CN105227445B - Recommend platform using methods and applications are recommended - Google Patents

Abstract

The present invention provides a kind of application and methods and applications is recommended to recommend business platform, wherein this method includes：Using the application data for recommending business platform to acquire acquisition user from carrier server；Using recommending business platform application data to carry out data mining analysis, the characteristic information of user is obtained；The list information of the application to be recommended to user is determined, and the list information of application is sent to user according to characteristic information using business platform is recommended.The demand according to user itself is realized to the recommendation list of user's sending application, meets individual demand of the different user for application.

Description

Application recommendation method and application recommendation platform

technical field

The invention relates to mobile Internet technology, in particular to an application recommendation method and an application recommendation service platform.

Background technique

With the continuous development of mobile Internet technology, various applications emerge as the times require. These applications can be applied to mobile terminals, and networks or application stores can recommend various applications to users.

In the prior art, the network or application store can obtain information such as the number of searches, download times, and user ratings of all users for each application, and then calculate these information to obtain a comprehensive score for each application; the network or application store according to each application The comprehensive score of each application is sorted, so that the application with a higher comprehensive score is recommended to the user.

However, the application recommendation method provided by the prior art can only recommend applications with a high number of searches and downloads by users, which cannot meet the personalized needs of different users for applications.

Contents of the invention

The present invention provides an application recommendation method and an application recommendation service platform, which are used to solve the problem that the application recommendation method provided by the prior art can only recommend applications with high search times and download times, and cannot meet the individual needs of different users for applications. The problem.

One aspect of the present invention is to provide an application recommendation method, including:

The application recommendation business platform collects and obtains the user's application data from the operator's server;

The application recommendation service platform performs data mining analysis on the application data to obtain characteristic information of the user;

The application recommendation service platform determines list information of applications to be recommended to the user according to the characteristic information, and sends the list information of the applications to the user.

One aspect of the present invention is to provide an application recommendation service platform, including:

The data acquisition module is used to collect and acquire the user's application data from the operator's server;

A data mining module, configured to perform data mining analysis on the application data to obtain characteristic information of the user;

An application recommending module, configured to determine list information of applications to be recommended to the user according to the characteristic information, and send the list information of the applications to the user.

The present invention collects and acquires the user's application data from the operator's server through the application recommendation service platform; the application recommendation service platform performs data mining and analysis on the application data to obtain the user's feature information; the application recommendation service platform determines to be recommended to the user according to the feature information The list information of the application, and send the list information of the application to the user. By analyzing the user's own characteristics, the application recommendation list is sent to the user according to the user's own needs, so as to meet the personalized needs of different users for the application.

Description of drawings

FIG. 1 is a flow chart of an application recommendation method provided in Embodiment 1 of the present invention;

FIG. 2 is a flow chart of the application recommendation method provided by Embodiment 1 of the present invention;

FIG. 3 is a schematic structural diagram of an application recommendation service platform provided by Embodiment 3 of the present invention;

FIG. 4 is a schematic structural diagram of an application recommendation service platform provided by Embodiment 4 of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Fig. 1 is a flow chart of the application recommendation method provided by Embodiment 1 of the present invention. As shown in Fig. 1, the method of this embodiment includes:

Step 101, the application recommendation service platform collects and acquires the user's application data from the operator's server.

In this embodiment, specifically, a collection device can be deployed on the operator's server to detect the user's behavior data and signaling data on the mobile Internet in real time, and perform in-depth analysis, so that the application recommendation service platform can obtain information from the operator's server Collect and obtain the user's application data, which can be the mobile phone number of the mobile terminal, access website, access start time, access end time, access duration, access traffic, service type, user base station/cell and other information. E.g:

Table 1 Application data

Step 102, the application recommendation service platform performs data mining analysis on the application data to obtain user characteristic information.

In this embodiment, specifically, the application recommendation service platform performs data mining analysis on the application data obtained in step 101, so as to obtain characteristic information of the user within a certain period of time.

Step 103, the application recommendation service platform determines the list information of the applications to be recommended to the user according to the characteristic information, and sends the list information of the applications to the user.

