Broadband potential user mining method and device and readable storage medium

Abstract

The invention provides a mining method and device for broadband potential users and a readable storage medium, and the method comprises the steps: screening out the broadband potential users from mobile network stock users, and obtaining a first target group; recognizing a resident night residential district of each user in the first target group based on a positioning technology of a mobile network and user behavior analysis; screening out users whose broadband coverage conditions are full coverage and partial coverage according to the broadband coverage conditions of the users resident in the night residential districts in the first target group, and obtaining a second target group; determining a third target group according to the active time and the number of active days of each user in the second target group in the preset period; and determining all users in the third target group as final broadband potential users. According to the method, the device and the medium, the problem that the mining accuracy of the broadband potential users is relatively low due to the fact that the users are analyzed only from the data perspective and the actual conditions of the users are not fully considered in the existing mining method of the broadband potential users can be solved.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and in particular, to a method, an apparatus, and a readable storage medium for mining potential broadband users. Background Art

[0002] The broadband service has always been one of the key services of communication operators. With the development of the communication industry, each family has a certain demand for broadband networks. To promote the development of the broadband service, operators need to deeply evaluate the broadband demand of users, accurately identify potential broadband users, and carry out broadband marketing activities for these specific users.

[0003] However, currently, the mining of potential broadband users mainly analyzes the characteristics of existing broadband users and then combines the data of in-network users to build a big data model to mine potential broadband users. But in the existing technology, only the users are analyzed from the data perspective, and the actual situation of the users is not fully considered, resulting in a low accuracy rate for mining potential broadband users. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method, an apparatus, and a readable storage medium for mining potential broadband users to solve the problem that the existing method for mining potential broadband users only analyzes the users from the data perspective and does not fully consider the actual situation of the users, resulting in a low accuracy rate for mining potential broadband users.

[0005] In a first aspect, the present invention provides a method for mining potential broadband users, the

[0006] method includes:

[0007] Screen potential broadband users from the existing mobile network users to obtain a first target group;

[0008] Based on the positioning technology of the mobile network and user behavior analysis, identify the residential communities where each user in the first target group resides at night;

[0009] According to the broadband coverage in the residential communities where each user in the first target group resides at night, screen out the users with full coverage and partial coverage of broadband, and obtain a second target group;

[0010] Determine a third target group according to the active time and active days of each user in the second target group within a preset period;

[0011] Determine all users in the third target group as the final potential broadband users.

[0012] Further, the step of screening potential broadband users from the existing mobile network users to obtain a first target group specifically includes:

[0013] Collect relevant data of broadband users and non-broadband users from the existing mobile network users as a data set, and label the data set;

[0014] Based on a preset supervised learning algorithm, use the labeled data set to train a broadband potential user prediction model;

[0015] Use the trained broadband potential user prediction model to predict the existing mobile network users, identify broadband potential users, and obtain a first target group.

[0016] Further, the permanent night residential communities of each user in the first target group are identified based on the positioning technology of the mobile network and user behavior analysis, specifically including:

[0017] Within a cycle time, according to the terminal reporting information of each user in the first target group, count the resident base station communities and residence durations of each user in a preset night time period every day;

[0018] For each user in the first target group, select the resident base station community where the user's residence duration per day is greater than a preset first duration and the number of residence days within the cycle time is greater than a preset first number of days as the user's permanent night base station community;

[0019] According to the permanent night base station communities of each user in the first target group and the corresponding residence frequencies, and / or residence days, and / or the most recent arrival time, determine the home permanent base station community of each user;

[0020] According to the border relationship between the home permanent base station community of each user and the border of the actual residential broadband electronic fence, determine the permanent night residential community of each user in the first target group.

[0021] Further, the determining of the home permanent base station community of each user according to the permanent night base station communities of each user in the first target group and the corresponding residence frequencies, and / or residence days, and / or the most recent arrival time specifically includes:

[0022] For a user in the first target group whose permanent night base station community is one, use the permanent night base station community as the corresponding home permanent base station community;

[0023] For a user in the first target group whose permanent night base station communities are at least two, select the permanent night base station community with a high residence frequency, a large number of residence days, and a recent arrival as the home permanent base station community in the order of residence frequency first, residence days second, and the most recent arrival time third.

[0024] Further, screening out users with full coverage and partial coverage of broadband in the permanent night residential communities of each user in the first target group to obtain a second target group, specifically including:

[0025] Obtaining the broadband coverage of each user in the permanent night residential community in the first target group, where the broadband coverage is divided into full coverage, partial coverage, single-point coverage, and no coverage. The single-point coverage means that only one building has broadband coverage within the broadband electronic fence, and the partial coverage means that there are multiple buildings with broadband coverage within the broadband electronic fence but not all buildings are covered;

[0026] Excluding users with single-point coverage and no coverage of broadband in the first target group to obtain the second target group.

