, the calculation sub-module includes: a demand value calculation unit, configured to calculate a user's demand value for each application according to the number of occurrences of the target event corresponding to the respective applications; and a matrix creation unit, configured to A user requirement matrix is created by the user, and the user requirement matrix includes a requirement value of each user for each application; a matrix decomposition unit is configured to decompose the user requirement matrix to obtain a user. a feature matrix and an application feature matrix, wherein the user feature matrix is used to characterize a degree of preference of the user for features included in the application, the application feature matrix being used to characterize the degree of similarity of each application to the feature; prediction And a value calculation unit, configured to calculate, according to the user feature matrix and the application feature matrix, a predicted predicted value of the user for each application.

Preferably, if the target event comprises: a startup event, and/or a download event, and/or a search event, and/or a browsing event, the demand value calculation unit occurs in accordance with a target event corresponding to the respective application The number of times, when calculating the user's demand value for each application, the following formula is used:

R _ij =t ₁ ×start _ij +t ₂ ×down _ij +t ₃ ×search _ij +t ₄ ×view _ij ;

Where R _ij represents the demand value of user i for application j, start _ij represents the number of times user i starts application j, down _ij represents the number of times user i downloads application j, and search _ij represents the number of times user i searches for application j , view _ij represents the number of times user i browses the application j, and t ₁ , t ₂ , t _{3 ,} and t ₄ are preset constants.

The predicted value calculation unit calculates the predicted value of the demand for each application according to the user feature matrix and the application feature matrix, and adopts the following formula:

Where P _ij represents the predicted value of the demand of the user i for the application j, U _ik is the element in the user feature matrix, represents the degree of preference of the user i for the feature of the k category, and V _kj is the element in the application feature matrix, indicating the application The degree of similarity between the program j and the features of the k category, and N indicates the number of categories of features.

According to an embodiment of the present invention, an application recommendation list can be obtained, the application recommendation list being generated by the server according to a user's demand forecast value for the application.

According to an embodiment of the present invention, an application recommended through an application list can meet user requirements, thereby more accurately predicting a user's demand for an application.

According to one embodiment of the present invention, it is possible to reduce the time and effort that a user takes to acquire a desired application.

The above general description and the following detailed description are intended to be illustrative and not restrictive.

DRAWINGS

The accompanying drawings, which are incorporated in the specification of FIG The drawings described herein are intended to provide a further understanding of the invention, and are intended to be a part of the invention.

1 is a block diagram showing the hardware structure of a push method of an application according to an exemplary embodiment of the present invention;

2 is a flowchart of a push method of an application according to an exemplary embodiment of the present invention;

FIG. 3 is a schematic diagram of an alternate application push method according to an exemplary embodiment of the present invention; FIG.

4 is a schematic diagram of an alternate application push method in accordance with an exemplary embodiment of the present invention;

FIG. 5 is a schematic diagram of a push device of an application according to an exemplary embodiment of the present invention; and

6 is a schematic diagram of a server in accordance with an exemplary embodiment of the present invention;

FIG. 7 is a schematic diagram of a workflow of an application recommendation method according to an exemplary embodiment;

FIG. 8 is a schematic diagram of a workflow of still another application recommendation method according to an exemplary embodiment;

FIG. 9 is a schematic diagram of a workflow for calculating a demand prediction value in an application recommendation method according to an exemplary embodiment;

FIG. 10 is a schematic structural diagram of an application recommendation apparatus according to an exemplary embodiment;

FIG. 11 is a schematic structural diagram of still another application recommendation method according to an exemplary embodiment.

detailed description

The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is apparent that the described embodiments are merely a part of the embodiments of the invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of the present invention.

It is to be understood that the terms "first", "second" and the like in the specification and claims of the present invention are used to distinguish similar objects, and are not necessarily used to describe a particular order or order. It is to be understood that the data so used may be interchanged where appropriate, so that the embodiments of the invention described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "comprises" and "comprises" and "the" and "the" are intended to cover a non-exclusive inclusion, for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to Those steps or units may include other steps or units not explicitly listed or inherent to such processes, methods, products or devices.

Example 1

According to an embodiment of the present invention, an embodiment of a push method of an application is also provided. It should be noted that the steps illustrated in the flowchart of the figures may be performed in a computer system such as a set of computer executable instructions. Although the order is schematically illustrated in the flowcharts, in some cases, the steps shown or described may be performed in a different order than the ones described herein.

The method embodiment provided in Embodiment 1 of the present application can be executed in a mobile terminal, a computer terminal or the like. Taking a computer terminal as an example, FIG. 1 is a block diagram showing the hardware structure of a computer terminal of an application program pushing method according to an embodiment of the present invention. As shown in FIG. 1, computer terminal 10 may include one or more (only one shown) processor 102 (processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) A memory 104 for storing data, and a transmission device 106 for communication functions. It will be understood by those skilled in the art that the structure shown in FIG. 1 is merely illustrative and does not limit the structure of the above electronic device. For example, computer terminal 10 may also include more or fewer components than those shown in FIG. 1, or have a different configuration than that shown in FIG.

The memory 104 can be used to store software programs and modules of the application software, such as program instructions/modules corresponding to the push method of the application program in the embodiment of the present invention. The processor 102 runs software stored in the memory 104 Programs and modules to perform various functional applications and data processing, that is, to implement the vulnerability detection method of the above application. Memory 104 may include high speed random access memory, and may also include non-volatile memory such as one or more magnetic storage devices, flash memory, or other non-volatile solid state memory. In some examples, memory 104 may further include memory remotely located relative to processor 102, which may be coupled to computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

Transmission device 106 is for receiving or transmitting data via a network. The network specific examples described above may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC) that can be connected to other network devices through a base station to communicate with the Internet. In one example, the transmission device 106 can be a Radio Frequency (RF) module for communicating with the Internet wirelessly.

In the above operating environment, the present invention provides a push method of an application as shown in FIG. 2. 2 is a flow chart of a method for pushing an application according to a first embodiment of the present invention. As shown in FIG. 2, the method can include the following steps.

Step S22: Receive a wireless network list sent by the terminal device, where the wireless network list records multiple wireless network identifiers scanned by the terminal device.

In the above step S22, the solution may use a server to receive a list of wireless networks sent by the terminal device. Preferably, the wireless network list may be a WIFI list. When there is a WIFI hotspot around the terminal device, a WIFI list is automatically generated in the terminal device. In the WIFI list, each wireless network identity can be recorded. The terminal device can upload the above WIFI list to the server at any time.