In this embodiment, specifically, the application recommendation service platform determines the characteristics of the user according to the characteristic information, so that the application list information can be recommended to the user according to the user's own characteristics, and the application list information is sent to the user.

In this embodiment, the application recommendation service platform is used to perform data mining analysis on the user's application data obtained from the operator's server to obtain the user's characteristic information, and then determine the list information of the application to be recommended to the user according to the characteristic information. And send the list information of the application to the user. In this way, by analyzing the user's own characteristics, the application recommendation list is sent to the user according to the user's own needs, so as to meet the personalized needs of different users for the application.

Fig. 2 is a flow chart of the application recommendation method provided by Embodiment 1 of the present invention. On the basis of Embodiment 1, as shown in Fig. 2, the method of this embodiment includes:

Step 101, the application recommendation service platform collects and acquires the user's application data from the operator's server.

In this embodiment, specifically, the application recommendation service platform can collect and obtain the user's application data from the operator's server, and these application data can be the mobile phone number of the mobile terminal, the access URL, the access start time, the access end time, and the access duration , access traffic, service type, user base station/cell and other information.

Step 201, the application recommendation service platform performs data mining analysis on the application data to obtain the user's hobbies and/or group attribute characteristics.

In this embodiment, specifically, the application recommendation service platform performs data mining analysis on the application data acquired in step 101. In this embodiment, the specific manner of data mining analysis is not limited. Thus, by mining the application data, we can analyze the user's hobbies and hobbies when using each application software. The hobbies and hobbies represent the user's preference for using each application software; the user's crowd attribute characteristics can also be analyzed, and the crowd attribute characteristics represent Which type of group the user belongs to; can also obtain the user's hobbies and group attribute characteristics at the same time.

Step 202, the application recommendation service platform determines the list information of applications to be recommended to the user according to the characteristics of hobbies and/or attributes of groups of people.

In this embodiment, specifically, the application recommendation service platform can determine the list information of the applications to be recommended to the user according to the user's hobbies and hobbies, for example, the user's hobbies If the characteristics show that the user is more interested in game-like and shopping-like application software, it is determined that the list information of applications to be recommended to the user includes game-like and shopping-like application software. The application recommendation service platform can also determine the list information of the applications to be recommended to the user according to the user's group attribute characteristics. For example, if the user's group attribute characteristics show that the user is an office worker, then determine the list information of the applications to be recommended to the user Including office application software. The application recommendation service platform can also comprehensively determine the list information of the applications to be recommended to the user according to the user's hobbies and group attributes. For example, the user's hobbies show that the user is more interested in shopping applications. At the same time, the group attribute characteristic of the user is an office worker, and it is determined that the list information of applications to be recommended to the user includes shopping and office application software.

In this embodiment, the application recommendation service platform determines the list information of applications to be recommended to the user according to the user's hobbies and/or group attribute characteristics. In this way, by analyzing the user's own characteristics, the application recommendation list is sent to the user according to the user's own needs, so as to meet the personalized needs of different users for the application.

Further, on the basis of the above embodiments, step 101 specifically includes:

The application recommendation service platform collects and obtains the user's access service type, access duration, access frequency, and access traffic corresponding to each application from the operator's server.

Correspondingly, the application recommendation service platform in step 201 performs data mining analysis on the application data to obtain the user's hobbies and interests, specifically including:

The application recommendation business platform determines the score of each application A _i = Q ₁ S _i /S _{total i} +Q ₂ F _i /F _total i according to the access duration S _i , access frequency F _i and access traffic M _i corresponding to each application +Q ₃ M _i /M _{total i} , wherein, Q ₁ , Q ₂ and Q ₃ are different first preset weight values, S _{total i} , F _{total i} and M _{total i} are the access totals of each application accessed by the user respectively. Duration, total visit frequency and total visit traffic, i∈[1,N], where i is a positive integer and N is a positive integer;

The application recommendation service platform determines the interest indices of the applications belonging to the same access service type according to the score A _i of each application and the access service type corresponding to each application Among them, b∈[1,B], b is a positive integer, and B is a positive integer;

The application recommendation service platform determines the basic recommendation index P _i of each application according to each interest index I _b and the pre-acquired scores of each application;

