WO2016180182A1

WO2016180182A1 - Service package recommendation method and device

Info

Publication number: WO2016180182A1
Application number: PCT/CN2016/079668
Authority: WO
Inventors: 丁岩; 吴超
Original assignee: 中兴通讯股份有限公司
Priority date: 2015-10-30
Filing date: 2016-04-19
Publication date: 2016-11-17
Also published as: CN106658434A

Abstract

Provided is a service package recommendation method. The method comprises the following steps: upon receiving basic information of terminals and service information of application program in the terminals transmitted by at least two terminals, establishing service models according to the basic information of the terminals and the service information of the application program; calculating, according to a preset algorithm, similarity between the service models, and assigning the service models having the similarity exceeding a preset threshold into a same group of the service models; and calculating a distance between different groups of the service models and a target point of interest, and determining, according to the distance, a recommended policy of service packages of the different groups of the service models. Also provided is a service package recommendation device. The above solution recommends, according to information such as a service direction of user traffic and a bandwidth occupancy condition, a corresponding service package to a user.

Description

Recommended method and device for business package

Technical field

The present application relates to, but is not limited to, the field of communication technologies, and in particular, to a method and apparatus for recommending a service package.

Background technique

With the explosive growth of mobile terminal applications, the network pressure of operators' servers continues to increase, and the traditional services are gradually shrinking. With the decline in operating income, carrier servers are gradually becoming the makers of OTT (Over The Top). pipeline. Therefore, the implementation of refined traffic management, mining data business value, leading the development of mobile information technology has become an urgent task for major operators. At present, the operator server can only grasp how much traffic each user uses. As for which application the data traffic is contributed to, it can only be indirectly judged by the download amount of the application of the operator's application store, thereby causing the operator server. It is not possible to customize a targeted traffic plan and quality of service package for users by locating the specific destination of the user's usage of the application.

The above content is only used to assist in understanding the technical solutions of the present application, and does not constitute an admission that the above is prior art.

Summary of the invention

The following is an overview of the topics detailed in this document. This Summary is not intended to limit the scope of the claims.

The main purpose of the present application is to provide a method and a device for recommending a service package, which is to solve the problem that the related technology cannot recommend a corresponding service package for the user according to the usage direction of the user traffic and the bandwidth occupation situation.

To achieve the above objective, the method for recommending a service package provided by the present application includes the following steps:

When receiving the basic information of the terminal and the usage information of the application in the terminal sent by the at least two terminals, establishing a usage model according to the basic information of the terminal and the usage information of the application;

Calculating the similarity between the usage models according to a preset algorithm, and dividing the usage model with the similarity greater than a preset threshold into the same group usage model;

Calculating the distance between the usage model of the different groups and the target interest point, and determining the recommendation strategy of the business package of the different group using the model according to the distance.

Optionally, the step of calculating a distance between the usage model of the different groups and the target interest point, and determining the recommendation policy of the service package of the different group using the model according to the distance includes:

Determining the coordinates of the used models of different groups and the coordinates of the target points of interest in the coordinate system;

Calculating a distance between the usage model of the different groups and the target interest point according to the coordinates of the use model of the different groups and the coordinates of the target interest point in the coordinate system;

Determining a recommendation policy of the business package of the different group of usage models according to the size of the distance.

Optionally, when the basic information of the terminal and the usage information of the application in the terminal that are sent by the at least two terminals are received, the basic information of the terminal and the usage information of the application are used. The steps of the model include:

Receiving basic information of the terminal sent by at least two terminals and usage information of an application in the terminal, and receiving latitude and longitude data of the terminal sent by the location service based server, according to the basic of the terminal The information, the usage information of the application, and the latitude and longitude data of the terminal are used to establish a usage model.

Optionally, the calculating the similarity between the usage models according to a preset algorithm, and dividing the usage model with the similarity greater than the preset similarity into the same group usage model includes:

Calculating the similarity between the usage models in the same cell according to a preset algorithm, and dividing the usage model in the same cell with the similarity greater than the preset similarity into the same group usage model, to obtain the usage model. The result of the grouping.

Obtaining an analysis result according to the grouping result of the usage model in the same cell and the usage model analyzing the network condition of the cell;

Sending the analysis result to the operator server for the operator server to The analysis result performs a corresponding operation on the cell.

Optionally, the step of analyzing the network status of the cell according to the grouping result of the usage model in the same cell and the usage model, and the step of obtaining an analysis result includes:

Determining whether a signal strength of the terminal corresponding to the usage model in the same group usage model in the same cell is smaller than a preset signal strength threshold;

When the signal strength of the terminal in the same cell is less than the preset signal strength threshold, and the proportion of the terminal in the same cell that is smaller than the preset signal strength threshold exceeds a preset ratio, determining The cell is in a network blind zone;

When the signal strength of the terminal in the same cell is greater than or equal to the preset signal strength threshold, determining that the cell is in a network normal state.

Optionally, the preset algorithm is an algorithm combining a collaborative filtering algorithm and a cosine similarity.

Optionally, the basic information of the terminal includes an international mobile subscriber identity, an operating system version number and vendor information of the terminal, and a list of applications installed by the terminal;

The usage information of the application includes the number of startups of the application, the running time of the application, traffic consumption data, occupied bandwidth data, the type of network used by the application, and network signal strength.

In addition, in order to achieve the above object, the present application further provides a recommendation device for a service package, where the recommendation device of the service package includes:

Establishing a module, configured to establish a usage model according to the basic information of the terminal and the usage information of the application when receiving basic information of the terminal sent by at least two terminals and usage information of the application in the terminal ;

a first calculating module, configured to calculate a similarity between the usage models according to a preset algorithm, and divide the usage model with the similarity greater than a preset threshold into a same group usage model;

The second calculating module is configured to calculate a distance between the usage model of the different groups and the target interest point, and determine a recommendation strategy of the service package of the different group using the model according to the distance.

