Broadband user mobile phone prediction method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a broadband user relocation prediction method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining feature data of a first broadband user in a first time period, the feature data comprising basic attribute data of the broadband user and behavior data related to broadband relocation; according to the feature data of the first broadband user and feature data of second broadband users included in a pre-established user population, a target distance between the first broadband user and the user population is determined, the user population comprises a plurality of second broadband users with similar feature data, and at least part of the second broadband users have broadband relocation requirements; each second broadband user has feature data in a second time period; and predicting whether the first broadband user has a broadband relocation demand according to the target distance. Therefore, according to the embodiment of the invention, the problem that the broadband user loss is easily caused because the broadband relocation demand of the broadband user cannot be predicted in advance in the prior art can be solved.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a method, device, electronic equipment and storage medium for predicting the migration of broadband users. Background Art

[0002] At present, the market competition among communication operators is very fierce, and the overall market of operators has become saturated. In addition, it is often difficult to develop a new user than to retain an old user. Therefore, it is very important to maintain the customer relationship between the users of the network and the enterprise. Among them, the problem of broadband user migration is a good bridge to establish customer relationship.

[0003] According to the analysis, among the operators' broadband users in a single city, at least 100,000 users have the need to move their devices every month, and the annual user relocation demand is close to 1.2 million. Calculated based on a package of 100 yuan per household, the impact on the company's annual revenue is as high as several hundred million yuan.

[0004] Therefore, if we can predict in advance the target users who may need to relocate their devices, provide them with free activities related to relocation escort, and provide them with care and help, we can effectively maintain and establish the relationship between the company and the users.

[0005] However, existing technologies generally directly predict user churn based on data generated within the operator, but there is no targeted research on user migration. Moreover, user migration does not mean user churn, and the response measures or strategies adopted by enterprises are even more different.

[0006] In summary, in the prior art, it is impossible to predict in advance the broadband migration needs of broadband users, which easily leads to the loss of broadband users. Summary of the invention

[0007] The embodiments of the present application provide a method, device, electronic device and storage medium for predicting the migration of broadband users, so as to solve the problem in the prior art that the broadband migration needs of broadband users cannot be predicted in advance, which easily leads to the loss of broadband users.

[0008] In a first aspect, an embodiment of the present application provides a method for predicting the migration of broadband users, the method comprising:

[0009] Acquiring characteristic data of a first broadband user in a first time period, wherein the characteristic data includes basic attribute data of the broadband user and behavior data related to broadband migration;

[0010] Determining a target distance between the first broadband user and the user population according to the characteristic data of the first broadband user and the characteristic data of the second broadband users included in the pre-established user population, wherein the user population includes a plurality of second broadband users with similar characteristic data, at least some of the second broadband users have a broadband migration demand, and each second broadband user has characteristic data within a second time period;

[0011] It is predicted whether the first broadband user has a broadband relocation demand according to the target distance.

[0012] In a second aspect, an embodiment of the present application provides a broadband user migration prediction device, the device comprising:

[0013] A first acquisition module, configured to acquire characteristic data of a first broadband user in a first time period, wherein the characteristic data includes basic attribute data of the broadband user and behavior data related to broadband migration;

[0014] a first determination module, configured to determine a target distance between the first broadband user and the user population according to feature data of the first broadband user and feature data of second broadband users included in a pre-established user population, wherein the user population includes a plurality of second broadband users with similar feature data, at least some of the second broadband users have a need to migrate broadband devices, and each second broadband user has feature data within a second time period;

[0015] The prediction module is used to predict whether the first broadband user has a broadband migration demand according to the target distance.

[0016] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a transceiver, and a processor:

[0017] A memory is used to store computer programs; a transceiver is used to send and receive data under the control of the processor; and a processor is used to read the computer program in the memory and execute the broadband user migration prediction method described in the first aspect above.

[0018] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the broadband user migration prediction method described in the first aspect is implemented.