The application recommendation service platform determines each similarity S _ij between each application and the application in the mobile terminal currently used by the user, where j∈[1,J], j is a positive integer, and J is a positive integer;

The application recommendation service platform adjusts the basic recommendation index P _i '={1-(S _i1 -S _threshold )}P _i Q ₁ ' ₊ ...+{1-(S _ij -S _threshold )}P _i Q _i '+...+{1-(S _iJ -S _threshold )}P _i Q' _N , where the S _{threshold is} the preset similarity threshold, and Q _i 'is the second preset weight value;

The application recommendation business platform sorts the applications according to the adjusted basic recommendation index, and obtains an interest recommendation list, which is used to characterize the user's hobbies and interests.

In this embodiment, specifically, the application recommendation service platform can collect various data corresponding to each application of the user from the operator's server, and these data include the type of access service, access duration, and access frequency corresponding to each application. and access traffic. In this way, each application, that is, the access service type, access duration, access frequency, and access traffic of each application can be obtained.

For each application, according to the application's access duration S _i , access frequency F _i and access traffic M _i , determine the total access duration S _total i and the total access frequency F _{total i} of the user's access to each application within the preset time and total access traffic M _{total i} , calculate the score A _i of each application = Q ₁ S _i /S _{total i} +Q ₂ F _i /F _{total i} +Q ₃ M _i /M _{total i} , where, Q ₁ , Q ₂ and Q ₃ are different first preset weight values set according to experience, and i∈[1,N], i is a positive integer, and N is a positive integer. Then count the sum of the scores of the applications belonging to the same access service type, and calculate the interest indexes of the applications belonging to the same access service type according to the obtained scores A _i of each application and the access service types corresponding to each application Among them, b∈[1, B], b is a positive integer, and B is a positive integer. By sorting the interest indexes, and then sorting the applications belonging to the same access service type according to the descending order of the values of the interest indexes, it is possible to obtain which types of applications the user is more interested in.

The application recommendation service platform will obtain the comprehensive ratings of all users for each application in advance, and use the commonly used recommendation algorithm to calculate the basic recommendation index P _i of each application based on these scores and each interest index I _b . The commonly used recommendation algorithm can be application Collaborative recommendation, user coordination recommendation and other algorithms. Then the application recommendation service platform sorts the basic recommendation index P _i of each application in descending order, and selects the top n applications and the basic recommendation indexes corresponding to the top n applications to form a basic recommendation list.

The application recommendation service platform counts the scores of each application installed in the mobile terminal of the current user. This score can be obtained through data mining analysis, or it can be the scores of all users for each application. The application recommendation service platform selects the score ranking as the top m-named applications. According to the tags of each application in the basic recommendation list and the tags of the top m applications in the mobile terminal currently used by the user, the application recommendation service platform determines the relationship between each application in the basic recommendation list and the mobile terminal currently used by the user. The similarity S _ij between the top m applications of , where j∈[1,J], j is a positive integer, and J is a positive integer. Tags can be converted into vectors, and the similarity can be calculated by calculating the cosine of the angle between two vectors.

Then the application recommendation business platform _adjusts the basic recommendation index P _i '={1-(S _i1 -S _threshold )}P _i Q ₁ '+...+{1-(S _ij - S _threshold )}P _i Q _i '+…+{1-(S _iJ -S _threshold )}P _i Q' _N , where S _{threshold is} the preset similarity threshold, and Q _i 'is the second Default weight value.

According to the adjusted basic recommendation index, the application recommendation business platform reorders the applications in the basic recommendation list in descending order according to the adjusted basic recommendation index, and obtains an interest recommendation list. It can be known that the current user is more interested in The applications are ranked at the front of the interest recommendation list, so that the interest recommendation list represents the characteristics of the user's hobbies.

In this embodiment, the hobbies and hobbies of the current user are determined according to the acquired access service type, access duration, access frequency, and access traffic corresponding to each application.