Optionally, the second calculating module includes:

a first determining unit configured to determine coordinates of a different group of usage models and coordinates of the target points of interest in the coordinate system;

a calculating unit, configured to calculate a distance between the usage model of the different group and the target interest point according to the coordinates of the usage model of the different groups and the coordinates of the target interest point in the coordinate system;

The second determining unit is configured to determine a recommendation policy of the service package of the different group of usage models according to the size of the distance.

Optionally, the establishing module is configured to: when receiving basic information of the terminal sent by at least two terminals and usage information of an application in the terminal, and receiving the sending by the location service based server When the latitude and longitude data of the terminal is used, the usage model is established based on the basic information of the terminal, the usage information of the application, and the latitude and longitude data of the terminal.

Optionally, the first calculating module is configured to calculate, according to a preset algorithm, a similarity between the usage models in the same cell, and the similarity in the same cell is greater than a preset similarity. Using the model to divide into the same set of usage models, the grouping results of the used models are obtained.

Optionally, the recommended device of the service package further includes:

The analyzing module is configured to analyze the network status of the cell according to the grouping result of the usage model in the same cell and the usage model, and obtain an analysis result;

And a sending module, configured to send the analysis result to the operator server, where the operator server performs a corresponding operation on the cell according to the analysis result.

Optionally, the analyzing module includes:

a determining unit, configured to determine whether a signal strength of the terminal corresponding to the usage model in the same group usage model in the same cell is smaller than a preset signal strength threshold;

a first determining unit, configured to: when the signal strength of the terminal in the same cell is less than the preset signal strength threshold, and the ratio of the terminal in the same cell that is smaller than the preset signal strength threshold When the preset ratio is exceeded, it is determined that the cell is in a network blind zone;

The second determining unit is configured to determine that the cell is in a network normal state when the signal strength of the terminal in the same cell is greater than or equal to the preset signal strength threshold.

The application further provides a computer readable storage medium storing computer executable instructions that are implemented when the computer executable instructions are executed.

The application establishes a usage model according to the basic information of the at least two terminals and the usage information of the application, calculates a similarity between the usage models, and divides the usage model with the similarity greater than a preset threshold. The model is used for the same group, and the distance between the usage model of the different groups and the target interest point is calculated, and the recommendation strategy of the business package of the different group using the model is determined according to the distance. The basic information of the terminal used by the user and the usage information of the application in the terminal are collected on the Android and iOS terminal platforms. According to the collected information, the usage model is established by using a preset algorithm to obtain the traffic destination of the user using the application. And data such as bandwidth occupation, determining, according to the data, a traffic policy and/or a recommendation policy of the QoS package of the usage model of the different group, and the operator server uses the traffic plan and/or QoS according to the usage model of the different group. The recommendation strategy of the package is to provide a data package and/or QoS package for the user, and provide data basis for cooperation between the operator server and the OTT manufacturer.

Other aspects will be apparent upon reading and understanding the drawings and detailed description.

BRIEF abstract

1 is a schematic flowchart of a first embodiment of a method for recommending a service package according to the present application;

2 is a schematic flowchart of an optional step of a first embodiment of a method for recommending a service package according to the present application;

3 is a schematic flowchart of a second embodiment of a method for recommending a service package according to the present application;

FIG. 4 is a flowchart of analyzing the network condition of the cell according to the grouping result of the usage model in the same cell and the usage model according to an embodiment of the present invention; schematic diagram;

5 is a schematic diagram of functional modules of a first embodiment of a device for recommending a service package according to the present application;

6 is a schematic diagram of a functional module of a second computing module according to an embodiment of the present invention;

7 is a schematic diagram of functional modules of a second embodiment of a device for recommending a service package according to the present application;

FIG. 8 is a schematic diagram of a functional module of an analysis module according to an embodiment of the present invention.

Preferred embodiment of the invention

The application provides a method of recommending a business package.

Referring to FIG. 1, FIG. 1 is a schematic flowchart diagram of a first embodiment of a method for recommending a service package according to the present application.

In this embodiment, the recommended method of the service package includes:

Step S10, when receiving the basic information of the terminal and the usage information of the application in the terminal sent by the at least two terminals, establishing a usage model according to the basic information of the terminal and the usage information of the application;

When the server receives the basic information of the terminal and the usage information of the application in the terminal, the server establishes a usage model according to the basic information of the terminal and the usage information of the application. The usage model includes a data usage model of the end user and a bandwidth usage model of the end user. Each end user has a corresponding data usage model and bandwidth usage model. The server is a server capable of accurately analyzing the end user data, and the server may exist in a cloud or a physical device. The minimum configuration of the server is: Intel Xeon E5 six-core processor Intel Xeon E5-2620, 8GB (Gigabyte, gigabytes) of memory and 2TB (Terabyte, terabyte) data disk. The terminal includes, but is not limited to, a smartphone and a tablet. When the terminal is an Android phone or an Android tablet, the minimum configuration of the Android phone or the Android tablet is an ARM (Advanced RISC Machines) architecture CPU (Central Processing Unit), 512 MB (MByte) The RAM, 1GB ROM (Read-Only Memory) and output device are capacitive touch screens with a resolution of 460*640. When the terminal is Ping When the mobile phone or the Apple tablet is used, the minimum configuration of the Apple mobile phone or the Apple tablet is: ARM Cortex-A8 CPU, 256 MB of RAM, 8 GB of ROM and output device is a capacitive touch screen with a resolution of 480*320.