Optionally, a certain time interval may be set in the terminal device to periodically read the current wifi scan list of the terminal device, and then upload the wifi scan list in real time in the format of the smart device unique identifier/wifi scan list/time.

The following is an example of pushing an APP to a user user. When the user user is in a commercial area, the user's terminal device automatically detects the surrounding wifi signal and generates a wifi list. The identity of all wifi around the terminal device of the user user can be recorded in the wifi list. An example of a wifi list is shown in Table 1 below. It should be noted here that only the wifi network around the user user can be displayed from Table 1 below, but the actual geographical location of the user user cannot be determined.

Table I:

WIFI logo
WIFI_id1
WIFI_id2
WIFI_id3
WIFI_id4

Step S24: Determine a real-time geographic location of the terminal device according to the multiple wireless network identifiers.

In the above step S24, the server may determine the real-time geographic location of the terminal device according to each wireless network identifier in the wireless network list. Optionally, the solution may select a geographic location of the business circle as a real-time geographic location of the user according to a predetermined algorithm from a plurality of business circle address locations preset in the server.

The user is still pushed by the user user as an example. The user device's terminal device sends the above wifi list to the server. From the above wifi list, you can get WIFI_id1, WIFI_id2, WIFI_id3, WIFI_id4 around the user. Four wifi logos. The server may obtain multiple geographic locations corresponding to the four wifi identifiers according to a preset algorithm, and filter out a geographic location as a real-time geographic location of the user user.

In step S26, the query obtains at least one application corresponding to the real-time geographic location.

Step S28, pushing at least one application to the terminal device.

In the above steps S26 to S28, a plurality of geographical locations may be pre-stored in the server, and each geographical location corresponds to an application. After determining the real-time geographic location of the user, the server may push multiple applications corresponding to the real-time geographic location of the user to the user's terminal device.

The user is still pushed by the user user as an example. The server can pre-store various types of APPs corresponding to geographical locations and geographic locations. For example, the APP corresponding to the Zhengjia Food Court is a gourmet app, the APP corresponding to the Gaode Fitness Plaza is a fitness app, and the APP corresponding to the human art theater is a drama. APP, the corresponding APP of Deyun Society is the comic voice APP. If the server determines that the real-time location of the user is “Zhengjia Food Court”, the server can query the “Jiangjia Food Court” corresponding APP as a food app, and then push the food app to the terminal device of the user user. It should be noted that, in this embodiment, since the real-time location of the user is in the “Zhengjia Food Court”, the success rate of the application recommendation is greatly improved when the food app is pushed to the user at this time.

In this embodiment, the wireless network list sent by the terminal device is received, where the wireless network list records each wireless network identifier scanned by the terminal device; the real-time geographic location of the terminal device is determined according to each wireless network identifier; At least one application corresponding to the real-time geographic location; pushing at least one application to the terminal device. It is easy to notice that when pushing the application to the user, the solution pays attention to the real-time geographic location where the user is located, and recommends a suitable application to the user according to the real-time geographic location of the user. Compared with the prior art, the solution solves the problem that only the user's interests are concerned when pushing the application to the user, and the success rate of the application recommendation is low. Furthermore, according to an embodiment of the invention. Pay attention to the user's scene (the user's geographic location) when pushing the application to the user, thereby increasing the success rate of the application recommendation.

In an optional embodiment, before the receiving the wireless network list sent by the terminal device, the method provided by the implementation may further include the following steps.

Step S200: Receive user behavior data sent by the terminal device.

Specifically, in the above step S200, the server may receive user behavior data sent by the terminal device. The above user behavior data may be a ticker log generated according to a target event generated by the client. The target event may include four types of events: (1) the user initiates the APP event, (2) the user downloads the APP event, (3) the user searches for the APP event, and (4) the user browses the APP event. When the terminal device listens to the target event, the target event can be logged to generate a ticker log. The dot log may include: a unique identifier of the user, a target event type (startup, download, search, browse), a target event occurrence time, and an APP package name of the target event operation. The terminal device then sends the ticker log to the server. It should be noted that the server may analyze the user's interests from the above-mentioned management log.

Step S201: Generate an application list according to user behavior data, where the application list includes a plurality of applications.

In the above step S201, after the server obtains the user behavior data of the user, the server may use the online application recommendation module to analyze and process the user behavior data to generate an application list that the user may be interested in. The solution of the above application recommendation module may be an application recommendation based on a leaderboard, a recommendation based on personalized application, an application recommendation based on a business package, and an application recommendation based on an operator configuration. It should be noted that the online application recommendation module can collect data from the terminal device, and generate a recommendation candidate list, that is, the application list, in an offline or real-time manner, in combination with an actual recommended recommendation manner.

Step S202: Obtain each application in the application list.

Step S203, establishing each of the application and each preset geographic location according to a preset configuration rule. Mapping relationship.

In the foregoing steps S202 to S203, the server may establish a mapping relationship between each of the applications and each preset geographic location according to a preset configuration rule, that is, push the user when the user is in a specific real-time location. Specific application.

Still taking the user user push APP as an example, after the server receives the user behavior data of the user user, the server can analyze that the user user has browsed the pages of food, fitness, cross talk, drama, and the like many times. The server can generate a list of applications based on the user's user behavior data. The app list can include a gourmet app, a fitness app, a cross talk app, and a drama app. It should be noted that the rules for application push can be pre-stored in the server. For example, the gourmet app needs to be pushed to the user user when the user is in the Zhengjia food court. In the above application list, the location corresponding to the food app is Zhengjia Food Court. According to the above program pushing rules, the geographic location corresponding to the fitness APP is Gaode Fitness Square, and the geographical location corresponding to the comic sound APP is Deyun Society, and the geographical location corresponding to the drama APP is the human art theater.

In an optional embodiment, in step S26, the step of querying at least one application query corresponding to the real-time geographic location to obtain at least one application corresponding to the real-time geographic location comprises the following steps.

Step S261: Obtain at least one application corresponding to the real-time geographic location from the application list according to the mapping relationship.

In the above step S261, the server may obtain at least one application corresponding to the real-time geographic location of the user from the application list according to the mapping relationship, for example, the real-time geographic location of the user user is a Zhengjia food plaza, and the server queries the Zhengjia cuisine. The app corresponding to the square is a gourmet app.