Further, on the basis of the above embodiments, step 101 specifically includes:

The application recommendation service platform collects and obtains the user's access start time, access end time, and access base station corresponding to each application from the operator's server;

Correspondingly, the application recommendation service platform in step 201 performs data mining analysis on the application data to obtain user group attributes, specifically including:

The application recommendation business platform determines the proportion of the user's daily online time according to the access start time and access end time corresponding to each application;

The application recommendation business platform determines the user's daily movement trajectory according to the access base station corresponding to each application;

The application recommendation business platform determines the user's group attribute characteristics according to the proportion of the user's daily online time and the daily movement track.

In this embodiment, specifically, the application recommendation service platform can also collect various data corresponding to each application of the user from the operator's server, and these data include the access start time, access end time, Access the base station.

The application recommendation service platform can calculate the distribution of users' online time in each preset time period according to the access start time and access end time corresponding to each application. Specifically, first divide the time of a day into several time periods of equal length, for example, in the morning (6:00-12:00), in the afternoon (12:00-18:00), in the evening (18:00-24 :00), night time (24:00-6:00). The application recommendation business platform determines the total time spent online by the user in each time period according to the access start time and end time corresponding to each application, so as to determine the proportion of the user's daily online time in each time period. For example, the user's daily online time The Internet surfing conditions of the user are 0.5 hours in the morning, 0.5 hours in the afternoon, 4 hours in the evening, and 0 hours in the night, so the ratio of the user's daily online time is {0.5/24, 0.5/24, 4/24, 0}. At the same time, the application recommendation business platform can compare the proportion of each user's daily online time to obtain the leisure index of each user in each time period. The descending order of the total time sorts the users in each time period, and the leisure index of each time period is preset to five levels, and the leisure index of the top 20% users in a certain time period is 5, 20%~40 % of the user leisure index is 4, 40% to 60% of the user leisure index is 3, 60% to 80% of the user leisure index is 2, 80% and above the user leisure index is 1, the current user A in each time period The percentages of online time are 70%, 70%, 15%, and 100%. Correspondingly, the user's leisure index in each time period is {2,2,5,1}.

The application recommendation service platform can determine the location information carried in the update signaling message of the visiting base station corresponding to each application according to the visiting base station corresponding to each application, and divide the user's location into several areas according to functions, such as business district, living area, etc. Districts, work areas, schools, hospitals, etc. The application recommendation service platform determines the location and environment of the user at each time point according to the location information corresponding to each application, and the access start time and end time corresponding to each application. For example, user A is at 9 o'clock in the morning, where The location is the workspace. The application recommendation service platform detects the user's location changes within the preset time according to the location environment of the user at each time point determined within the preset time, so as to obtain the user's daily movement track, for example, user A At nine in the morning in the work area, at twelve in the noon in the commercial area, and at eight in the evening in the living area.

The application recommendation service platform judges the user's group attribute characteristics according to the determined user's daily online time ratio and daily movement trajectory. Among them, the user's group attribute characteristics can be office workers, students, idlers, etc. For example, user A is in the work area during the day and in the living area at night from Monday to Friday, so it can be inferred that user A is an office worker.

In this embodiment, according to the obtained user's access start time, access end time, and access base station corresponding to each application, the user's daily online time ratio and daily movement trajectory can be determined, and then the user's crowd attribute characteristics can be determined.

Further, on the basis of the above-mentioned embodiments, in step 103, the application recommendation service platform determines the list information of the applications to be recommended to the user according to the feature information, specifically including:

The application recommendation service platform determines the list information of the applications to be recommended to the user according to the characteristic information, the current geographic location of the user and the current time.

In this embodiment, specifically, the application recommendation service platform determines the application to be recommended to the user according to the characteristic information of the user's hobbies and/or group attribute characteristics, and at the same time according to the user's current geographical location and current time list information for .

Specifically, the application recommendation service platform can determine the list information of the applications to be recommended to the user based on the user's hobbies and interests, as well as the user's current geographical location and current time. The application recommendation service platform detects that the user is currently in the evening time period and is located in the residential area, and the determined application list information to be recommended to the user includes various game applications.