The basic information of the terminal includes an international mobile subscriber identity, an operating system version number and vendor information of the terminal, and a list of applications installed by the terminal. In this embodiment, the terminal of the Android operating system is taken as an example for description. The terminal starts the background information collection service of the APP (Application) after receiving the boot broadcast android.intent.action.BOOT_COMPLETED. The terminal acquires the IMSI (International Mobile Subscriber Identification Number) of the terminal by using the information acquisition service (Application Programming Interface) of the Android (TelephonyManager.getSubscriberId). The unique identification information of the message push; call android.os.Build.VERSION.RELEASE to obtain the version number of the terminal operating system; call android.os.Build.MODEL to obtain the model of the terminal; call android.os.Build.MANUFACTURER Obtain the manufacturer of the terminal; call PackageManager.getInstalledPackages to obtain information about all the apps installed in the terminal. And traversing the APP installation information parameter one by one, if the terminal detects that the value of packageInfo.applicationInfo.flags&ApplicationInfo.FLAG_SYSTEM is 0, the APP is marked as a non-system pre-installed APP, and if the terminal detects the packageInfo If the value of .applicationInfo.flags&ApplicationInfo.FLAG_SYSTEM is 1, the APP is marked as an APP pre-installed by the system, and the installation list information of the APP in the terminal is obtained.

The usage information of the application includes the number of startups of the application, the running time of the application, the traffic consumption data, the occupied bandwidth data, the type of network used by the application, and the network signal strength. The terminal, according to the installation list information of the APP, reflects com.android.internal.os.PkgUsageStats, and reads the PkgUsageStats.launchCount and PkgUsageStats.usageTime attributes respectively, to obtain the startup times and running times of different APPs, that is, The number of startups and running time of the different applications of the terminal; the terminal acquires the current network type and signal strength through the TelephonyManager.subtype, and writes the current network type and the signal strength into the sqllite database.

Step S20: Calculate the similarity between the usage models according to a preset algorithm, and divide the usage model whose similarity is greater than a preset threshold into the same group usage model;

The server calculates the similarity between the usage models according to a preset algorithm, and divides the usage model with the similarity greater than a preset threshold into the same group usage model, that is, calculates the interaction between the terminal users according to a preset algorithm. The similarity between the data traffic usage model and/or the bandwidth usage model, and the data traffic usage model in which the data traffic usage model of the end user is greater than a preset threshold is divided into the same group usage model, that is, the The similarity of the data traffic usage model is greater than the preset data traffic. The terminal users corresponding to the model similarity threshold are divided into the same group, and/or the similarity of the bandwidth usage model between the terminal users is greater than a preset. The data traffic using the threshold of the model similarity is divided into the same group usage model, that is, the terminal users whose similarity of the bandwidth usage model is greater than the threshold of the preset bandwidth usage model similarity are divided into the same group, and / Or comparing the similarity of the bandwidth usage model between the end users to a preset bandwidth usage model phase Data traffic using the threshold value of the model into the model using the same group, i.e. the bandwidth usage model similarity is larger than a preset dividing end-user bandwidth usage model similarity threshold value corresponding to the same group.

The preset algorithm is an algorithm combining a collaborative filtering algorithm and a cosine similarity, and the collaborative filtering algorithm is a main algorithm of the e-commerce recommendation system, and the collaborative filtering algorithm filters and analyzes user interests in the user group. Finding users of similar users (interests), synthesizing the evaluation of certain information by these similar users, and forming a prediction of how much the e-commerce recommendation system likes the designated users. Compared with the traditional text filtering algorithm, the collaborative filtering algorithm can filter the information of the machine based on content analysis automatically, and can filter based on some complicated and difficult to express concepts (information quality, taste), with recommended novelty. . The cosine similarity measures the similarity between them by measuring the cosine of the angle between the product spaces in the two vectors. The cosine of the 0 degree angle is 1, while the cosine of any other angle is not greater than 1; and its minimum is -1. Thus the cosine of the angle between the two vectors determines whether the two vectors generally point in the same direction. When two vectors have the same pointing, the value of cosine similarity is 1; when the angle between two vectors is 90°, the value of cosine similarity is 0; when the two vectors point to the opposite direction, the value of cosine similarity Is -1. In the comparison process, the size of the vector is not considered, only the direction of the vector is considered. Cosine similarity is usually used when the angle between two vectors is less than 90°, so the value of cosine similarity is 0 to 1. between.

The preset threshold may be set according to a specific situation. When more terminal users need to be divided into a group, the preset threshold may be set smaller, such as 80%, when more precise division is needed. When the terminal user is used, the preset threshold may be set to be larger, such as 90%. The preset data traffic usage threshold of the model similarity and the threshold of the bandwidth usage model similarity may be the same or different. For example, when the preset data traffic usage threshold of the model similarity and the threshold of the bandwidth usage model similarity are both 90%, the server compares the usage model between the end users by more than 90. The usage model of % is divided into the same group usage model. For example, when the similarity between the usage model of the A terminal user and the usage model of the B terminal user is greater than 90%, the server divides the usage model of the A terminal user and the usage model of the B terminal user into the same group. Use the model.

Step S30, calculating a distance between the usage model of the different groups and the target interest point, and determining a recommendation strategy of the service package of the different group using the model according to the distance.

The server calculates a distance between the usage model of the different groups and the target interest point, and determines a recommendation strategy of the service package of the different group using the model according to the distance. The target point of interest is an application that consumes data traffic in the terminal, such as a social application and a game application.

Referring to FIG. 2, FIG. 2 is a schematic flowchart diagram of an optional step of a first embodiment of a method for recommending a service package according to the present application.

Optionally, the step S30 includes:

Step S35, determining coordinates of different use models of the different groups and coordinates of the target interest points in the coordinate system;

Step S37, calculating distances between the usage models of the different groups and the target interest points according to the coordinates of the usage models of the different groups and the coordinates of the target interest points in the coordinate system;

Step S39: Determine a recommendation policy of the service package of the different group of usage models according to the size of the distance.