[0027] Further, determining a third target group according to the active time and active days of each user in the second target group within a preset period, specifically including:

[0028] Counting the active time and active days of each user in the second target group within a preset period;

[0029] Screening out users with active days greater than a preset number of days and active time greater than a preset number of hours from the second target group to obtain the third target group.

[0030] Further, after determining all users in the third target group as the final broadband potential users, the method further includes:

[0031] Formulating corresponding broadband marketing strategies according to the location scene attributes of the permanent night residential communities of the final broadband potential users.

[0032] In a second aspect, the present invention provides a device for mining broadband potential users, and the device includes:

[0033] A first screening module, configured to screen out broadband potential users from mobile network stock users to obtain a first target group;

[0034] A first identification module, connected to the first screening module, configured to identify the permanent night residential community of each user in the first target group based on the positioning technology of the mobile network and user behavior analysis;

[0035] A second screening module, connected to the first identification module, configured to screen out users with full coverage and partial coverage of broadband in the permanent night residential communities of each user in the first target group to obtain a second target group;

[0036] A third screening module, connected to the second screening module, is configured to determine a third target group according to the active time and active days of each user in the second target group within a preset period;

[0037] A potential user determination module, connected to the third screening module, is configured to determine all users in the third target group as the final broadband potential users.

[0038] In a third aspect, the present invention provides a device for mining broadband potential users, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to implement the method for mining broadband potential users described in the first aspect above.

[0039] In a fourth aspect, the present invention provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements the method for mining broadband potential users described in the first aspect above.

[0040] The method, device, and readable storage medium for mining broadband potential users provided by the present invention. First, broadband potential users are screened out from the mobile network stock users to obtain a first target group; then, based on the positioning technology of the mobile network and user behavior analysis, the permanent night residential communities of each user in the first target group are identified; and according to the broadband coverage in the permanent night residential communities of each user in the first target group, users with full coverage and partial coverage of broadband are screened out to obtain a second target group; then, a third target group is determined according to the active time and active days of each user in the second target group within a preset period; and all users in the third target group are determined as the final broadband potential users. By comprehensively using the mobile network positioning technology and user behavior analysis, the present invention accurately identifies the permanent night residential communities of users, and combines the broadband coverage and user activity data to effectively screen out the real broadband potential users. The present invention not only screens out broadband potential users from a data perspective, but also fully considers the actual living conditions and active behaviors of users, thereby effectively improving the accuracy of mining broadband potential users. It solves the problem that the existing methods for mining broadband potential users only analyze users from a data perspective and fail to fully consider the actual situation of users, resulting in a low accuracy of mining broadband potential users. Description of the Drawings

[0041] Figure 1 It is a flowchart of a method for mining broadband potential users according to Embodiment 1 of the present invention;

[0042] Figure 2 It is a schematic diagram showing the relationship between the base station coverage and the actual residential broadband electronic fence according to an embodiment of the present invention;

[0043] Figure 3Schematic structural diagram of a device for mining potential broadband users in Embodiment 2 of the present invention;

[0044] Figure 4 Schematic structural diagram of a device for mining potential broadband users in Embodiment 3 of the present invention. Detailed implementation manners

[0045] To enable those skilled in the art to better understand the technical solutions of the present invention, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0046] It can be understood that the specific embodiments and the accompanying drawings described herein are only used to explain the present invention, rather than limiting the present invention.

[0047] It can be understood that, without conflict, the various embodiments and the various features in the embodiments of the present invention can be combined with each other.

[0048] It can be understood that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings of the present invention, and the parts not related to the present invention are not shown in the accompanying drawings.

[0049] It can be understood that each unit and module involved in the embodiments of the present invention may correspond to only one physical structure, or may be composed of multiple physical structures, or multiple units and modules may also be integrated into one physical structure.

[0050] It can be understood that the terms "first", "second", etc. in the embodiments of the present invention are used to distinguish different objects, or to distinguish different processes for the same object, rather than to describe a specific order of the objects.

[0051] It can be understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of the present invention may occur in an order different from that marked in the accompanying drawings.

[0052] It can be understood that in the flowcharts and block diagrams of the present invention, the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to the embodiments of the present invention are shown. Among them, each block in the flowchart or block diagram may represent a unit, module, program segment, or code, which contains executable instructions for implementing the specified function. Moreover, each block or combination of blocks in the block diagram and flowchart can be implemented by a hardware-based system for implementing the specified function, or by a combination of hardware and computer instructions.