In an optional embodiment, in step S24, the step of determining the real-time geographic location of the terminal device according to each wireless network identifier may include the following steps.

Step S241, matching multiple wireless network identifiers with multiple pre-stored wireless network identifiers. In a case that the first wireless network identifier of the plurality of wireless network identifiers matches the first pre-stored wireless network identifier of the plurality of pre-stored wireless network identifiers, determining that the first pre-stored business circle identifier corresponding to the first pre-stored wireless network identifier is Hit the location.

In the above step S241, the first wifi list may be pre-stored in the server. The first wifi list is used to store the plurality of pre-stored wireless network identifiers. It should be noted that the server may collect and label the wifi identification list of the target locations such as popular business districts, stores, and office buildings in advance. An example of the first wifi list pre-stored in the server is shown in Table 2. It should be noted that the first wifi list in Table 2 is pre-stored by the server, and the wifi list in Table 1 is obtained by real-time scanning of the user's terminal device, and the two are different. The server may first obtain any one of the first wireless network identifiers, such as WIFI_id1, WIFI_id2, WIFI_id3, and WIFI_id4, from the foregoing Table 1, and then respectively obtain the identifiers of the obtained four wifis in Table 2 (the first wifi list). All the pre-stored wifi identifiers are matched, and the first pre-stored business circle geographic location corresponding to the first pre-stored wireless wifi identifier that is successfully matched is obtained, and the first pre-stored business circle geographic location is determined as the hit geographic location. For example, in this embodiment, the wireless network identifiers WIFI_id1 and WIFI_id3 in Table 1 are successfully matched with the pre-stored wireless network identifiers WIFI_id1, WIFI_id2, and WIFI_id3 in Table 2. In this embodiment, the server determines, in Table 2, the Zhengjia Food Plaza corresponding to WIFI_id1, WIFI_id2, and WIFI_id3, and the Gaode Fitness Plaza as the hitting geographic location.

Table II:

Pre-stored WIFI logo	Booking business location
WIFI_id1	Zhengjia Food Court
WIFI_id2	Zhengjia Food Court

WIFI_id3	Gaode Fitness Square
WIFI_id6	Gaode Fitness Square

Step S242, counting the number of times the first pre-stored business circle geographic location is determined to be the hit geographic location.

Step S243: If the number of times is greater than a preset threshold, determine that the geographic location of the first business circle is a real-time geographic location of the terminal device.

In the above steps S242 to S243, the server may count the number of hits f(i) in the first predetermined business circle geographic location in Table 2, and the first predetermined business circle geography whose hit count f(i) is greater than a preset threshold. The location is determined as the user's real-time geographic location. Preferably, the solution may determine the first business circle geographic location whose hit count f(i) is greater than 0 and the largest is the real-time geographic location of the user.

For example, the user can push the APP as an example. The server can count the number of times the predetermined geographic location is hit. In this embodiment, the Zhengjia Food Plaza is hit twice, and the Gaode Fitness Plaza is hit once, so the server will be a good food. The square is determined as the real-time geographic location of the user user, and the gourmet app is pushed to the user user according to the above mapping relationship.

In an optional embodiment, when it is detected that the terminal device is connected to the wireless hotspot, the user behavior data and/or the wireless network list recorded by the terminal device is acquired.

Specifically, the server can detect whether the terminal device is connected to a wireless hotspot (eg, a wifi router). When the terminal device is connected to the wifi, the server sends an acquisition instruction to the terminal device for acquiring the user behavior data or the wireless network list recorded by the terminal device.

A preferred embodiment of the present application is described below in conjunction with FIG.

As shown in FIG. 3, the present application can first collect two types of data through a client. The first type of data is the log of the mobile client. The second type of data is the wifi scan list of the mobile phone. The first type of data is sent by the client to the online application recommendation module of the server. The second type of data is sent to the offline scene recognition module of the server. The online application recommendation module is configured to calculate a user's demand for each APP according to the dot data of the user data, and obtain an APP recommendation candidate set. The offline scene recognition module is configured to analyze the current scene of the user according to the wifi scan list of the user. When the offline scene recognition module recognizes the target scene, the online application recommendation module is triggered. The application recommendation module pushes the recommended application list to the client of the user according to the APP recommendation candidate set, thereby implementing the online and offline APP recommendation.

Optionally, in combination with FIG. 3, the step of the client collecting the first type of data may be as follows.

Step A: Monitor the target event of the mobile client. The target event includes four types of events: a. the user initiates the APP event; b. the user downloads the APP event; c. the user searches for the APP event; d. the user browses the APP event.

Step B: When the target event is detected, the target event is recorded, including: the unique identifier of the user, the target event type (start, download, search, browse), the target event occurrence time, and the APP package name of the target event operation.

Step C: When the connection wifi is detected, the log is uploaded to the server.

Optionally, the step of the client collecting the second type of data may be as follows:

Step A: Step B is started periodically according to a certain time interval.

Step B: Read the current wifi scan list.

Step C: The wifi scan list read in step B is uploaded in real time in the format of the smart device unique identifier/wifi scan list/time.

A further preferred embodiment of the present application is described below with reference to FIG. 4, and the steps of the embodiment may be as follows.

Step S41, the client identifies the unique identifier of the user's mobile phone (such as the sim card of the mobile phone), the type of the target event, The time of occurrence and the package name of the APP that the user has paid attention to are uploaded to the big data platform of the server.

Specifically, in the above step S41, the data uploaded by the client to the server is the first type of data.

In step S42, the client uploads the unique identifier of the user's mobile phone, the wifi scan list, and the current time to the big data platform of the server.

Specifically, in the above step S42, the data uploaded by the client to the server is the second type of data.

Step S43, the big data platform sends the first type of data to the recommendation platform by using the offline or real-time calculation method (based on the recommendation of the leaderboard, based on the recommendation of the business package, based on the personalized recommendation and based on the manual The recommendation platform generates a candidate recommendation list based on the first type of data described above.

Step S44: The big data platform sends the second type of data to the wifi address matching unit, and the wifi address matching unit calculates the actual scene where the user is located according to the second type of data.

In step S45, the recommendation platform pushes the corresponding APP recommendation list to the user according to the actual scene where the user is located.

In summary, the present application proposes an APP recommendation framework that combines online APP recommendation and offline scene recognition triggering. Specifically, the online APP recommendation includes a plurality of APP recommendation methods. One of the offline scene recognitions includes comparing the current wifi scan list of the user with the wifi business circle (location) mapping table of the server, thereby achieving the purpose of identifying the business circle (location). The APP recommendation is triggered by the identified business circle (place) to select the APP to be recommended according to the configuration rule based on the online APP candidate recommendation list.