[0019] In an embodiment of the present application, characteristic data of a first broadband user within a first time period can be obtained, and based on the characteristic data of the first broadband user and the characteristic data of second broadband users included in a pre-established user population, a target distance between the first broadband user and the user population can be determined, thereby predicting whether the first broadband user has a need for broadband relocation based on the target distance; wherein the characteristic data includes basic attribute data of the broadband user and behavior data related to broadband relocation, the user population includes multiple second broadband users with similar characteristic data, and at least some of the second broadband users have a need for broadband relocation, and each second broadband user has characteristic data within a second time period.

[0020] It can be seen that in the embodiment of the present application, a user population can be constructed based on the above-mentioned multiple second broadband users, and at least some of the multiple second broadband users have broadband relocation needs, and each second broadband user has corresponding characteristic data. Among them, the characteristic data includes basic attribute data of broadband users and behavioral data related to broadband relocation, and each second broadband user is similar in basic attributes and behavioral data related to broadband relocation, and in terms of broadband relocation needs. In this way, for any broadband user to be predicted whether there is a need for broadband relocation, the degree of similarity between the broadband user and each second broadband user with broadband relocation needs can be obtained based on the distance between the broadband user and the user population, so that it can be predicted whether the broadband user has a need for broadband relocation.

[0021] It can be seen that the embodiment of the present application provides a method for predicting whether a broadband user has a need to relocate, thereby reducing the loss of broadband users. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0023] Figure 1 A flowchart of a broadband user migration prediction method provided in an embodiment of the present application;

[0024] Figure 2 A flowchart of a specific implementation of the broadband user migration prediction method according to an embodiment of the present application;

[0025] Figure 3 A structural block diagram of a broadband user migration prediction device provided in an embodiment of the present application;

[0026] Figure 4A structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] In the embodiments of the present application, the term "and / or" describes the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0028] In the embodiments of the present application, the term "plurality" refers to two or more than two, and other quantifiers are similar.

[0029] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0030] The embodiments of the present application provide a method and device for predicting the migration of broadband users, so as to solve the problem in the prior art that the broadband migration needs of broadband users cannot be predicted in advance, which easily leads to the loss of broadband users.

[0031] Among them, the method and the device are based on the same application concept. Since the method and the device solve the problem in a similar principle, the implementation of the device and the method can refer to each other, and the repeated parts will not be repeated.

[0032] Figure 1 A schematic diagram of a process flow of a broadband user migration prediction method provided by an embodiment of the present application is shown. Figure 1 As shown, the method may include the following steps 101 to 103:

[0033] Step 101: Acquire characteristic data of a first broadband user in a first time period.

[0034] The characteristic data includes basic attribute data of broadband users and behavior data related to broadband migration. It should be noted that the behavior data related to broadband migration can also be understood as behavior preference data related to broadband migration.

[0035] In addition, the first broadband user may include one or more broadband users, which can be understood as: the first broadband user includes one or more broadband users to be predicted.

[0036] In addition, the first time period may be preset, for example, set to the latest six months, or the latest three months.

[0037] Step 102: Determine a target distance between the first broadband user and the user population based on the characteristic data of the first broadband user and the characteristic data of second broadband users included in the pre-established user population.

[0038] The user population includes a plurality of second broadband users with similar characteristic data, and at least some of the second broadband users have a need to migrate broadband devices, and each second broadband user has characteristic data within a second time period.

[0039] It is understandable that whether the feature data of different second broadband users are similar can be determined based on the similarity between the feature data, which can be determined by the distance between the feature data (such as Euclidean distance) or by the value of the specific content included in the feature data.

[0040] It can be seen that in the embodiment of the present application, a user population can be constructed in advance based on the above-mentioned multiple second broadband users, and at least some of the multiple second broadband users have broadband migration needs, and each second broadband user has corresponding characteristic data. Among them, the characteristic data includes basic attribute data of broadband users and behavior data related to broadband migration. Then, each second broadband user is similar in basic attributes and behavior data related to broadband migration, and in terms of broadband migration needs, that is, it can also be understood that the user population is a group of people with similar basic attribute data and behavior data related to broadband migration, and in terms of broadband migration needs.