The application recommendation service platform can also determine the list information of the applications to be recommended to the user based on the user's crowd attributes, as well as the user's current geographic location and current time. For example, the user's crowd attributes show that the user is an office worker, When the application recommendation service platform detects that the user is currently in the work area and is currently working, the determined list of applications to be recommended to the user includes tool applications such as news, weather, investment software, office software, and chat software. For another example, if the user's group attribute characteristics show that the user is an office worker, and the application recommendation service platform detects that the user is currently in a work area or a business area, and is currently in a rest period, the determined list information of the applications to be recommended to the user includes Service applications such as takeaway, meal delivery, and restaurant recommendations. For another example, if the user's group attribute characteristics show that the user is an office worker, and the application recommendation service platform detects that the user is currently located in the living area, and it is currently in the evening time period, the determined list information of the application to be recommended to the user includes video, Music, games, Taobao and other entertainment and shopping applications. Users can be divided into groups, users' time can be segmented, locations can be divided into functions, and context-aware recommendation strategies can be formulated. Examples are as follows:

Table 2 List information of applications to be recommended to users

The application recommendation service platform can also determine the list information of applications to be recommended to the user according to the user's hobbies and group attributes, as well as the user's current geographic location and current time. For example, the user's hobbies feature shows that the user should be more interested in finance and music, the user's crowd attribute feature shows that the user is an office worker, and the application recommendation service platform detects that the user is currently located in the living area, and it is currently in the evening time period , the determined list information of applications to be recommended to the user includes financial and music applications.

In this embodiment, the application recommendation service platform determines the list information of applications to be recommended to the user according to the user's hobbies or/and crowd attribute characteristics, the user's current geographical location and the current time. According to each user's own characteristics, and according to the current time and location, the application recommendation list is specified, so as to accurately recommend applications that meet their interests and take into account their current location and time information, and satisfy different users at different times and in different places. For the individual needs of the application.

Further, on the basis of the above embodiments, in step 103, the application list information is sent to the user, including:

After receiving the recommendation request sent by the user, the application recommendation service platform sends the list information of the application to the user;

or,

After the user logs in to the application recommendation service platform, the application recommendation service platform sends the application list information to the user;

or,

The application recommendation service platform sends the application list information to the user within a preset time period.

In this embodiment, specifically, the application recommendation service platform sends the determined application list information to the user. The user can send a recommendation request to the application recommendation service platform, and the application recommendation service platform will send the list information of the applications analyzed in real time to the user, or the application recommendation service platform will send the list information of the pre-determined applications to the user.

It may also be that the user logs in to the application recommendation service platform, and then the application recommendation service platform sends the list information of the determined applications to the user, or after the user logs in to the application recommendation service platform, the application recommendation service platform sends the determined applications in real time The list information is sent to the user.

It may also be that the user specifies a first push schedule, the first push schedule includes various time periods, and the application recommendation service platform sends the application list information to the user within each preset time period,

It may also be that the application recommendation service platform specifies a second push schedule, and the second push schedule includes various time periods, and the application recommendation service platform sends the determined information to the user in different time periods according to the user's current location and time. Apps that rank higher in the app's listing information.

Alternatively, the application recommendation service platform may send applications with higher ranks in the determined application list information to users with higher leisure index within different time periods.

FIG. 3 is a schematic structural diagram of the application recommendation service platform provided by Embodiment 3 of the present invention. As shown in FIG. 3 , the application recommendation service platform provided by this embodiment includes:

The data acquisition module 31 is used to collect and acquire the user's application data from the operator's server;

The data mining module 32 is used to perform data mining analysis on the application data to obtain user characteristic information;

The application recommendation module 33 is configured to determine the list information of the applications to be recommended to the user according to the characteristic information, and send the list information of the applications to the user.

The application recommendation service platform in this embodiment can execute the application recommendation method provided in Embodiment 1 of the present invention, and its implementation principles are similar, so details will not be repeated here.

In this embodiment, the application recommendation service platform is used to perform data mining analysis on the user's application data obtained from the operator's server to obtain the user's characteristic information, and then determine the list information of the application to be recommended to the user according to the characteristic information. And send the list information of the application to the user. In this way, by analyzing the user's own characteristics, the application recommendation list is sent to the user according to the user's own needs, so as to meet the personalized needs of different users for the application.