The server determines coordinates of different groups of usage models in a three-dimensional coordinate system and coordinates of target interest points in a three-dimensional coordinate system, and the server uses coordinates of the different groups according to the usage model Calculating a distance between the usage model of the different group and the target interest point according to the coordinates of the target interest point in the coordinate system, and determining the distance according to the usage model of the different group and the distance before the target interest point The recommendation strategy of the different groups of business packages using the model. When the distance between the usage model of the different group and the target interest point is smaller, the recommended degree value of the target interest point is higher, and when the recommendation degree value is higher, the different group is When using the model for business package recommendation, first recommend the business package corresponding to the high value. For example, the server calculates that the distance between the first data traffic usage model and the social application WeChat is 0.2, the distance from the video playback application Youku is 0.18, and the distance between the second data traffic usage model and the social application WeChat is 0.15. The distance from the Youku is 0.22, and the server determines, according to the distance between the first group of data traffic usage models and the second group of data traffic usage models and the WeChat and the Youku. The first set of data traffic usage models and the WeChat of the second set of data traffic usage models and the recommended value of the Youku, in the first set of data traffic usage models, the value of the recommendation of the Youku is greater than The recommendation value of the WeChat, in the second group data usage model, the recommendation value of the Youku is smaller than the recommendation value of the WeChat, and the server determines the corresponding service package according to the recommendation value. Recommended strategy.

Optionally, the server sends, to the operator server, a recommended policy of the service group of the different group of usage models, by the operator server, according to the OSS (Operation Support System) The recommendation strategy of the business package of the different groups using the model is to formulate corresponding business packages for the terminals corresponding to the usage models of the different groups. The service package includes a data flow package, a QoS (quality of service) package, and a combination of data traffic and QoS. And the operator server determines, for the terminal user corresponding to the first group of data traffic usage models, that the data traffic applied to the application for video playback is greater than the data traffic of the social software, and is used by the second group of data traffic. The end user corresponding to the model formulates a package whose data traffic applied to the video playback application is smaller than the data traffic of the social software.

In this embodiment, the usage model is established according to the basic information of the at least two terminals and the usage information of the application, and the similarity between the usage models is calculated, and the usage model with the similarity greater than a preset threshold is used. Dividing into the same group of usage modalities, and calculating the distance between the usage models of the different groups and the target interest points, and determining the recommendation strategies of the service packages of the different groups using the models according to the distances. Realize the collection of terminals used by users on Android and iOS terminal platforms The basic information and the usage information of the application in the terminal, according to the collected information, establish a usage model by using a preset algorithm to obtain data such as traffic direction and bandwidth occupation of the user using the application, and determine the difference according to the data. The group's use of the model's traffic plan and / or QoS package recommendation strategy, the operator server to set the data package and / or QoS package for the user according to the different group of usage model of the traffic plan and / or QoS package recommendation policy Provide data basis for cooperation between carrier servers and OTT vendors.

Referring to FIG. 3, FIG. 3 is a schematic flowchart of a second embodiment of a method for recommending a service package according to the present application. The second embodiment of the present application is based on the first embodiment of the service package establishment method of the present application.

In this embodiment, optionally, the step S10 includes:

Step S11, when receiving basic information of the terminal sent by at least two terminals and usage information of an application in the terminal, and receiving latitude and longitude data of the terminal sent by the location service based server, according to the The basic information of the terminal, the usage information of the application, and the latitude and longitude data of the terminal establish a usage model.

When the server receives the basic information of the terminal sent by at least two terminals and the usage information of the application in the terminal, and receives the latitude and longitude data of the terminal sent by the location service based server, the server A usage model is established based on basic information of the terminal, usage information of the application, and latitude and longitude data of the terminal. The usage model includes the end user network type model, a signal strength model, a user geographic location information model, and the terminal application usage model. Each end user has a corresponding network type model, a signal strength model, a user geographic location information model, and the terminal application usage model. The network type model refers to whether the end user uses a 2G (2nd-Generation wireless telephone technology) network, or a 3G (3rd-Generation, 3rd generation mobile communication technology) network or 4G. (the 4th Generation mobile communication technology) network.

The location information is obtained by the terminal by using a method of calling the TelephonyManager.getCellLocation().getCid() and the TelephonyManager.getCellLocation().getLac() to obtain the identification information of the location where the cell where the terminal is located and the LAC (location area) Code, location area code) data. The terminal sets the identification information of the location where the cell is located and The LAC data is reported to the LBS (Location Based Service) server in the cloud, so that the LBS server obtains the latitude and longitude data of the location of the terminal according to the identifier information of the location of the cell and the LAC data. And transmitting latitude and longitude data of the location of the terminal to the server. The minimum configuration of the LBS server is: Xeon E3-1230v3 CPU; 8 GB of memory; and 1 TB data disk, the LBS server may exist in the cloud or may exist in the physical device. After the server receives the latitude and longitude data sent by the LBS, the geographic location of the terminal is determined according to the latitude and longitude data, and the geographic location of the terminal is written into the sqllite database.

Optionally, the step S20 includes:

Step S21: Calculate the similarity between the usage models in the same cell according to a preset algorithm, and divide the usage model in which the similarity in the same cell is greater than the preset similarity into the same group usage model, and obtain the The result of the grouping using the model.

And the server calculates, according to the algorithm that is combined with the cosine filtering algorithm and the cosine similarity, a similarity between the usage models in the same cell, and the similarity of the same cell is greater than a preset similarity. The model is divided into the same group of users to use the model, and the grouping result of the usage model is obtained. For example, the server divides the network type model corresponding to the terminal users in the same cell using the 2G network into a group, and divides the network type model corresponding to the terminal users using the 3G network into a group, and uses the terminal of the 4G network. The network type models corresponding to users are divided into a group.