It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations. However, it should be understood by those skilled in the art that the present invention is not limited by the described order of the acts. In accordance with the present invention, certain steps may be performed in other sequences or concurrently. Those skilled in the art should also appreciate that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a general hardware platform. In addition, it can also be implemented by hardware. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product. The computer software product can be stored in a storage medium (such as a ROM/RAM, a magnetic disk, an optical disk), and includes a plurality of instructions for causing a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to perform various implementations of the present invention. Example method.

Example 2

According to an embodiment of the present invention, there is further provided a push device for an application for implementing a push method of the above application. As shown in FIG. 5, the apparatus may include: a first receiving unit 50, configured to receive a wireless network list sent by the terminal device, where the wireless network list records the plurality of wireless network identifiers scanned by the terminal device; the determining unit 52, And determining, by the plurality of wireless network identifiers, a real-time geographic location of the terminal device; the querying unit 54 is configured to query at least one application corresponding to the real-time geographic location; and the pushing unit 56 is configured to push the at least one application to the terminal device .

In this embodiment, the wireless network list sent by the terminal device is received, wherein the wireless network list records the plurality of wireless network identifiers scanned by the terminal device; the real-time geographic location of the terminal device is determined according to the plurality of wireless network identifiers; At least one application corresponding to the real-time geographic location; pushing at least one application to the terminal device.

It is easy to note that, according to an embodiment of the present invention, when an application is pushed to a user, the real-time geographic location where the user is located is focused, and a suitable application is recommended to the user according to the real-time geographic location of the user. Compared to the prior art, according to one embodiment of the present invention, not only user interest is concerned when pushing an application to a user. Alternatively, the success rate of the application recommendation may be improved in accordance with an embodiment of the present invention. a according to the invention Embodiments may focus on the user's location (the user's geographic location) when pushing the application to the user to increase the success rate of the application recommendation.

Optionally, the foregoing apparatus may further include: a second receiving unit, receiving user behavior data sent by the terminal device; and a generating unit, configured to generate an application list according to the user behavior data, where the application list includes multiple applications; the acquiring unit, For acquiring each application in the application list; establishing a unit, configured to establish a mapping relationship between each application and each preset geographic location according to a preset configuration rule.

Optionally, the querying unit may further include: a first obtaining module, configured to acquire, according to the mapping relationship, at least one application corresponding to the real-time geographic location from the application list.

Optionally, the determining unit includes: a first matching module, configured to match the multiple wireless network identifiers with the plurality of pre-stored wireless network identifiers, the first wireless network identifier of the plurality of wireless network identifiers, and the plurality of pre-stored wireless identifiers The first pre-stored wireless network identifier corresponding to the first pre-stored wireless network identifier is determined to be a hit geographic location, and the statistical module is configured to collect the first pre-stored commercial circle geographic location. Determine the number of times to hit the location. And a determining module, configured to determine, in a case that the number of times is greater than a preset threshold, a geographical location of the first pre-stored business circle as a real-time geographic location of the terminal device.

Optionally, the foregoing apparatus may further include: a first acquiring module, configured to acquire user behavior data and/or a wireless network list recorded by the terminal device when detecting that the terminal device is connected to the wireless hotspot.

Example 3

The application also provides a server. As shown in FIG. 6, the server may include: a receiver 60, configured to receive a wireless network list sent by the terminal device, where the wireless network list records a plurality of wireless network identifiers scanned by the terminal device; and the processor 62 is configured to The wireless network identifier determines a real-time geographic location of the terminal device and queries for at least one application corresponding to the real-time geographic location; the transmitter 64 is configured to push the at least one application to the terminal device.

In this embodiment, the wireless network list sent by the terminal device is received, wherein the wireless network list records the plurality of wireless network identifiers scanned by the terminal device; the real-time geographic location of the terminal device is determined according to the plurality of wireless network identifiers; At least one application corresponding to the real-time geographic location; pushing at least one application to the terminal device.

According to an embodiment of the present invention, when the application is pushed to the user, the solution pays attention to the real-time geographic location where the user is located, and recommends a suitable application to the user according to the real-time geographic location of the user. Compared to the prior art, according to one embodiment of the present invention, not only user interest is concerned when pushing an application to a user. Alternatively, the success rate of the application recommendation may be improved in accordance with an embodiment of the present invention. According to an embodiment of the present invention, the user's scene (the user's geographic location) can be focused on when the application is pushed to the user, so as to improve the success rate of the application recommendation.

In addition, the present application also discloses an application recommendation method and apparatus to solve the problem that the existing application recommendation method cannot meet the needs of the user, and the user may consume a lot of time and effort when acquiring the application required by the user.

One embodiment of the present application discloses an application recommendation method. Referring to the workflow diagram shown in FIG. 7, the application recommendation method includes the following steps:

Step S71: Listening to target events for each application.

For example, the target event typically includes a launch event, and/or a download event, and/or a search event, and/or a browse event. Of course, the target event may also be another event that performs an operation on the application, which is not limited in this application. When the terminal has a target event for the application, it often indicates that the user is interested in the application.

Step S72: Generate a monitoring record according to the monitoring result, where the monitoring record includes: a user identifier, a target event occurrence time of each type, and a name of an application targeted by the target event.

The user identifier may be an account used by the user to log in to the website, such as a Taobao account, a microblog account, or the like, or may be another identifier that can distinguish the user.

Step S73: Transfer the monitoring record to the server.

Step S74: Receive an application recommendation list generated by the server according to the interception record. After receiving the monitoring record, the server calculates a predicted value of the user's demand for each application according to the information included in the monitoring record, and generates the application according to the predicted value of the user's demand for each application. Program recommendation list.

The demand forecast value is used to characterize the degree of user demand for the application. The higher the demand forecast value, the higher the user's demand for the application.

After the terminal generates the monitoring record, the terminal transmits the monitoring record to the server. After receiving the monitoring record transmitted by each terminal, the server analyzes and processes the monitoring record, calculates a predicted value of the user's demand for each application, and generates a corresponding application recommendation list according to the same, and pushes the application recommendation list. To the terminal. The terminal can download and install the corresponding application according to the application recommendation list.