FIG. 4 is a schematic structural diagram of the application recommendation service platform provided by Embodiment 4 of the present invention. On the basis of Embodiment 3, as shown in FIG. 4 , in the application recommendation service platform provided by this embodiment, the data mining module 32 includes:

The data mining sub-module 321 is used to perform data mining analysis on the application data to obtain the user's hobbies and/or crowd attribute characteristics;

Correspondingly, the application recommendation module 33 includes:

The application determination sub-module 331 is used to determine the list information of applications to be recommended to the user according to the characteristics of hobbies and/or attributes of the crowd;

The application sending sub-module 332 is configured to send the application list information to the user.

The data acquisition module 31 is specifically used to: collect and acquire the user's access business type, access duration, access frequency, and access traffic corresponding to each application from the operator's server;

Correspondingly, when the data mining sub-module 321 is used to perform data mining analysis on the application data to obtain the characteristics of the user's hobbies, it is specifically used for:

According to the access duration S _i , access frequency F _i and access traffic M _i corresponding to each application, determine the score A _i of each application = Q ₁ S _i /S _{total i} +Q ₂ F _i /F _{total i} +Q ₃ M _i /M _{total i} , wherein, Q ₁ , Q ₂ and Q ₃ are different first preset weight values, S _{total i} , F _{total i} and M _{total i} are the total access time and total access time of each application respectively. Frequency and total access traffic, i∈[1,N], i is a positive integer, N is a positive integer;

According to the score A _i of each application and the access service type corresponding to each application, determine the interest index of each application belonging to the same access service type Among them, b∈[1,B], b is a positive integer, and B is a positive integer;

Determine the basic recommendation index P _i of each application according to each interest index I _b and the pre-acquired scores of each application;

Determine each similarity S _ij between each application and the application in the mobile terminal currently used by the user, where j∈[1,J], j is a positive integer, and J is a positive integer;

According to each similarity S _ij , adjust the basic recommendation index P _i '={1-(S _i1 -S _threshold )}P _i Q ₁ '+...+{1-(S _ij -S _threshold )}P _i Q _i '+...+{1-(S _iJ -S _threshold )}P _i Q' _N , where the S _{threshold is} the preset similarity threshold, and Q _i 'is the second preset weight value;

According to the adjusted basic recommendation index, each application is sorted to obtain an interest recommendation list, and the interest recommendation list is used to characterize the user's hobbies and interests.

The data acquisition module 31 is specifically used to: acquire the user's access start time, access end time, and access base station corresponding to each application from the operator's server;

Correspondingly, when the data mining sub-module 321 is used to perform data mining analysis on the application data to obtain user group attributes, it is specifically used for:

According to the access start time and access end time corresponding to each application, determine the proportion of the user's daily online time;

According to the access base station corresponding to each application, determine the user's daily movement trajectory;

According to the user's daily online time ratio and daily movement trajectory, the user's crowd attribute characteristics are determined.

The application determination submodule 331 is specifically used for:

According to the feature information, the current geographic location of the user, and the current time, list information of applications to be recommended to the user is determined.

The application sending submodule 332 is specifically used for:

After receiving the recommendation request sent by the user, send the list information of the application to the user; or, after the user logs in to the application recommendation service platform, send the list information of the application to the user; or, within a preset time period, send the application list information to the user; The list information of the application is sent to the user.

The application recommendation service platform in this embodiment can execute the application recommendation method provided in Embodiment 2 of the present invention and the above-mentioned implementation manners, and the implementation principles thereof are similar, and will not be repeated here.

In this embodiment, the application recommendation service platform determines the list information of applications to be recommended to the user according to the user's hobbies and/or group attribute characteristics. In this way, by analyzing the user's own characteristics, the application recommendation list is sent to the user according to the user's own needs, so as to meet the personalized needs of different users for the application. In addition, in this embodiment, the application recommendation service platform determines the list information of applications to be recommended to the user according to the user's hobbies or/and crowd attribute characteristics, the user's current geographical location and the current time. According to each user's own characteristics, and according to the current time and location, the application recommendation list is specified, so as to accurately recommend applications that meet their interests and take into account their current location and time information, and satisfy different users at different times and in different places. For the individual needs of the application.