Optionally, the step 30 includes:

Step S31, analyzing network conditions of the cell according to the grouping result of the usage model in the same cell and the usage model, and obtaining an analysis result;

The server analyzes the network status of the cell according to the grouping result of the usage model in the same cell and the usage model, and obtains an analysis result. The server analyzes the status of the network type of the terminal user corresponding to the 3G in the same cell, and combines the network type model with the signal strength model of the terminal user corresponding to the 3G, and combines the network The type model is a geographical location information model of the end user corresponding to the 3G, and the network type model is determined to be the network status of the cell where the terminal user corresponding to the 3G is located.

Step S32, sending the analysis result to the operator server for the operator The server performs a corresponding operation on the cell according to the analysis result.

The server sends the analysis result to the operator server, where the operator server receives the analysis result, and performs a corresponding operation on the cell according to the analysis result. When the analysis result indicates that the cell is in a network poor state, the operator server performs a capacity expansion and/or network optimization operation on the cell according to the analysis result, and when the analysis result indicates that the cell When the cell is in a network good state, the operator server continues to perform the currently performing operation on the cell.

In this embodiment, the usage model is grouped according to latitude and longitude data of at least two terminals, basic information of the terminal, and usage information of an application in the terminal, and the usage model is grouped according to the grouping result and the usage model. The network status of the cell in which the terminal is located, and the network status of the cell is sent to the operator server, so that when the network where the terminal is located has a poor network condition, the operator server according to the cell The network condition expands the capacity and/or optimizes the operation of the cell, thereby improving the user experience.

Referring to FIG. 4, FIG. 4 is a schematic flowchart of analyzing results according to the grouping result of the usage model in the same cell and the usage model to analyze the network status of the cell according to an embodiment of the present invention.

In this embodiment, optionally, the step S31 includes:

Step S311, determining whether the signal strength of the terminal corresponding to the usage model in the same group usage model in the same cell is less than a preset signal strength threshold;

Step S312, when the signal strength of the terminal in the same cell is less than the preset signal strength threshold, and the proportion of the terminal in the same cell that is smaller than the preset signal strength threshold exceeds a preset ratio. Determining that the cell is in a network blind zone;

Step S313, when the signal strength of the terminal in the same cell is greater than or equal to the preset signal strength threshold, determining that the cell is in a network normal state.

The server determines whether the signal strength of the terminal corresponding to the usage model in the same group usage model in the same cell is less than a preset signal strength threshold. When the server determines that the signal strength of the terminal in the same cell is less than the preset signal strength threshold, and the When the proportion of the terminal that is smaller than the preset signal strength threshold in the same cell exceeds a preset ratio, it is determined that the cell is in a network blind zone. When the signal strength of the terminal in the same cell is greater than or equal to the preset signal strength threshold, the server determines that the cell is in a network normal state. For example, when the server determines that the same cell, the network type model is a terminal that has a signal strength less than the preset signal strength threshold in the terminal corresponding to the 4G, and the same cell is smaller than the pre- When the ratio of the terminal with the signal strength threshold exceeds a preset ratio, the server determines that the 4G network in the cell has a network blind zone and is in an abnormal state.

Embodiments of the present invention further provide a computer readable storage medium storing computer executable instructions that are implemented when the computer executable instructions are executed.

The application also provides a recommendation device for a business package.

Referring to FIG. 5, FIG. 5 is a schematic diagram of functional modules of a first embodiment of a device for recommending a service package according to the present application.

In this embodiment, the recommended device of the service package includes:

The establishing module 10 is configured to establish, according to the basic information of the terminal and the usage information of the application, when receiving the basic information of the terminal and the usage information of the application in the terminal sent by the at least two terminals model;

When the server receives the basic information of the terminal and the usage information of the application in the terminal, the server establishes a usage model according to the basic information of the terminal and the usage information of the application. The usage model includes a data usage model of the end user and a bandwidth usage model of the end user. Each end user has a corresponding data usage model and bandwidth usage model. The server is a server capable of accurately analyzing the end user data, and the server may exist in a cloud or a physical device. The minimum configuration of the server is: Intel Xeon E5 six-core processor Intel Xeon E5-2620, 8GB (Gigabyte, gigabytes) of memory and 2TB (Terabyte, terabyte) data disk. The terminal includes, but is not limited to, a smartphone and a tablet. When the terminal is an Android phone or an Android tablet, the minimum configuration of the Android phone or the Android tablet is an ARM (Advanced RISC Machines) architecture CPU (Central Processing Unit, The central processing unit), 512 MB (MByte) of RAM, 1 GB of ROM (Read-Only Memory, read only memory) and output devices are capacitive touch screens with a resolution of 460*640. When the terminal is an Apple phone or an Apple tablet, the minimum configuration of the Apple phone or the Apple tablet is: ARM Cortex-A8 CPU, 256 MB of RAM, 8 GB of ROM and output device with resolution of 480*320. Capacitive touch screen.

The usage information of the application includes the number of startups of the application, the running time of the application, the traffic consumption data, the occupied bandwidth data, the type of network used by the application, and the network signal strength. The terminal, according to the installation list information of the APP, reflects com.android.internal.os.PkgUsageStats, and reads the PkgUsageStats.launchCount and PkgUsageStats.usageTime attributes respectively, to obtain the startup times and running times of different APPs, that is, Number of starts and running time of different applications of the terminal; the terminal passes The TelephonyManager.subtype obtains the current network type and signal strength, and writes the current network type and the signal strength into the sqllite database.