[0041] Therefore, the similarity between the first broadband user and the above user population indicates the similarity between the first broadband user and a user group with similar broadband migration needs. Thus, based on the similarity between the first broadband user and the user population, it can be predicted whether the first broadband user has a broadband migration need.

[0042] Step 103: predicting whether the first broadband user has a broadband migration demand according to the target distance.

[0043] Among them, broadband relocation refers to the situation where a user submits an application for relocation service to the operator company when the broadband installed by the user needs to be relocated to a new installation address. The operator company will promptly relocate the address of the broadband service for the user based on the user's requirements and resource conditions.

[0044] In addition, when it is predicted that the first broadband user has a need to relocate broadband, the first broadband user who has a need to relocate broadband can be output so that relevant messages about the broadband relocation service can be pushed to the first broadband user, such as prompting the relevant procedures for handling the relocation service, or the communication number bound to the first broadband user can be manually dialed to inquire whether the customer needs to handle the broadband relocation service, thereby helping to reduce customer churn.

[0045] It can be seen from the above steps 101 to 103 that in the embodiment of the present application, the characteristic data of the first broadband user in the first time period can be obtained, and based on the characteristic data of the first broadband user and the characteristic data of the second broadband users included in the pre-established user population, the target distance between the first broadband user and the user population is determined, thereby predicting whether the first broadband user has a need for broadband relocation based on the target distance; wherein the characteristic data includes basic attribute data of the broadband user and behavior data related to broadband relocation, the user population includes multiple second broadband users with similar characteristic data, and at least some of the second broadband users have a need for broadband relocation, and each second broadband user has characteristic data in the second time period.

[0046] It can be seen that in the embodiment of the present application, a user population can be constructed based on the above-mentioned multiple second broadband users, and at least some of the multiple second broadband users have broadband relocation needs, and each second broadband user has corresponding characteristic data. Among them, the characteristic data includes basic attribute data of broadband users and behavioral data related to broadband relocation, and the basic attributes and behavioral data related to broadband relocation of each second broadband user are similar and related in terms of broadband relocation needs. In this way, for any broadband user to be predicted whether there is a need for broadband relocation, the similarity between the broadband user and each second broadband user with broadband relocation needs can be obtained based on the distance between the broadband user and the user population, so that it can be predicted whether the broadband user has a need for broadband relocation.

[0047] It can be seen that the embodiments of the present application provide a method for predicting whether broadband users have a need to relocate their devices, thereby reducing the loss of broadband users and increasing corporate revenue.

[0048] Optionally, the basic attribute data includes at least one of the following:

[0049] Broadband account, target communication number bound to the broadband account, age of broadband user, broadband size, broadband online time, broadband package balance, and broadband migration time.

[0050] The target communication number bound to the broadband account may include at least one of a mobile communication number and a landline telephone number.

[0051] In addition, the above basic attribute data can be obtained through the operator's internal system.

[0052] It should be noted that the historical record data left by broadband users before the mobile phone migration behavior is very important. Therefore, the basic attribute data generated by the operator's internal system in the first time period (eg, the last six months) can be obtained as part of the feature data.

[0053] It is understandable that if the above-mentioned contents included in the basic attribute data are empty, it means that the corresponding contents do not exist. For example, if the broadband relocation time is "empty", it means that the corresponding broadband user has not applied for the broadband relocation service in the first time period; if the broadband relocation time is not "empty", for example, it is a time, it means that the corresponding broadband user has applied for the broadband relocation service during the time.

[0054] Optionally, the behavior data includes at least one of the following items A-1 to A-7:

[0055] Item A-1: ​​The first indication information is used to indicate whether the target communication number bound to the broadband account has dialed the communication number of the broadband customer service.

[0056] After obtaining the target communication number bound to the broadband account, the operator's internal system records can be used to check whether the target communication number has called the customer service number in the first time period. For example, if it has, it can be marked as 1, otherwise, it can be marked as 0.