[0053] It can be understood that the units and modules involved in the embodiments of the present invention can be implemented in software or in hardware. For example, the units and modules can be located in the processor.

[0054] Embodiment 1:

[0055] This embodiment provides a method for mining potential broadband users, as Figure 1 shown, the method includes:

[0056] Step S101: Screen potential broadband users from the existing mobile network users to obtain a first target group.

[0057] In this embodiment, first, potential broadband users are initially identified from the existing mobile network users through a machine learning algorithm to obtain a first target group. Among them, the existing mobile network users refer to the user group that has registered or used the mobile network (such as mobile phone network, mobile Internet service, etc.) within a certain time period.

[0058] Optionally, the step of screening potential broadband users from the existing mobile network users to obtain a first target group specifically includes:

[0059] Collect relevant data of broadband users and non-broadband users from the existing mobile network users as a data set, and label the data set;

[0060] Based on a preset supervised learning algorithm, use the labeled data set to train a potential broadband user prediction model;

[0061] Use the trained potential broadband user prediction model to predict the existing mobile network users, identify potential broadband users, and obtain a first target group.

[0062] In this embodiment, the relevant data includes user basic information and data strongly related to broadband usage, such as traffic usage, Internet access duration, whether to use high-definition video services, whether to be a broadband user, etc. The machine learning algorithm preferably adopts a supervised learning algorithm, such as decision tree, random forest, etc. By using the labeled data set to train a potential broadband user prediction model, and using the trained potential broadband user prediction model to predict the existing mobile network users, high-quality broadband marketing users (i.e., potential broadband users) are mined from non-broadband users, and this part of users constitutes the first target group.

[0063] Step S102: Identify the permanent night residential communities of each user in the first target group based on the positioning technology of the mobile network and user behavior analysis.

[0064] In this embodiment, utilize the precise positioning ability of the mobile network to determine the permanent night residential community of the user through user behavior analysis.

[0065] Optionally, the step of identifying the permanent night residential communities of each user in the first target group based on the positioning technology of the mobile network and user behavior analysis specifically includes:

[0066] Within the cycle time, based on the terminal reporting information of each user in the first target group, count the resident base station cells and residence duration of each user during the preset nightly time period every day;

[0067] For each user in the first target group, select the resident base station cells where the user's residence duration every day is greater than the preset first duration and the number of residence days within the cycle time is greater than the preset first number of days as the user's regular nightly base station cells;

[0068] Based on the regular nightly base station cells of each user in the first target group, as well as the corresponding residence frequency, and / or residence days, and / or the most recent arrival time, determine the home regular base station cells of each user;

[0069] Based on the border relationship between the home regular base station cells of each user and the border of the actual residential broadband electronic fence, determine the regular nightly residential communities of each user in the first target group.

[0070] In this embodiment, the cycle time can be one week, 15 days, one month, etc. The preset nightly time period can be in the later part of each day until early morning of the next day. Specifically, it can be any continuous time period between 20:00 and 9:00 of the next day (for example, 21:00 to 24:00 and 0:00 to 8:00), or it can also be 0:00 to 8:00 of each day. Based on the terminal reporting information of the users, count the resident base station cells and residence duration of each user in the first target group during the preset nightly time period every day. Select the resident base station cells where the user's residence duration every day is greater than the preset first duration (such as 3 hours) and the number of residence days within the cycle time is greater than the preset first number of days (such as 3 days) as the user's regular nightly base station cells. When a user has multiple regular nightly base station cells, the home regular base station cells of each user can be further determined based on the residence frequency, and / or residence days, and / or the most recent arrival time.

[0071] In this embodiment, after determining the home resident base station cells of each user, according to the border relationship between the home resident base station cells of each user and the border of the actual residential broadband electronic fence, the home resident base station cells are placed into the broadband electronic fence. If the home resident base station cell is completely contained within the broadband electronic fence, it is directly determined that the home resident base station cell belongs to the broadband electronic fence; if the home resident base station cell is not completely contained within any broadband electronic fence, its belonging is determined according to the proportion of the signal coverage range of the home resident base station cell covering the border of the broadband electronic fence. The specific steps may include: First, calculate the intersection area between the signal coverage range of the home resident base station cell and each broadband electronic fence, and then calculate the percentage of these intersection areas in the total signal coverage area of the home resident base station cell. Then, for the home resident base station cell, find the broadband electronic fence with the largest proportion within its signal coverage range. If the coverage proportion of the broadband electronic fence with the largest proportion is greater than or equal to the preset proportion value (such as 30%), the home resident base station cell is determined to belong to the broadband electronic fence; if the coverage proportion of all broadband electronic fences is less than the preset proportion value (such as 30%), the home resident base station cell does not belong to any broadband electronic fence, and further location information or manual intervention may be required to determine its belonging.