The first calculating module 20 is configured to calculate a similarity between the usage models according to a preset algorithm, and divide the usage model whose similarity is greater than a preset threshold into the same group usage model;

The server calculates the similarity between the usage models according to a preset algorithm, and divides the usage model with the similarity greater than a preset threshold into the same group usage model, that is, calculates the interaction between the terminal users according to a preset algorithm. The similarity between the data traffic usage model and/or the bandwidth usage model, and the data traffic usage model in which the data traffic usage model of the end user is greater than a preset threshold is divided into the same group usage model, that is, the The similarity of the data traffic usage model is greater than the preset data traffic. The terminal users corresponding to the model similarity threshold are divided into the same group, and/or the similarity of the bandwidth usage model between the terminal users is greater than a preset. The data traffic usage threshold of the bandwidth usage model similarity is divided into the same group usage model, that is, the end users corresponding to the bandwidth usage model whose degree of similarity is greater than the preset bandwidth usage model similarity threshold are divided into the same group.

The preset algorithm is an algorithm combining a collaborative filtering algorithm and a cosine similarity, and the collaborative filtering algorithm is a main algorithm of the e-commerce recommendation system, and the collaborative filtering algorithm filters and analyzes user interests in the user group. Finding users of similar users (interests), synthesizing the evaluation of certain information by these similar users, and forming a prediction of how much the e-commerce recommendation system likes the designated users. Compared with the traditional text filtering algorithm, the collaborative filtering algorithm can filter the information of the machine based on content analysis automatically, and can filter based on some complicated and difficult to express concepts (information quality, taste), with recommended novelty. . The cosine similarity measures the similarity between them by measuring the cosine of the angle between the product spaces in the two vectors. The cosine of the 0 degree angle is 1, while the cosine of any other angle is not greater than 1; and its minimum is -1. Thus the cosine of the angle between the two vectors determines whether the two vectors generally point in the same direction. When two vectors have the same pointing, the value of cosine similarity is 1; when the angle between two vectors is 90°, the value of cosine similarity is 0; when the two vectors point to the opposite direction, the value of cosine similarity Is -1. In the comparison process, the size of the vector is not considered, only the direction of the vector is considered. Cosine similarity is usually used when the angle between two vectors is less than 90°, so the value of cosine similarity is between 0 and 1.

The second calculating module 30 is configured to calculate a distance between the usage model of the different groups and the target interest point, and determine a recommendation strategy of the service package of the different group using the model according to the distance.

Referring to FIG. 6, FIG. 6 is a schematic diagram of a functional module of a second computing module 30 according to an embodiment of the present invention.

Optionally, the second calculating module 30 includes:

The first determining unit 31 is configured to determine coordinates of the use model of different groups and coordinates of the target interest point in the coordinate system;

The calculating unit 32 is configured to calculate a distance between the usage model of the different group and the target interest point according to the coordinates of the different group of usage models and the coordinates of the target interest point in the coordinate system;

The second determining unit 33 is configured to determine a recommendation policy of the service package of the different group of usage models according to the size of the distance.

The server determines coordinates of different groups of usage models in a three-dimensional coordinate system and coordinates of a target interest point in a three-dimensional coordinate system, and the server is based on coordinates of the usage model of the different groups and the target interest points Coordinates in the coordinate system calculate different usage models and target points of interest The distance between the two groups is determined according to the size of the distance between the usage model of the different groups and the target interest point. When the distance between the usage model of the different group and the target interest point is smaller, the recommended degree value of the target interest point is higher, and when the recommendation degree value is higher, the different group is When using the model for business package recommendation, first recommend the business package corresponding to the high value. If the server calculates that the distance between the first group of data traffic usage models and the social application WeChat is 0.2, the distance from the Youku is 0.18, and the distance between the second group of data traffic usage models and the social application WeChat is 0.15, and The distance of the Youku is 0.22, and the server determines the first according to the distance between the first set of data traffic usage models and the second set of data traffic usage models and the WeChat and the Youku. The group data traffic usage model and the WeChat of the second set of data traffic usage models and the recommended value of the Youku, in the first set of data traffic usage models, the value of the recommendation of the Youku is greater than the The recommendation value of the WeChat, in the second group of data traffic usage models, the recommendation value of the Youku is smaller than the recommendation value of the WeChat, and the server determines the recommendation of the corresponding service package according to the recommendation value. Strategy.

In this embodiment, the usage model is established according to the basic information of the at least two terminals and the usage information of the application, and the similarity between the usage models is calculated, and the usage model with the similarity greater than a preset threshold is used. Dividing into the same group of usage modalities, and calculating the distance between the usage models of the different groups and the target interest points, and determining the recommendation strategies of the service packages of the different groups using the models according to the distances. The basic information of the terminal used by the user and the usage information of the application in the terminal are collected on the Android and iOS terminal platforms, and the collected information is used according to the collected information. The preset algorithm establishes a usage model to obtain data such as traffic destination and bandwidth occupation of the user using the application, and determines a traffic policy and/or a QoS package recommendation policy of the different group usage model according to the data, and the operator server according to the The recommended policies for the traffic tiers and/or QoS packages of the different groups of usage models are to provide traffic tiers and/or QoS tiers for the users, and provide data basis for cooperation between the operator servers and the OTT vendors.

Referring to FIG. 7, FIG. 7 is a schematic diagram of functional modules of a first embodiment of a service package of the present application. The second embodiment of the recommendation device for the service package of the present application is proposed based on the first embodiment of the recommendation device of the service package of the present application. example.

In this embodiment, the establishing module 10 is configured to receive basic information of the terminal sent by at least two terminals and usage information of an application in the terminal, and receive a location service based When the latitude and longitude data of the terminal is sent by the server, the usage model is established according to the basic information of the terminal, the usage information of the application, and the latitude and longitude data of the terminal.