Those of ordinary skill in the art can understand that all or part of the steps for implementing the above method embodiments can be completed by program instructions and related hardware. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above-mentioned method embodiments; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other various media that can store program codes.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims (8)

1. An application recommendation method, comprising:

the application recommendation service platform acquires and acquires application data of a user from an operator server;

the application recommendation service platform carries out data mining analysis on the application data to obtain the characteristic information of the user;

the application recommendation service platform determines list information of applications to be recommended to the user according to the characteristic information and sends the list information of the applications to the user;

the method for acquiring the application data of the user from the operator server by the application recommendation service platform comprises the following steps:

the application recommendation service platform acquires and acquires access service types, access duration, access frequency and access flow of the user corresponding to each application from an operator server;

correspondingly, the data mining analysis is performed on the application data by the application recommendation service platform to obtain the interest and hobby characteristics of the user, and the method comprises the following steps:

the application recommendation service platform is used for recommending the service according to the access duration S corresponding to each application_iFrequency of access F_iAnd access traffic M_iDetermining the score A of each application_i＝Q₁S_i/S_{Total of i}+Q₂F_i/F_{Total of i}+Q₃M_i/M_{Total of i}Wherein Q is₁、Q₂And Q₃Are different first preset weight values, S_{Total of i}、F_{Total of i}And M_{Total of i}Respectively providing total access time length, total access frequency and total access flow for the user to access each application, wherein i belongs to [1, N ∈]I is a positive integer, and N is a positive integer;

the application recommendation service platform scores A according to each application_iAnd access service types corresponding to the applications, determining interest indices of the applications belonging to the same access service typeWherein B is ∈ [1, B ]]B is a positive integer, B is a positive integer;

the application recommendation service platform is used for recommending the service according to the interest indexes I_bAnd pre-obtaining the scores of all the applications to determine the basic recommendation index P of all the applications_i；

The application recommendation service platform determines each similarity S between each application and the application in the mobile terminal currently used by the user_ijWherein J is ∈ [1, J ∈ ]]J is a positive integer, J is a positive integer;

the application recommendation service platform recommends the service platform according to the similaritiesDegree S_ijAdjusting the basic recommendation index P of each application_i'＝{1-(S_i1-S_{Threshold value})}P_iQ₁'+…+{1-(S_ij-S_{Threshold value})}P_iQ_i'+…+{1-(S_iJ-S_{Threshold value})}P_iQ'_NWherein S is_{Threshold value}To preset a similarity threshold, Q_i' is a second preset weight value;

and the application recommendation service platform sequences all applications according to the adjusted basic recommendation index to obtain an interest recommendation list, wherein the interest recommendation list is used for representing the interest and hobby characteristics of the user.

2. The method of claim 1, wherein the data mining analysis of the application data by the application recommendation service platform to obtain the feature information of the user comprises:

the application recommendation service platform carries out data mining analysis on the application data to obtain interest and hobby characteristics and/or crowd attribute characteristics of the user;

correspondingly, the determining, by the application recommendation service platform, the list information of the applications to be recommended to the user according to the feature information includes:

and the application recommendation service platform determines list information of the applications to be recommended to the user according to the interest and hobby characteristics and/or the crowd attribute characteristics.

3. The method of claim 2, wherein the application recommendation service platform collects and obtains application data of the user from an operator server, and comprises:

the application recommendation service platform acquires and acquires access start time, access end time and access base station of a user corresponding to each application from an operator server;

correspondingly, the data mining analysis is performed on the application data by the application recommendation service platform to obtain the crowd attribute characteristics of the user, and the method comprises the following steps:

the application recommendation service platform determines the daily internet access time proportion of the user according to the access start time and the access end time corresponding to each application;

the application recommendation service platform determines the daily motion track of the user according to the access base station corresponding to each application;

and the application recommendation service platform determines the crowd attribute characteristics of the user according to the daily internet surfing time proportion and the daily motion trail of the user.