One embodiment of the present application discloses an application recommendation method. In the method, the terminal listens to target events for each application, generates a listening record according to the monitoring result, and transmits the monitoring record to the server. After receiving the monitoring record, the server calculates a predicted value of the user's demand for each application according to the information included in the monitoring record, and generates the application recommendation list according to the predicted value of the user's demand for each application. And transmitting the application recommendation list to the terminal, so that the terminal determines the application required by the user according to the application recommendation list.

The solution disclosed in the present application is capable of obtaining an application recommendation list. The application recommendation list is generated by the server according to the user's demand forecast value of the application, so the recommended application of the application list meets the user's requirements, so that the user's demand for the application can be predicted more accurately. According to an embodiment of the present invention, user requirements can be met. In addition, you can reduce the time and effort users spend on getting the applications they need.

In step S73, the monitoring record is transmitted to the server. In this step, the terminal can transmit the monitoring record through its own traffic. In addition, in order to save traffic, it can also be transmitted over a wireless network. In this case, the transmitting the monitoring record to the server includes the following steps.

Check if the terminal is connected to the wireless network. If the terminal is connected to the wireless network, the monitoring record is transmitted to the server through the wireless network; if the terminal is not connected to the wireless network, the monitoring record is transmitted through the terminal self-contained traffic, or if it is determined that the terminal is not connected The wireless network detects whether the terminal is connected to the wireless network at regular intervals until it is determined that the terminal is connected to the wireless network according to the detection result, and then transmits the monitoring record through the wireless network.

Through the above operations, the terminal can preferentially transmit the wireless network through the wireless network, thereby saving the traffic of the terminal and saving the user the expenditure of the network tariff.

With the development of technology, there are many types of applications. In particular, in smart mobile terminals such as smart phones, multiple applications are often installed to meet user needs. In this case, the smart mobile terminal may use the application recommendation method disclosed in the present application to generate a monitoring record, transmit the monitoring record to the server, and receive an application recommendation list generated by the server according to the monitoring record, according to the application. The recommendation list downloads the corresponding program, and through the application recommendation list, the user's demand for the application can be more accurately predicted, and the user experience is improved.

Accordingly, another embodiment of the present application discloses an application recommendation method. Referring to the workflow diagram shown in FIG. 8, the application recommendation method includes the following steps.

Step S81: Obtain a monitoring record transmitted by each terminal, where the monitoring record includes: a user identifier, a target event occurrence time of each type, and a name of an application for which the target event is targeted.

The target event typically includes a launch event, and/or a download event, and/or a search event, and/or a browse event. Of course, the target event may also be another event that performs an operation on the application, which is not limited in this application. When the terminal has a target event for the application, it often indicates that the user is interested in the application.

Step S82: Calculate a predicted value of the demand of each user for each application according to each piece of information included in the monitoring record.

Step S83: Generate an application recommendation list according to the predicted value of the user's demand for each application, and transmit the application recommendation list to the target terminal.

The demand forecast value is used to characterize the degree of user demand for the application. The higher the demand forecast value, the higher the user's demand for the application.

Another embodiment of the present application discloses an application recommendation method. In the method, after obtaining the monitoring record transmitted by each terminal, the server calculates a predicted value of the demand of each application according to the monitoring record, and then generates an application recommendation list according to the predicted value of each application. And transmitting the application recommendation list to the terminal, so that the terminal downloads an application required for installation according to the application recommendation list.

The application recommendation list can be obtained by the solution disclosed in the present application. The application recommendation list is generated by the server based on the user's predicted demand value for the application. The application list recommended by the application list is more in line with the user's needs, so as to more accurately predict the user's demand for the application. According to an embodiment of the present invention, user requirements can be met. In addition, you can reduce the time and effort users spend on getting the applications they need.

According to the application recommendation method disclosed in the present application, in step S82, the predicted demand value of the user for each application is calculated based on each piece of information included in the monitoring record. The calculating the predicted value of the user's demand for each application according to each piece of information included in the monitoring record includes the following steps.

And determining, according to the target event occurrence time and the name of the application for the target event, the number of times the target event corresponding to each application is generated within a preset time period. The predicted value of the demand of the user for each application is calculated according to the number of occurrences of the target event corresponding to the respective applications.

Referring to the workflow diagram shown in FIG. 9, the calculating the predicted value of the demand for each application by the user according to the number of occurrences of the target event corresponding to the respective applications includes the following steps.

Step S91: Calculate a demand value of the user for each application according to the number of occurrences of the target event corresponding to each application.

Step S92: Create a user requirement matrix according to the user's demand value for each application, where the user requirement matrix includes a demand value of each user for each application.

The user requirement matrix can be as shown in Table 3. In the table, the demand value of each user for each application is filled in. For example, in Table 3, user 1 has a demand value of 3 for application 1, user 1 has a demand value of 0 for application 2, and user 2 has application 1 The demand value is 0.

Table 3

Of course, the user requirement matrix may also adopt other forms, which is not limited in this application.

In addition, if the required value is calculated to be 0 in most cases, the user demand matrix may also be referred to as a user demand sparse matrix.

Step S93: Decompose the user requirement matrix to obtain a user feature matrix and an application feature matrix. The user feature matrix is used to characterize a degree of preference of the user to features included in the application, the application feature matrix being used to characterize the degree of similarity of each application to the feature.

In the present application, the user requirement matrix is decomposed into two matrixes: a user feature matrix and an application feature matrix. For example, the user demand matrix is usually decomposed by methods such as least squares method and stochastic gradient descent method. Of course, other methods of matrix decomposition may also be used, which is not limited in this application.

The feature is a concept in the field of machine learning. Through the features, the characteristics of the application can be reflected. For example, if an application is more similar to the feature of the video, the application is related to the video.

Step S94: Calculate, according to the user feature matrix and the application feature matrix, a predicted value of the demand of the user for each application.

If the target event includes: launching an event, and/or downloading an event, and/or searching for an event, and/or browsing an event. When calculating the user's demand value for each application according to the number of occurrences of the target event corresponding to each application, the following formula is adopted:

R _ij =t ₁ ×start _ij +t ₂ ×down _ij +t ₃ ×search _ij +t ₄ ×view _ij ;

Where R _ij represents the demand value of user i for application j, start _ij represents the number of times user i starts application j, down _ij represents the number of times user i downloads application j, and search _ij represents the number of times user i searches for application j , view _ij represents the number of times user i browses the application j, and t ₁ , t ₂ , t _{3 ,} and t ₄ are preset constants.