The location information is obtained by the terminal by using a method of calling the TelephonyManager.getCellLocation().getCid() and the TelephonyManager.getCellLocation().getLac() to obtain the identification information of the location where the cell where the terminal is located and the LAC (location area) Code, location area code) data. The terminal reports the identification information of the location of the cell and the LAC data to the LBS (Location Based Service) of the cloud. And a server, wherein the LBS server obtains latitude and longitude data of the location where the terminal is located according to the identifier information of the location of the cell and the LAC data, and sends the latitude and longitude data of the location where the terminal is located to the server. The minimum configuration of the LBS server is: Xeon E3-1230v3 CPU; 8 GB of memory; and 1 TB data disk, the LBS server may exist in the cloud or may exist in the physical device. After the server receives the latitude and longitude data sent by the LBS, the geographic location of the terminal is determined according to the latitude and longitude data, and the geographic location of the terminal is written into the sqllite database.

Optionally, the first calculating module 20 is configured to calculate, according to a preset algorithm, a similarity between the usage models in the same cell, where the similarity in the same cell is greater than a preset similarity. The usage model is divided into the same group usage model, and the grouping result of the usage model is obtained.

The recommended device of the service package further includes:

The analyzing module 40 is configured to analyze the network status of the cell according to the grouping result of the usage model in the same cell and the usage model, and obtain an analysis result;

The server analyzes the network status of the cell according to the grouping result of the usage model in the same cell and the usage model, and obtains an analysis result. The server analyzes the status of the network type of the terminal user corresponding to the 3G in the same cell, and combines the network type model with the signal strength model of the terminal user corresponding to the 3G, and combines the network The type model is a geographical location information model of the end user corresponding to the 3G, and determines the network status of the cell in which the network type model is the terminal user corresponding to the 3G.

The sending module 50 is configured to send the analysis result to the operator server, so that the operator server performs a corresponding operation on the cell according to the analysis result.

Sending, by the server, the analysis result to the operator server for the operator The server receives the analysis result, and performs a corresponding operation on the cell according to the analysis result. When the analysis result indicates that the cell is in a network poor state, the operator server performs a capacity expansion and/or network optimization operation on the cell according to the analysis result, and when the analysis result indicates that the cell When the cell is in a network good state, the operator server continues to perform the currently performing operation on the cell.

Referring to FIG. 8, FIG. 8 is a schematic diagram of a functional module of an analysis module according to an embodiment of the present invention.

In this embodiment, optionally, the analyzing module 40 includes:

The determining unit 41 is configured to determine whether the signal strength of the terminal corresponding to the usage model in the same group usage model in the same cell is less than a preset signal strength threshold;

The first determining unit 42 is configured to: when the signal strength of the terminal in the same cell is less than the preset signal strength threshold, and the terminal in the same cell is smaller than the preset signal strength threshold When the ratio exceeds a preset ratio, determining that the cell is in a network blind zone;

The second determining unit 43 is configured to determine that the cell is in a network normal state when the signal strength of the terminal in the same cell is greater than or equal to the preset signal strength threshold.

The server determines whether the signal strength of the terminal corresponding to the usage model in the same group usage model in the same cell is less than a preset signal strength threshold. When the server determines that the signal strength of the terminal in the same cell is less than the preset signal strength threshold, and the proportion of the terminal in the same cell that is smaller than the preset signal strength threshold exceeds a preset At the time of the ratio, it is determined that the cell is in a network dead zone. When the signal strength of the terminal in the same cell is greater than or equal to the preset signal strength threshold, the server determines that the cell is in a network normal state. And when the server determines that the same cell is within the network When the type model is a terminal in which the signal strength is less than the preset signal strength threshold in the terminal corresponding to the 4G, and the proportion of the terminal in the same cell that is smaller than the preset signal strength threshold exceeds a preset ratio, The server determines that the 4G network in the cell has a network dead zone and is in an abnormal state.

It is to be understood that the term "comprises", "comprising", or any other variants thereof, is intended to encompass a non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes those elements. It also includes other elements that are not explicitly listed, or elements that are inherent to such a process, method, article, or device. An element that is defined by the phrase "comprising a ..." does not exclude the presence of additional equivalent elements in the process, method, item, or device that comprises the element.

The serial numbers of the embodiments of the present invention are merely for the description, and do not represent the advantages and disadvantages of the embodiments.

Through the description of the above embodiments, those skilled in the art can clearly understand that the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is better. Implementation. Based on such understanding, the technical solution of the present application, which is essential or contributes to the related art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk, CD-ROM). The instructions include a number of instructions for causing a terminal device (which may be a cell phone, computer, server, air conditioner, or network device, etc.) to perform the methods described in various embodiments of the present invention.

One of ordinary skill in the art will appreciate that all or a portion of the above steps may be performed by a program to instruct related hardware, such as a processor, which may be stored in a computer readable storage medium, such as a read only memory, disk or optical disk. Wait. Alternatively, all or part of the steps of the above embodiments may also be implemented using one or more integrated circuits. Correspondingly, each module/unit in the above embodiment may be implemented in the form of hardware, for example, by implementing an integrated circuit to implement its corresponding function, or may be implemented in the form of a software function module, for example, executing a program stored in the memory by a processor. / instruction to achieve its corresponding function. Embodiments of the invention are not limited to any specific form of combination of hardware and software.

The above is only an alternative embodiment of the present invention, and thus does not limit the scope of the patent application, and the equivalent structure or equivalent process transformation, or directly or between the contents of the specification and the drawings. The use of other related technical fields is equally included in the scope of patent protection of this application.