[0072] It should be noted that when determining the cycle time for screening the user's resident night base station cells, we need to consider the requirements for the stability and reliability of the location information in actual applications, which is similar to the consideration of location accuracy. The longer the cycle time, the larger the amount of user residence data collected, so that the activity patterns of users at different time periods and locations can be more comprehensively reflected, making the user location information more stable. Therefore, a longer time cycle helps to improve the accuracy and reliability of the location information. Usually, in order to obtain sufficiently stable and reliable user location information, it is preferably to select 1 month as the time cycle for analysis, but the specific time length should also be determined according to the actual application scenario and requirements.

[0073] It should be noted that the broadband electronic fence is a polygon composed of a series of longitude and latitude points, used to mark the areas where the operator's broadband may need to develop and construct in reality. It includes location scenarios such as residential communities, commercial buildings, company parks, street shops, stations, parks, etc.

[0074] Optionally, determining the home resident base station cells of each user according to the resident night base station cells of each user in the first target group and the corresponding residence frequencies, and / or residence days, and / or the most recent arrival time specifically includes:

[0075] For users in the first target group with only one resident night base station cell, the resident night base station cell is used as the corresponding home resident base station cell;

[0076] For users in the first target group who have at least two resident night base station cells, select the resident night base station cell with a high residence frequency, a large number of residence days, and a recent arrival time as the home resident base station cell in the order of residence frequency first, residence days second, and the most recent arrival time third.

[0077] In this embodiment, the user residence frequency can be determined on a daily granularity, that is, for a certain resident night base station cell, how many days on average does the user come to this place once, and try to select the resident night base station cell with a smaller interval days for the user.

[0078] For example, an average interval days can be calculated as the residence frequency. From the start of residence to the current time, how many days has the user resided in total. That is, if the residence frequency in area A = 7 / 3 ≈ 2.3 days, and the residence frequency in area B = 7 / 4 = 1.75 days. Then select area B as the home resident base station cell.

[0079]

[0080] In this embodiment, in the case where a user has multiple resident night base station cells, according to the importance degree: residence frequency > residence days > the most recent arrival time, select the user's unique home resident base station cell. For example, taking 2 resident night base station cells as an example, if the residence frequencies of the two resident night base station cells are the same, we will select according to the number (or length) of residence days, and the one with more days wins. If the residence days are also the same, then select the resident night base station cell that the user visited most recently. It should be noted that a high residence frequency means fewer residence days within a cycle (such as a month). If the residence frequencies are the same, it means that within this month, the same user has the same number of residence days in two places. Therefore, at this time, the historical records can be compared, comparing the total number of times the user has been to the two places respectively in the historical records, and select the place with more visits as the user's home resident base station cell.

[0081] Step S103: According to the broadband coverage situation in the resident night residential areas of each user in the first target group, screen out the users with full coverage and partial coverage of broadband, and obtain the second target group.

[0082] In this embodiment, the broadband coverage situation is divided into full coverage, partial coverage, single-point coverage, and no coverage. The single-point coverage means that only one building in the broadband electronic fence has broadband coverage. The partial coverage means that there are multiple buildings in the broadband electronic fence with broadband coverage but not all covered. No coverage means that there is no broadband coverage in the broadband electronic fence, and full coverage means that the broadband is completely covered in the broadband electronic fence.

[0083] Optionally, screening out users with full coverage and partial coverage of broadband in the night residential communities where the users in the first target group are resident to obtain a second target group, specifically including:

[0084] Obtaining the broadband coverage of each user in the night residential community where the user in the first target group is resident;

[0085] Excluding users with single-point coverage and no coverage of broadband in the first target group to obtain the second target group.

[0086] In this embodiment, by excluding users with single-point coverage and no coverage of broadband in the first target group, the remaining users with full coverage and partial coverage form the second target user group.

[0087] Step S104: Determining a third target group according to the active time and active days of each user in the second target group within a preset period.

[0088] Specifically, counting the active time and active days of each user in the second target group within a preset period; screening out users in the second target group whose active days are greater than a preset number of days and whose active time is greater than a preset number of hours to obtain the third target group.

[0089] In this embodiment, the active days refer to the number of days when a user has activity records on the network. The active time refers to the total duration of a user logging in to the network and continuously using it (such as browsing, interacting, etc.) during these active days. Users in the second target group whose active days are greater than a preset number of days and whose active time is greater than a preset number of hours within a preset period are defined as the third target group.

[0090] Step S105: Determining all users in the third target group as the final broadband potential users.