In the above formula, t ₁ , t ₂ , t ₃ and t ₄ are constants, and the value of the constant can be set in advance according to the emphasis on different target events.

Further, when calculating the predicted value of the user's demand for each application according to the user feature matrix and the application feature matrix, the following formula is adopted:

Where P _ij represents the predicted value of the demand of the user i for the application j, U _ik is the element in the user feature matrix, represents the degree of preference of the user i for the feature of the k category, and V _kj is the element in the application feature matrix, indicating the application The degree of similarity between the program j and the features of the k category, and N indicates the number of categories of features.

When calculating the user's demand forecast value for each application, the required elements are sequentially extracted from the user feature matrix and the application feature matrix, respectively, and then the user's demand forecast value for each application can be obtained according to the above formula.

After calculating the predicted demand value of the user for each application, the server generates an application recommendation list for the user according to the demand prediction value, so that the user downloads and installs the application required by the user according to the application recommendation list.

Wherein, in the application recommendation list, each application according to the calculated demand prediction value is from a large Arranged in small order. Each application included in the application recommendation list may include all applications whose demand prediction values are within a preset range. The server may also determine an application that has been installed in the terminal and remove the application already installed in the terminal from the application recommendation list.

Accordingly, in another embodiment of the present application, an application recommendation device is disclosed. Referring to the structural diagram shown in FIG. 10, the application recommendation apparatus includes: a listening module 110, a record generating module 120, a transmitting module 130, and a receiving module 140.

The listening module 110 is configured to listen to target events for each application. The target event typically includes a launch event, and/or a download event, and/or a search event, and/or a browse event. Of course, the target event may also be another event that performs an operation on the application, which is not limited in this application.

The record generation module 120 is configured to generate a listen record according to the monitoring result. The monitoring record includes: a user identifier, a time of occurrence of each type of target event, and a name of an application targeted by the target event. The user identifier may be an account used by the user to log in to the website, such as a Taobao account, a microblog account, or the like, or may be another identifier that can distinguish the user.

The transmission module 130 is configured to transmit the monitoring record to a server.

The receiving module 140 is configured to receive an application recommendation list generated by the server according to the interception record. After receiving the monitoring record, the server calculates a predicted value of the user's demand for each application according to the information included in the monitoring record, and generates the application according to the predicted value of the user's demand for each application. Program recommendation list.

. The demand forecast value is used to characterize the degree of user demand for the application. The higher the demand forecast value, the higher the user's demand for the application.

Further, the transmission module 130 includes: a detecting unit, configured to detect whether the terminal is connected to the wireless network, and a wireless transmission unit, configured to transmit the monitoring record to the server by using the wireless network if the terminal is connected to the wireless network.

The device disclosed in the present application enables the terminal to obtain an application recommendation list. The application recommendation list is generated by the server based on the user's predicted demand value for the application. The application list recommended application meets the user's needs to more accurately predict the user's needs for the application. According to an embodiment of the present invention, user requirements can be met. In addition, you can reduce the time and effort users spend on getting the applications they need.

The specific manner in which the operations performed by the various modules in the apparatus in the above-described embodiments have been described in detail in the embodiments relating to the method will not be described in detail herein.

Accordingly, another embodiment of the present application discloses an application recommendation device. Referring to the structural diagram shown in FIG. 11, the application recommendation apparatus includes: an acquisition module 210, a calculation module 220, and a recommendation module 230.

The obtaining module 210 is configured to acquire a monitoring record transmitted by each terminal. The monitoring record includes: a user identifier, a time of occurrence of each type of target event, and a name of an application targeted by the target event. The target event typically includes a launch event, and/or a download event, and/or a search event, and/or a browse event. Of course, the target event may also be another event that performs an operation on the application, which is not limited in this application.

The calculation module 220 is configured to calculate, according to each piece of information included in the monitoring record, a predicted value of the demand of the user for each application, wherein the demand predicted value is used to represent the degree of demand of the user for the application. The higher the demand forecast value indicates the higher the user's demand for the application;

The recommendation module 230 is configured to generate an application recommendation list according to the predicted value of the user's demand for each application, and transmit the application recommendation list to the target terminal.

Further, the calculation module 220 includes: a statistics sub-module and a calculation sub-module.

The statistic sub-module is configured to count, according to the target event occurrence time and the name of the application program targeted by the target event, the target terminal corresponding to each user identifier in a preset time period, and the target corresponding to each application The number of times the event occurred;

The calculation sub-module is configured to calculate a predicted value of the demand of each user for each application according to the number of occurrences of the target event corresponding to the respective applications.

Further, the calculation sub-module includes: a requirement value calculation unit, configured to calculate a user's demand value for each application according to the number of occurrences of the target event corresponding to each application; a matrix creation unit, configured to A user requirement matrix is created by the user, and the user requirement matrix includes a requirement value of each user for each application; a matrix decomposition unit is configured to decompose the user requirement matrix to obtain a user. a feature matrix and an application feature matrix, wherein the user feature matrix is used to characterize a degree of preference of the user for features included in the application, the application feature matrix being used to characterize the degree of similarity of each application to the feature; prediction And a value calculation unit, configured to calculate, according to the user feature matrix and the application feature matrix, a predicted predicted value of the user for each application.

Further, if the target event includes: launching an event, and/or downloading an event, and/or searching for an event, and/or browsing an event. The demand value calculation unit calculates the user's demand value for each application according to the number of occurrences of the target event corresponding to the respective applications, and adopts the following formula:

R _ij =t ₁ ×start _ij +t ₂ ×down _ij +t ₃ ×search _ij +t ₄ ×view _ij ;

Where R _ij represents the demand value of user i for application j, start _ij represents the number of times user i starts application j, down _ij represents the number of times user i downloads application j, and search _ij represents the number of times user i searches for application j , view _ij represents the number of times user i browses the application j, and t ₁ , t ₂ , t _{3 ,} and t ₄ are preset constants.

In the above formula, t ₁ , t ₂ , t ₃ and t ₄ are constants, and the value of the constant can be set in advance according to the focus of different target events.

Further, when the predicted value calculation unit calculates the predicted demand value of each user for each application according to the user feature matrix and the application feature matrix, the following formula is adopted:

Where P _ij represents the predicted value of the demand of the user i for the application j, U _ik is the element in the user feature matrix, represents the degree of preference of the user i for the feature of the k category, and V _kj is the element in the application feature matrix, indicating the application The degree of similarity between the program j and the features of the k category, and N indicates the number of categories of features.