Industrial applicability

Claims

A recommended method for a business package, including the following steps:

When receiving the basic information of the terminal and the usage information of the application in the terminal sent by the at least two terminals, establishing a usage model according to the basic information of the terminal and the usage information of the application;

Calculating the similarity between the usage models according to a preset algorithm, and dividing the usage model with the similarity greater than a preset threshold into the same group usage model;

Calculating the distance between the usage model of the different groups and the target interest point, and determining the recommendation strategy of the business package of the different group using the model according to the distance.
The method of recommending a service package according to claim 1, wherein said calculating a distance between a usage model of a different group and a target point of interest, and determining a service package of the usage model of the different group according to the distance The steps to recommend a strategy include:

Determining the coordinates of the used models of different groups and the coordinates of the target points of interest in the coordinate system;

Calculating a distance between the usage model of the different group and the target interest point according to the coordinates of the use model of the different groups and the coordinates of the target interest point in the coordinate system;

Determining a recommendation policy of the business package of the different group of usage models according to the size of the distance.
The method of recommending a service package according to claim 1, wherein when the basic information of the terminal and the usage information of the application in the terminal are received by at least two terminals, according to the basic of the terminal The steps of establishing information and using the usage information of the application include:

Receiving basic information of the terminal sent by at least two terminals and usage information of an application in the terminal, and receiving latitude and longitude data of the terminal sent by the location service based server, according to the basic of the terminal The information, the usage information of the application, and the latitude and longitude data of the terminal establish a usage model.
The method for recommending a service package according to claim 3, wherein the calculating the similarity between the usage models according to a preset algorithm, and dividing the usage model having a similarity greater than a preset similarity into the same group The steps of the model include:

Calculating the similarity between the usage models in the same cell according to a preset algorithm, and dividing the usage model in the same cell with the similarity greater than the preset similarity into the same group usage model, to obtain the usage model. The result of the grouping.
The method of recommending a service package according to claim 4, wherein said calculating a distance between a usage model of a different group and a target point of interest, and determining a service package of the usage model of the different group according to the distance The steps to recommend a strategy include:

And analyzing, according to the grouping result of the usage model in the same cell and the usage model, analyzing a network condition of the cell, and obtaining an analysis result;

And sending the analysis result to the operator server, where the operator server performs a corresponding operation on the cell according to the analysis result.
The method of recommending a service package according to claim 5, wherein said step of obtaining an analysis result based on a grouping result of said usage model in said same cell and said usage model analyzing said network condition of said cell include:

Determining whether a signal strength of the terminal corresponding to the usage model in the same group usage model in the same cell is smaller than a preset signal strength threshold;

When the signal strength of the terminal in the same cell is less than the preset signal strength threshold, and the proportion of the terminal in the same cell that is smaller than the preset signal strength threshold exceeds a preset ratio, determining The cell is in a network blind zone;

When the signal strength of the terminal in the same cell is greater than or equal to the preset signal strength threshold, determining that the cell is in a network normal state.
The method for formulating a service package according to claim 1, wherein the preset algorithm is an algorithm combining a collaborative filtering algorithm and a cosine similarity.
The method for recommending a service package according to any one of claims 1 to 7, wherein the basic information of the terminal includes an international mobile subscriber identity, an operating system version number and vendor information of the terminal, and the terminal installation Application list;

The usage information of the application includes the number of startups of the application, the running time of the application, traffic consumption data, occupied bandwidth data, the type of network used by the application, and network signal strength.
A recommended device for a business package, including:

Establishing a module, configured to establish a usage model according to the basic information of the terminal and the usage information of the application when receiving basic information of the terminal sent by at least two terminals and usage information of the application in the terminal ;

a first calculating module, configured to calculate a similarity between the usage models according to a preset algorithm, and divide the usage model with the similarity greater than a preset threshold into a same group usage model;

The second calculating module is configured to calculate a distance between the usage model of the different groups and the target interest point, and determine a recommendation strategy of the service package of the different group using the model according to the distance.
The recommendation device of the service package of claim 9, wherein the second calculation module comprises:

a first determining unit configured to determine coordinates of a different group of usage models and coordinates of the target points of interest in the coordinate system;

a calculating unit, configured to calculate a distance between the usage model of the different group and the target interest point according to the coordinates of the usage model of the different groups and the coordinates of the target interest point in the coordinate system;

The second determining unit is configured to determine a recommendation policy of the service package of the different group of usage models according to the size of the distance.
The device for recommending a service package according to claim 9, wherein the establishing module is configured to receive basic information of the terminal and usage information of an application in the terminal when receiving at least two terminals, and Upon receiving the latitude and longitude data of the terminal transmitted by the location service based server, the usage model is established according to the basic information of the terminal, the usage information of the application, and the latitude and longitude data of the terminal.
The device for recommending a service package according to claim 11, wherein the first calculating module is configured to calculate a similarity between the usage models in the same cell according to a preset algorithm, and the same cell The usage model in which the internal similarity is greater than the preset similarity is divided into the same group usage model, and the grouping result of the usage model is obtained.
The recommendation device of the service package according to claim 12, wherein the recommendation device of the service package further comprises:

The analyzing module is configured to analyze the network status of the cell according to the grouping result of the usage model in the same cell and the usage model, and obtain an analysis result;

And a sending module, configured to send the analysis result to the operator server, where the operator server performs a corresponding operation on the cell according to the analysis result.
The recommendation device of the service package according to claim 13, wherein the analysis module comprises:

a determining unit, configured to determine whether a signal strength of the terminal corresponding to the usage model in the same group usage model in the same cell is smaller than a preset signal strength threshold;

a first determining unit, configured to: when the signal strength of the terminal in the same cell is less than the preset signal strength threshold, and the ratio of the terminal in the same cell that is smaller than the preset signal strength threshold When the preset ratio is exceeded, it is determined that the cell is in a network blind zone;

The second determining unit is configured to determine that the cell is in a network normal state when the signal strength of the terminal in the same cell is greater than or equal to the preset signal strength threshold.
The recommendation device of the service package according to claim 9, wherein the preset algorithm is an algorithm combining a collaborative filtering algorithm and a cosine similarity.
The recommendation device of the service package according to any one of claims 9 to 15, wherein the basic information of the terminal includes an international mobile subscriber identity, an operating system version number and vendor information of the terminal, and the terminal installation Application list;

The usage information of the application includes the number of startups of the application, the running time of the application, traffic consumption data, occupied bandwidth data, the type of network used by the application, and network signal strength.