[0057] It should be noted that users who have broadband relocation needs may call the customer service number to inquire about the broadband relocation service before processing the broadband relocation needs. Therefore, the above-mentioned first indication information (i.e., whether the target communication number bound to the broadband account has called the broadband customer service number) can also be used as partial characteristic data to determine whether the broadband user has broadband relocation needs.

[0058] Item A-2: The second indication information is used to indicate whether the broadband user has visited the mobile interface of the broadband application APP.

[0059] The operator's internal system records may be used to check whether the broadband user has visited the broadband APP migration interface within the first time period. For example, if the broadband user has visited the broadband APP migration interface, it may be marked as 1, otherwise, it may be marked as 0.

[0060] It should be noted that users who have broadband relocation needs may have visited the relocation interface of the broadband APP before handling the broadband relocation needs to understand how to handle the broadband relocation service. Therefore, the above second indication information (i.e., whether the broadband user has visited the relocation interface of the broadband APP) can also be used as part of the characteristic data for judging whether the broadband user has broadband relocation needs.

[0061] Item A-3: The third indication information is used to indicate whether there is call record data between the target communication number and a specific communication number at the residence of the broadband user, wherein the specific communication number includes at least one of a moving communication number, a renting communication number, and a buying communication number.

[0062] The call bill data records the user's communication behavior, including making calls, sending text messages, etc. The third indication information is used to indicate whether the target communication number has contacted a specific communication number at the residence of the broadband user.

[0063] It should be noted that the communication number mentioned in this article may include at least one of a mobile communication number (such as a mobile phone number) and a fixed telephone number.

[0064] Optionally, the process of determining the third indication information includes:

[0065] Obtaining call record data of the target communication number within a first time period;

[0066] Determine whether the target communication number includes a communication number in the first table in the call list data within the first time period, thereby obtaining the third indication information;

[0067] Among them, the acquisition process of the first table includes: obtaining a specific communication number whose location is the residence of the broadband user; according to the user data of the specific communication number in the operator, eliminating invalid communication numbers (such as communication numbers in arrears, dismantled, shut down, etc.) to obtain a second communication number; through the operator's factor verification interface, eliminating abnormal communication numbers (such as untrustworthy and illegal communication numbers); saving the remaining specific communication numbers whose location is the residence of the broadband user in the first table.

[0068] In order to facilitate understanding of the determination process of the third indication information, the following specific implementation methods for determining the third indication information are now introduced:

[0069] If broadband users need to move their phones, they will most likely be connected to local numbers related to moving, renting, buying a house, etc. Therefore, the Internet, crawlers and other technologies can be used to batch obtain the communication numbers related to moving, renting, buying a house, etc., which are at the residence of broadband users, and save such numbers in the database; then, through big data analysis, these numbers are analyzed and compared with the operator's user data to eliminate invalid users such as those who owe fees, dismantle their phones, or shut down their phones; finally, through the element verification interface, abnormal numbers such as dishonesty and fraud are eliminated to complete the data cleaning of communication numbers related to moving, renting, buying a house, etc.

[0070] Afterwards, a dimension table of mobile communication numbers for moving, renting, and buying a house at the place of residence of broadband users and a dimension table of fixed-line telephone numbers for moving, renting, and buying a house at the place of residence can be established based on the cleaned communication numbers. The two dimension tables are saved in the database, solidified into an automatically running model through script code, and regularly maintained and updated to ensure the freshness of the data. At this point, the acquisition of valid communication numbers for moving, renting, and buying a house at the place of residence of broadband users is completed.

[0071] Furthermore, the call record data of the target communication number can be obtained, and the data within the first time period can be cleaned based on the data generation time. Then, based on the target communication number, big data analysis technology is used to perform relevant analysis on the calling, called, and text message exchanges based on the cleaned operator call record data.

[0072] Among them, according to the analysis results, it is determined whether the target communication number has had contact with the numbers in the two dimension tables of moving, renting, and buying houses where the broadband user lives in the first time period. If there is contact, it is marked as 1, otherwise, it is marked as 0. It is understandable that the number of calls, call duration, etc. can also be counted.