When calculating the user's demand forecast value for each application, the required elements are sequentially extracted from the user feature matrix and the application feature matrix, respectively, and then the user's demand forecast value for each application can be obtained according to the above formula.

After the user predicts the demand forecast value of each application, the recommendation module 240 generates an application recommendation list for the user according to the demand prediction value, so that the user downloads and installs the application required by the user according to the application recommendation list. program.

In the application recommendation list, each application is sequentially arranged in accordance with the calculated demand prediction values from large to small. Each application included in the application recommendation list may include all applications whose demand prediction values are within a preset range. In addition, the recommendation module 240 may also determine an application that has been installed in the terminal, and remove an application already installed in the terminal from the application recommendation list.

With regard to the apparatus in the above embodiments, the specific manner in which the respective modules perform the operations has been described in detail in the embodiment relating to the method, and will not be explained in detail herein. The above embodiments can also be summarized as follows

EEEE1, an application recommendation method, comprising: monitoring a target event for each application; generating a monitoring record according to the monitoring result, wherein the monitoring record includes: a user identifier, each type of target event a name of the application for which the time and target event is generated; transmitting the monitoring record to the server; receiving an application recommendation list generated by the server according to the monitoring record, wherein the server receives the monitoring record And calculating a demand forecast value of the user for each application according to each piece of information included in the monitoring record, and generating the application recommendation list according to the predicted value of the user's demand for each application.

EEEE 2. The application recommendation method according to EEEE1, wherein the transmitting the monitoring record to a server comprises: detecting whether a terminal is connected to a wireless network; and if the terminal is connected to a wireless network, passing the wireless network The monitoring record is transmitted to the server.

EEEE 3, an application recommendation method, comprising: acquiring a monitoring record transmitted by each terminal, wherein the monitoring record includes: a user identifier, each type of target event occurrence time, and an application targeted by the target event. a name; a predicted value of the user's demand for each application is calculated according to each piece of information included in the monitoring record; and an application recommendation list is generated according to the predicted value of the user's demand for each application, and transmitted to the target terminal The application recommendation list.

EEEE 4. The application recommendation method according to EEEE3, wherein the calculating a predicted value of a user's demand for each application according to each piece of information included in the monitoring record comprises: according to the monitoring record The target event occurrence time and the name of the application targeted by the target event, and the number of times the target event corresponding to each application is generated in the preset time period corresponding to the target terminal corresponding to each user identifier; according to the corresponding application The number of times the target event occurs, and the predicted value of the demand of the user for each application is calculated.

EEEE 5. The application recommendation method according to EEEE4, wherein the calculating a predicted value of the user's demand for each application according to the number of occurrences of the target event corresponding to the respective applications, including: Describe the number of occurrences of the target event corresponding to each application, calculate a user's demand value for each application, and create a user requirement matrix according to the user's demand value for each application, where the user requirement matrix includes each user a demand value for each application; decomposing the user requirement matrix to obtain a user feature matrix and an application feature matrix, wherein the user feature matrix is used to characterize the degree of preference of the user for features included in the application, The application feature matrix is used to characterize the similarity degree of each application with the feature; and the predicted value of the demand of the user for each application is calculated according to the user feature matrix and the application feature matrix.

EEEE 6. The application recommendation method according to EEEE5, characterized in that, if the target event comprises: a startup event, and/or a download event, and/or a search event, and/or a browsing event, according to each The number of occurrences of the target event corresponding to the application. When calculating the user's demand value for each application, the following formula is used:

R _ij =t ₁ ×start _ij +t ₂ ×down _ij +t ₃ ×search _ij +t ₄ ×view _ij ;

Where R _ij represents the demand value of user i for application j, start _ij represents the number of times user i starts application j, down _ij represents the number of times user i downloads application j, and search _ij represents the number of times user i searches for application j , view _ij represents the number of times user i browses the application j, and t ₁ , t ₂ , t _{3 ,} and t ₄ are preset constants.

EEEE 7. The application recommendation method according to EEEE 5 or 6, characterized in that The household feature matrix and the application feature matrix, when calculating the predicted demand value of the user for each application, adopt the following formula:

Where P _ij represents the predicted value of the demand of the user i for the application j, U _ik is the element in the user feature matrix, represents the degree of preference of the user i for the feature of the k category, and V _kj is the element in the application feature matrix, indicating the application The degree of similarity between the program j and the features of the k category, and N indicates the number of categories of features.

EEEE 8. An application recommendation device, comprising: a monitoring module, configured to monitor a target event for each application; a record generation module, configured to generate a monitoring record according to the monitoring result, wherein the monitoring The record includes: a user identifier, a time of occurrence of each type of target event, and a name of an application for which the target event is targeted; a transmission module configured to transmit the monitoring record to the server; and a receiving module configured to receive the basis returned by the server The application recommendation list generated by the monitoring record, wherein, after receiving the monitoring record, the server calculates a predicted value of the user's demand for each application according to each piece of information included in the monitoring record, and according to the The user's demand forecast for each application generates the application recommendation list.

EEEE9, the application recommendation device according to EEEE8, wherein the transmission module comprises: a detecting unit, configured to detect whether the terminal is connected to the wireless network; and a wireless transmission unit, configured to: if the terminal is connected to the wireless network, The wireless network transmits the monitoring record to the server.

EEEE10, an application recommendation device, comprising: an acquisition module, configured to acquire a monitoring record transmitted by each terminal, wherein the monitoring record includes: a user identifier, each type of target event occurrence time, and a target event. a name of the application; a calculation module, configured to calculate a predicted value of the user's demand for each application according to each piece of information included in the monitoring record; and a recommendation module, configured to: according to the user's demand for each application Predicting the value, generating an application recommendation list, and transmitting the application recommendation list to the target terminal.

EEEE11, the application recommendation device according to the EEEE10, wherein the calculation module comprises: a statistics sub-module, configured to: according to the target event occurrence time and the name of the application targeted by the target event included in the monitoring record, Counting the number of times the target event corresponding to each application is generated by the target terminal corresponding to each user identifier in a preset time period; and calculating a sub-module, configured to calculate the user according to the number of occurrences of the target event corresponding to each application The predicted value of the demand for each application.