[0091] Optionally, after determining all users in the third target group as the final broadband potential users, the method further includes:

[0092] Formulating corresponding broadband marketing strategies according to the location scene attributes of the night residential communities where the final broadband potential users are resident.

[0093] In this embodiment, the location scene attributes of the night residential community include the nature of the residence (such as residential community, rural community, apartment dormitory, etc.), campus area (such as college campus area, etc.), and industrial park (such as factory park, commercial building, etc.). Since the broadband marketing strategies and methods for different location scene attributes are different, corresponding broadband marketing strategies can be formulated based on the location scene attributes of the night residential community to achieve precise marketing.

[0094] It should be noted that the method for mining potential broadband users provided by the present invention starts from the mobile network positioning ability and combines big data modeling to identify valuable broadband users, thus solving the problems of inaccurate target users and low accuracy in identifying potential broadband users in current broadband marketing.

[0095] In a specific embodiment, in order to realize the mining of potential broadband users, accurately output potential broadband users, and guide broadband business marketing, a method for mining potential broadband users based on accurate mobile network positioning is provided. The method mainly includes the following steps:

[0096] 1. Screen high-quality broadband marketing users from mobile network existing users

[0097] First, initially identify potential broadband users from mobile network existing users through machine learning algorithms. The machine learning algorithm adopts a supervised learning algorithm, and decision trees or random forests can be used. Supervised learning is to train the algorithm through a labeled data set so that the algorithm can perform accurate classification or prediction functions. An input object and an expected output value are used as an instance, and the input object contains various parameters and features of the object. The supervised learning algorithm analyzes and trains this part of the data to obtain a judgment ability for mapping new instances. Thus, by selecting some features of existing broadband users for model training, high-quality broadband marketing users among non-broadband users are mined. This part of users is the first target group.

[0098] 2. Accurately locate mobile network users' permanent residence in night residential communities

[0099] This step provides a method for identifying users' permanent residence for multiple days to calculate users' multi-day permanent residence information, and locates users' permanent residence in night residential communities through mobile network call records and signaling.

[0100] 2.1 Statistically analyze the information reported by user terminals (such as mobile phones), including the residence information from 0:00 to 8:00 and 21:00 - 24:00 every day, including the residence area (i.e., the base station cell where the user stays, and the base station cell can be associated with the province, city, district, longitude, and latitude, etc. where the base station cell is located through the operator's cell engineering parameters) and the residence duration. Select the residence area where the residence duration of the user at a certain residence location reaches a certain number of hours (such as 3 hours) every day and the number of residence days > 1 day within the cycle time. When the user has multiple permanent residences (i.e., permanent night base station cells), further screen the family permanent residence (i.e., family permanent base station cell).

[0101] 2.2 Determine the user's residence frequency by day granularity, that is, how often the user comes to a certain permanent night base station cell on average, and try to select the permanent night base station cell with a smaller interval days for the user.

[0102] For example, an average number of days between stays can be calculated as the dwelling frequency. Count the total number of days since the start of the stay until the current time. That is, if the dwelling frequency in Area A = 7 / 3 ≈ 2.3 days and the dwelling frequency in Area B = 7 / 4 = 1.75 days. Then select Area B.

[0103]

[0104] 2.3 Statistically analyze the permanent night base station cells where the user sets the dwelling days to be greater than a certain number of days according to the requirements. Incorporate the permanent night base station cells where the user's location has not been updated for more than a month into the historical data, and at the same time roll the history to calculate the dwelling days of each permanent night base station cell.

[0105] 2.4 Determine the time when the user last arrived at the permanent residence.

[0106] 2.5 According to the importance level: dwelling frequency > dwelling days > the time of the last arrival, select the user's unique home permanent base station cell. That is, select the permanent night base station cell with the shortest dwelling interval time (i.e., high dwelling frequency), longer dwelling days, and the one the user has visited recently as the home permanent base station cell.

[0107] 2.6 According to the relationship between the home permanent base station cell and the border of the actual residential broadband electronic fence, place the home permanent base station cell within the broadband electronic fence. If the home permanent base station cell is completely contained within the broadband electronic fence, then directly determine that this home permanent base station cell belongs to this broadband electronic fence. If this home permanent base station cell is not within any electronic fence, then it is determined according to the percentage of the electronic border covered by the signal range of this home permanent base station cell in the cell signal range. Select the broadband electronic fence with the largest percentage and a percentage greater than 30% as the corresponding broadband electronic fence for this home permanent base station cell.