4. The method according to claim 2 or 3, wherein the determining, by the application recommendation service platform, the list information of the applications to be recommended to the user according to the feature information comprises:

the application recommendation service platform determines list information of applications to be recommended to the user according to the feature information, the current geographic position of the user and the current time;

the sending the list information of the application to the user includes: after receiving a recommendation request sent by a user, the application recommendation service platform sends the list information of the application to the user;

or after a user logs in an application recommendation service platform, the application recommendation service platform sends the list information of the application to the user;

or, the application recommendation service platform sends the list information of the application to the user within a preset time period.

5. An application recommendation service platform, comprising:

the data acquisition module is used for acquiring and acquiring application data of a user from an operator server;

the data mining module is used for carrying out data mining analysis on the application data to obtain the characteristic information of the user;

the application recommending module is used for determining list information of the applications to be recommended to the user according to the characteristic information and sending the list information of the applications to the user;

the data acquisition module is specifically configured to:

acquiring access service types, access durations, access frequencies and access flow of users corresponding to all applications from an operator server;

correspondingly, the data mining submodule, when being used for performing data mining analysis on the application data to obtain the interest and hobby characteristics of the user, is specifically used for:

according to the access duration S corresponding to each application_iFrequency of access F_iAnd access traffic M_iDetermining the score A of each application_i＝Q₁S_i/S_{Total of i}+Q₂F_i/F_{Total of i}+Q₃M_i/M_{Total of i}Wherein Q is₁、Q₂And Q₃Are different first preset weight values, S_{Total of i}、F_{Total of i}And M_{Total of i}Respectively providing total access time length, total access frequency and total access flow for the user to access each application, wherein i belongs to [1, N ∈]I is a positive integer, and N is a positive integer;

score A according to each application_iAnd access service types corresponding to the applications, determining interest indices of the applications belonging to the same access service typeWherein B is ∈ [1, B ]]B is a positive integer, B is a positive integer;

according to the interest indexes I_bAnd pre-obtaining the scores of all the applications to determine the basic recommendation index P of all the applications_i；

Determining similarity S between each application and the application in the mobile terminal currently used by the user_ijWherein J is ∈ [1, J ∈ ]]J is a positive integer, J is a positive integer;

according to the similarity S_ijAdjusting the basic recommendation index P of each application_i'＝{1-(S_i1-S_{Threshold value})}P_iQ₁'+…+{1-(S_ij-S_{Threshold value})}P_iQ_i'+…+{1-(S_iJ-S_{Threshold value})}P_iQ'_NWherein S is_{Threshold value}To preset a similarity threshold, Q_i' is a second preset weight value;

and sequencing the applications according to the adjusted basic recommendation index to obtain an interest recommendation list, wherein the interest recommendation list is used for representing the interest and hobby characteristics of the user.

6. The application recommendation service platform according to claim 5, wherein said data mining module comprises:

the data mining submodule is used for carrying out data mining analysis on the application data to obtain the interest and hobby characteristics and/or the crowd attribute characteristics of the user;

correspondingly, the application recommendation module comprises:

the application determining submodule is used for determining list information of applications to be recommended to the user according to the interest and hobby characteristics and/or the crowd attribute characteristics;

and the application sending submodule is used for sending the list information of the application to the user.

7. The application recommendation service platform of claim 6, wherein the data acquisition module is specifically configured to:

acquiring access start time, access end time and access base station corresponding to each application of a user from an operator server;

correspondingly, the data mining submodule, when being used for performing data mining analysis on the application data to obtain the crowd attribute characteristics of the user, is specifically used for:

determining the proportion of the internet surfing time of the user every day according to the access starting time and the access ending time corresponding to each application;

determining the movement track of the user every day according to the access base station corresponding to each application;

and determining the crowd attribute characteristics of the user according to the daily internet surfing time proportion and the daily motion trail of the user.

8. The application recommendation service platform according to claim 6 or 7, wherein the application determination submodule is specifically configured to:

determining list information of applications to be recommended to the user according to the feature information, the current geographical position of the user and the current time;

the application sending submodule is specifically configured to: after a recommendation request sent by a user is received, sending the list information of the application to the user;

or after a user logs in an application recommendation service platform, the list information of the application is sent to the user;

or, within a preset time period, sending the list information of the application to the user.