EEEE12, the application recommendation device according to the EEEE11, wherein the calculation submodule comprises: a requirement value calculation unit, configured to calculate a user for each application according to the number of occurrences of the target event corresponding to the respective application programs a demand value; a matrix creation unit, configured to create a user requirement matrix according to the user's demand value for each application, wherein the user requirement matrix includes a demand value of each user for each application; a matrix decomposition unit, And a method for decomposing the user requirement matrix to obtain a user feature matrix and an application feature matrix, wherein the user feature matrix is used to represent a degree of preference of the user for a feature included in an application, where the application feature matrix is used for Determining the degree of similarity between the respective applications and the features; the predicted value calculating unit is configured to calculate, according to the user feature matrix and the application feature matrix, the predicted predicted value of the user for each application.

EEEE13. The application recommendation device according to EEEE12, wherein the target value event comprises: a startup event, and/or a download event, and/or a search event, and/or a browsing event, the demand value calculation unit Calculating the user's needs for each application according to the number of occurrences of the target events corresponding to the respective applications When evaluating, the following formula is used:

R _ij =t ₁ ×start _ij +t ₂ ×down _ij +t ₃ ×search _ij +t ₄ ×view _ij ;

Where R _ij represents the demand value of user i for application j, start _ij represents the number of times user i starts application j, down _ij represents the number of times user i downloads application j, and search _ij represents the number of times user i searches for application j , view _ij represents the number of times user i browses the application j, and t ₁ , t ₂ , t _{3 ,} and t ₄ are preset constants.

EEEE 14. The application recommending apparatus according to EEEE 12 or 13, wherein the predicted value calculating unit calculates a demand forecast value of each user for each application according to the user feature matrix and the application feature matrix , using the following formula:

Where P _ij represents the predicted value of the demand of the user i for the application j, U _ik is the element in the user feature matrix, represents the degree of preference of the user i for the feature of the k category, and V _kj is the element in the application feature matrix, indicating the application The degree of similarity between the program j and the features of the k category, and N indicates the number of categories of features.

Other embodiments of the invention will be apparent to those skilled in the <RTIgt; The present application is intended to cover any variations, uses, or adaptations of the present invention, which are in accordance with the general principles of the present invention and include common general knowledge or conventional technical means in the art that are not disclosed in the present disclosure. . The specification and examples are to be considered as illustrative only,

It is to be understood that the invention is not limited to the details of the details of The scope of the invention is limited only by the appended claims.

Claims (11)

1. A method for pushing an application, comprising:

   Receiving, by the terminal device, a wireless network list, where the wireless network list records multiple wireless network identifiers scanned by the terminal device;

   Determining a real-time geographic location of the terminal device according to the plurality of wireless network identifiers;

   Querying to obtain at least one application corresponding to the real-time geographic location;

   Pushing the at least one application to the terminal device.
2. The method according to claim 1, wherein the method further comprises: before receiving the list of wireless networks sent by the terminal device, the method further comprising:

   Receiving user behavior data sent by the terminal device;

   Generating an application list according to the user behavior data, wherein the application list includes a plurality of the applications;

   Obtaining each application in the list of applications;

   Establish a mapping relationship between each application and each preset geographic location according to a preset configuration rule.
3. The method according to claim 2, wherein the step of obtaining at least one application corresponding to the real-time geographic location comprises:

   Obtaining the at least one application corresponding to the real-time geographic location from the application list according to the mapping relationship.
4. The method according to any one of claims 1 to 3, wherein the determining the real-time geographic location of the terminal device according to the plurality of wireless network identifiers comprises:

   Matching the plurality of wireless network identifiers with the plurality of pre-stored wireless network identifiers, wherein the first wireless network identifier of the plurality of wireless network identifiers matches the first pre-stored wireless network identifier of the plurality of pre-stored wireless network identifiers In the case that the first pre-stored business circle corresponding to the first pre-stored wireless network identifier is determined to be a hit geographic location;

   Counting, by the first pre-stored business circle, the number of times the location is determined as the hit location;

   If the number of times is greater than a preset threshold, determining that the first pre-stored business circle location is a real-time geographic location of the terminal device.
5. The method according to claim 2, wherein the user behavior data and/or the wireless network list recorded by the terminal device is acquired when detecting that the terminal device is connected to the wireless hotspot.
6. A push device for an application, comprising:

   a first receiving unit, configured to receive a wireless network list sent by the terminal device, where the wireless network list records multiple wireless network identifiers scanned by the terminal device;

   a determining unit, configured to determine a real-time geographic location of the terminal device according to the multiple wireless network identifiers;

   a query unit, configured to query to obtain at least one application corresponding to the real-time geographic location;

   a pushing unit, configured to push the at least one application to the terminal device.
7. The device according to claim 6, wherein the device further comprises:

   a second receiving unit, receiving user behavior data sent by the terminal device;

   a generating unit, configured to generate an application list according to the user behavior data, where the application list includes a plurality of the applications;

   An obtaining unit, configured to acquire each application in the application list;

   The establishing unit is configured to establish a mapping relationship between each application and each preset geographic location according to a preset configuration rule.
8. The device according to claim 7, wherein the query unit comprises:

   a first acquiring module, configured to acquire, according to the mapping relationship, the at least one application corresponding to the real-time geographic location from the application list.
9. The apparatus according to any one of claims 6-8, wherein the determining unit comprises:

   a first matching module, configured to match the multiple wireless network identifiers with the plurality of pre-stored wireless network identifiers, where the first wireless network identifier and the plurality of pre-stored wireless network identifiers of the plurality of wireless network identifiers And determining, by the pre-stored wireless network identifier, the first pre-stored business circle corresponding to the first pre-stored wireless network identifier as a hitting geographic location;

   a statistics module, configured to count the number of times that the first pre-stored business circle geographic location is determined to be the hit geographic location;

   And a determining module, configured to determine, in the case that the number of times is greater than a preset threshold, that the first pre-stored business circle location is a real-time geographic location of the terminal device.
10. The device according to claim 7, wherein the device further comprises: a first obtaining module, configured to acquire the user behavior recorded by the terminal device when detecting that the terminal device is connected to a wireless hotspot Data and/or the list of wireless networks.
11. A server, comprising:

    a receiver, configured to receive a wireless network list sent by the terminal device, where the wireless network list records multiple wireless network identifiers scanned by the terminal device;

    a processor, configured to determine a real-time geographic location of the terminal device according to the multiple wireless network identifiers, and query to obtain at least one application that corresponds to the real-time geographic location;

    a transmitter for the terminal device to push the at least one application.

Images