[0108] For example, the schematic diagram of the relationship between the base station coverage and the actual residential broadband electronic fence can be as Figure 2 shown, where the coverage range of the base station cell (represented by a sector) partially overlaps with the electronic fence of Community A (represented by a quadrilateral). As Figure 2 shown, 50% of the coverage range of the base station cell is within Community A. Therefore, users staying at this base station cell will be located in Community A. This is based on the rule we set, that is, if the intersection percentage of the signal coverage range of the user's home permanent base station cell and the broadband electronic fence of a certain residential community exceeds 30%, then it is considered that this base station cell belongs to this residential community. This 30% threshold is set to ensure the accuracy of positioning, but in actual applications, according to specific requirements, this threshold can be adjusted, or even this condition can be reduced or removed. If the intersection percentage of the coverage range of the base station cell and the electronic fences of all residential communities does not reach 30%, then the users of this base station cell will have no clear positioning information.

[0109] Thus, the positioning of the home locations of the first target group of users is completed.

[0110] It should be noted that there are multiple broadband electronic fences, and one broadband electronic fence is one area (such as a residential community, a commercial building, a company park, etc.). "Does not exist" means that the location of the base station cell is not within the border range. Therefore, it is necessary to determine which broadband electronic fence the coverage area of the base station cell covers.

[0111] 3. Match the broadband coverage situation in the residential community where the user resides at night

[0112] According to the broadband construction situation, match the actual community of the user with the broadband electronic fence situation, and obtain the nature of the user's resident location (i.e., the location scenario attribute). For example, the residential nature includes residential communities, rural communities, apartment dormitories, etc.; the campus area includes university campuses, etc.; the industrial park includes factory parks, commercial buildings, etc. At the same time, obtain the broadband coverage situation of the user's resident location, which is divided into full coverage, partial coverage, single-point coverage, and no coverage. No coverage means that there is no broadband coverage within the area (broadband electronic fence area), and these users are directly excluded; single-point coverage means that there is broadband coverage in a certain building within the broadband electronic fence area, and these users are directly excluded; partial coverage means that most of the area is covered, and these users are retained; full coverage means that the area is completely covered by broadband, and these users are retained.

[0113] Thus, the second target user group is composed of users with full coverage and partial coverage. The nature of the resident location is provided to the marketing department for formulating marketing strategies by classification.

[0114] 4. Count the user's online time and active days to obtain the third target group

[0115] Based on the second target group, count the online time (i.e., the time when the user uses the mobile phone, which is also the active time) and active days of this part of users. Define that the active days of the user within the preset period need to be greater than a certain number of days, and the active time needs to be greater than a certain number of hours for user quantity screening, and define the third target user group to guide broadband marketing.

[0116] The method for mining broadband potential users provided by the embodiments of the present invention first screens out broadband potential users from the existing mobile network users to obtain a first target group; then, based on the positioning technology of the mobile network and user behavior analysis, identifies the permanent night residential communities of each user in the first target group; and according to the broadband coverage in the permanent night residential communities of each user in the first target group, screens out users with full coverage and partial coverage of broadband to obtain a second target group; then determines a third target group according to the active time and active days of each user in the second target group within a preset period; and determines all users in the third target group as the final broadband potential users. The present invention effectively screens out real broadband potential users by comprehensively using mobile network positioning technology and user behavior analysis, accurately identifying the permanent night residential communities of users, and combining broadband coverage and user activity data. The present invention not only screens out broadband potential users from a data perspective, but also fully considers the actual living conditions and active behaviors of users, thereby effectively improving the accuracy rate of mining broadband potential users. It solves the problem that the existing methods for mining broadband potential users only analyze users from a data perspective and fail to fully consider the actual situation of users, resulting in a low accuracy rate of mining broadband potential users.

[0117] Embodiment 2:

[0118] As Figure 3 shown, the present embodiment provides a device for mining broadband potential users, which is used to execute the above-mentioned method for mining broadband potential users, and includes:

[0119] A first screening module 11, configured to screen out broadband potential users from the existing mobile network users to obtain a first target group;

[0120] A first identification module 12, connected to the first screening module 11, and configured to identify the permanent night residential communities of each user in the first target group based on the positioning technology of the mobile network and user behavior analysis;

[0121] A second screening module 13, connected to the first identification module 12, and configured to screen out users with full coverage and partial coverage of broadband according to the broadband coverage in the permanent night residential communities of each user in the first target group to obtain a second target group;

[0122] A third screening module 14, connected to the second screening module 13, and configured to determine a third target group according to the active time and active days of each user in the second target group within a preset period;

[0123] A potential user determination module 15, connected to the third screening module 14, and configured to determine all users in the third target group as the final broadband potential users.

[0124] Optionally, the first screening module 11 includes:

[0125] A data set acquisition unit, configured to collect relevant data of broadband users and non-broadband users from the existing mobile network users as a data set, and mark the data set;

[0126] A supervised training unit, configured to train a broadband potential user prediction model by using the marked data set based on a preset supervised learning algorithm;

[0127] A model prediction unit, configured to use the trained broadband potential user prediction model to predict the existing mobile network users, identify broadband potential users, and obtain a first target group.

[0128] Optionally, the first identification module 12 includes:

[0129] A first statistics unit, configured to, within a periodic time, according to the terminal reporting information of each user in the first target group, statistics the resident base station cell and the residence duration of each user in a preset night time period every day;

[0130] A first selection unit, configured to, for each user in the first target group, select a resident base station cell where the residence duration of the user is greater than a preset first duration and the number of residence days within the periodic time is greater than a preset first number of days as the user's regular night base station cell;

[0131] A first determination unit, configured to determine the home regular base station cell of each user according to the regular night base station cell of each user in the first target group, as well as the corresponding residence frequency, and / or residence days, and / or the most recent arrival time;

[0132] A second determination unit, configured to determine the regular night residential community of each user in the first target group according to the border relationship between the home regular base station cell of each user and the border of the actual residential broadband electronic fence.

[0133] Optionally, the first determination unit includes:

[0134] A first processing unit, configured to, for a user in the first target group whose regular night base station cell is one, use the regular night base station cell as the corresponding home regular base station cell;

[0135] A second processing unit, configured to, for a user in the first target group whose regular night base station cells are at least two, select a regular night base station cell with a high residence frequency, a large number of residence days, and a recent arrival as the home regular base station cell in the order of residence frequency first, residence days second, and the most recent arrival time third.

[0136] Optionally, the second screening module 13 includes:

[0137] A coverage acquisition unit, configured to acquire the broadband coverage of each user in the first target group in the night residential community where they are resident. The broadband coverage is divided into full coverage, partial coverage, single-point coverage, and no coverage. The single-point coverage means that only one building has broadband coverage within the broadband electronic fence, and the partial coverage means that there are multiple buildings with broadband coverage within the broadband electronic fence but not all buildings are covered.

[0138] An elimination unit, configured to eliminate the users with single-point coverage and no coverage in the broadband coverage from the first target group to obtain the second target group.

[0139] Optionally, the third screening module 14 includes:

[0140] A second statistics unit, configured to count the active time and active days of each user in the second target group within a preset period.

[0141] A user screening unit, configured to screen out the users with active days greater than a preset number of days and active time greater than a preset number of hours from the second target group to obtain the third target group.

[0142] Optionally, the device further includes:

[0143] A marketing module, configured to formulate corresponding broadband marketing strategies according to the location scenario attributes of the night residential community where the final broadband potential users are resident.

[0144] Embodiment 3:

[0145] Reference Figure 4 , this embodiment provides a device for mining broadband potential users, including a memory 21 and a processor 22. A computer program is stored in the memory 21, and the processor 22 is configured to run the computer program to execute the method for mining broadband potential users in Embodiment 1.

[0146] Wherein, the memory 21 is connected to the processor 22. The memory 21 can adopt flash memory, read-only memory, or other memories, and the processor 22 can adopt a central processing unit or a single-chip microcomputer.

[0147] Embodiment 4:

[0148] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method for mining broadband potential users in Embodiment 1 above.

[0149] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information such as computer-readable instructions, data structures, computer program modules, or other data. The computer-readable storage medium includes, but is not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash memory or other memory technologies, CD-ROM (Compact Disc Read-Only Memory), digital versatile disc (DVD) or other optical disc storage, magnetic cassette, tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0150] In summary, the method, device, and readable storage medium for mining broadband potential users provided by the embodiments of the present invention first screen out broadband potential users from the existing mobile network users to obtain a first target group; then identify the permanent night residential communities of each user in the first target group based on the positioning technology of the mobile network and user behavior analysis; and screen out users with full coverage and partial coverage of broadband according to the broadband coverage in the permanent night residential communities of each user in the first target group to obtain a second target group; then determine a third target group according to the active time and active days of each user in the second target group within a preset period; and determine all users in the third target group as the final broadband potential users. By comprehensively using the mobile network positioning technology and user behavior analysis, the present invention accurately identifies the permanent night residential communities of users, and combines the broadband coverage and user activity data to effectively screen out the real broadband potential users. The present invention not only screens out broadband potential users from a data perspective, but also fully considers the actual living conditions and active behaviors of users, thereby effectively improving the accuracy of mining broadband potential users. It solves the problem that the existing methods for mining broadband potential users only analyze users from a data perspective and fail to fully consider the actual situation of users, resulting in a low accuracy of mining broadband potential users.

[0151] It can be understood that the above embodiments are merely exemplary embodiments adopted to illustrate the principles of the present invention, and the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered within the protection scope of the present invention.

Claims

1. A method for mining potential broadband users, characterized in that: The method comprises: Screen out potential broadband users from the existing users of the mobile network to obtain the first target group; Identify the nighttime residential quarters where each user in the first target group resides based on mobile network positioning technology and user behavior analysis; According to the broadband coverage of each user in the first target group who resides in the residential area at night, users with full and partial broadband coverage are screened out to obtain a second target group; Determine a third target group according to the active time and active days of each user in the second target group within a preset period; All users in the third target group are determined as final broadband potential users.

2. The method according to claim 1, characterized in that The step of selecting potential broadband users from the existing users of the mobile network to obtain the first target group specifically includes: Collect relevant data of broadband users and non-broadband users from the existing users of the mobile network as a data set, and mark the data set; Based on a preset supervised learning algorithm, the broadband potential user prediction model is trained using the labeled data set; The trained broadband potential user prediction model is used to predict the mobile network stock users, identify broadband potential users, and obtain a first target group.

3. The method according to claim 1, characterized in that: The mobile network-based positioning technology and user behavior analysis identify the nighttime residential quarters where each user in the first target group resides, specifically including: During the period, according to the terminal reporting information of each user in the first target group, statistics are collected on the base station cell and the length of stay of each user in the preset night time period every day; For each user in the first target group, selecting a resident base station cell in which the user's daily resident duration is greater than a preset first duration and the number of resident days in the cycle time is greater than a preset first number of days as the user's permanent nighttime base station cell; Determine the home resident base station cell of each user according to the resident night base station cell of each user in the first target group and the corresponding resident frequency, and / or resident days, and / or the latest arrival time; The permanent nighttime residential area of ​​each user in the first target group is determined according to the border relationship between the home permanent base station area of ​​each user and the actual residential broadband electronic fence.

4. The method according to claim 3, characterized in that: The step of determining the home resident base station cell of each user according to the resident nighttime base station cell of each user in the first target group and the corresponding resident frequency, and / or resident days, and / or the latest arrival time specifically includes: For users in the first target group who have one permanent nighttime base station cell, the permanent nighttime base station cell is used as the corresponding home permanent base station cell; For users in the first target group who have at least two permanent night base station cells, the permanent night base station cells with high residence frequency, many residence days and recent visits are selected as home permanent base station cells in the order of residence frequency first, residence days second and most recent arrival time.

5. The method according to claim 3, characterized in that: According to the broadband coverage of each user in the first target group who resides in the residential area at night, users with full and partial broadband coverage are screened out to obtain the second target group, specifically including: Obtain broadband coverage conditions in residential quarters where each user in the first target group resides at night, wherein the broadband coverage conditions are divided into full coverage, partial coverage, single-point coverage, and no coverage. Single-point coverage means that only one building within the broadband electronic fence has broadband coverage, and partial coverage means that multiple buildings within the broadband electronic fence have broadband coverage but are not fully covered. The second target group is obtained by eliminating users whose broadband coverage is single-point coverage or no coverage in the first target group.

6. The method according to claim 1, characterized in that The determining of the third target group according to the active time and active days of each user in the second target group within a preset period specifically includes: Counting the active time and number of active days of each user in the second target group within a preset period; The users whose active days are greater than a preset number of days and whose active time is greater than a preset number of hours are screened out from the second target group to obtain the third target group.

7. The method according to claim 1, characterized in that After determining all users in the third target group as final broadband potential users, the method further includes: A corresponding broadband marketing strategy is formulated according to the location scene attributes of the final nighttime residential quarters where the potential broadband users reside.

8. A device for mining potential broadband users, characterized in that: The device comprises: A first screening module is used to screen out potential broadband users from the existing users of the mobile network to obtain a first target group; A first identification module, connected to the first screening module, for identifying the nighttime residential quarters where each user in the first target group resides based on the positioning technology of the mobile network and the user behavior analysis; A second screening module, connected to the first identification module, is used to screen out users with full coverage and partial coverage of broadband coverage according to the broadband coverage of each user in the first target group who resides in the residential area at night, to obtain a second target group; a third screening module, connected to the second screening module, for determining a third target group according to the active time and active days of each user in the second target group within a preset period; The potential user determination module is connected to the third screening module and is used to determine all users in the third target group as final broadband potential users.

9. A device for mining potential broadband users, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the method for mining potential broadband users according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for mining potential broadband users according to any one of claims 1 to 7 is implemented.

Images