[0073] Item A-4: The fourth indication information is used to indicate whether the broadband user has visited a specific website address at the broadband user's residence, wherein the specific network address includes at least one of a moving website address, a renting website address, and a buying website address.

[0074] Item A-5: The fifth indication information is used to indicate whether the broadband user has visited a page related to the broadband user's residence in a specific APP, wherein the specific APP includes at least one of a moving APP, a rental APP, and a home buying APP.

[0075] The following is an introduction to the process of obtaining the contents of A-4 and A-5:

[0076] In the Internet age, we cannot do without various network applications, such as websites or apps for buying, renting, moving, etc. Visiting them will generate message data. Message data refers to the data unit transmitted in the network. It contains information such as the source address, destination address, type, etc. The operator's unique message data has very great value.

[0077] Deep packet inspection (DPI) technology can obtain user behavior information, such as visited websites, used applications, online time, etc. through in-depth analysis of network data packets. It plays an important role in analyzing broadband mobile user behavior.

[0078] If a broadband user has never visited websites or apps for buying, renting, moving, etc. before, and now suddenly visits such websites or apps, then there is a high probability that the broadband user will need to relocate, and relocation also means the need to move the machine.

[0079] Therefore, it is possible to use the Internet, crawlers and other technologies to batch obtain websites or apps for renting, buying, moving, etc. in the user's place of residence, save the various websites and apps obtained into a database, and then perform data cleaning work such as inspection, deduplication, and classification on them.

[0080] Based on the cleaned website and APP data, a dimension table of websites related to moving, renting and buying houses at the place of residence of broadband users and a dimension table of APPs related to moving, renting and buying houses at the place of residence are established. The two dimension tables are saved in the database, solidified into an automatically running model through script code, and regularly maintained and updated to ensure the freshness of the data. At this point, the data acquisition of websites and APPs related to moving, renting and buying houses at the place of residence of broadband users is completed.

[0081] After that, DPI can be used to parse the messages of websites, apps and other online behaviors of broadband users, and the parsing results can be compared with the two dimension tables of websites and apps to determine whether the broadband user has visited the websites or apps recorded in the dimension tables of house purchase, house rental, and house moving in the first time period. If so, it will be marked as 1, and the access time, number of visits, and access duration will be counted. Finally, the data with the access mark and the access time difference from the current time less than three months can be used as the basis.

[0082] Item A-6: Sixth indication information, used to indicate whether the residential area of ​​the broadband user is in a demolished community or a newly built community where the broadband user lives;

[0083] Item A-7: The seventh indication information is used to indicate whether the broadband installation address of the broadband user is located in the demolished community where the broadband user resides.

[0084] Optionally, the process of determining the residential area of ​​the broadband user includes:

[0085] Acquiring base station access information of the target communication number within a third time period;

[0086] Determining, according to the base station access information, a target base station whose access frequency exceeds a second threshold;

[0087] The residential area of ​​the broadband user is determined according to the coverage area of ​​the target base station.

[0088] The following is an introduction to the process of obtaining the contents of A-6 and A-7:

[0089] If the residential address of a broadband user is a demolished community or a newly built community in the local city, or if the installation address of a broadband user is a demolished community, then the broadband user will most likely have a need to move the device.

[0090] Therefore, the names of the demolished and newly built communities and their corresponding addresses in the user's place of residence can be obtained in batches through the Internet, crawlers and other technologies, and the obtained demolished and newly built community names and addresses can be saved in the database, and then deduplication, verification and other data cleaning work can be performed on them.

[0091] Based on the cleaned community name and address data, a dimension table of demolished communities and a dimension table of newly built communities where broadband users live are established, and the two dimension tables are saved in the database. They are solidified into an automatically running model through script code, and regular maintenance and updates are carried out to ensure the freshness of the data. At this point, the data acquisition of newly built and demolished communities where broadband users live is completed.

[0092] Among them, the base station signaling data is the record information generated by the communication between the terminal and the mobile network base station, and these record information can provide the user location. Therefore, the access information of broadband users to the base station in the third time period (for example, the last month) (for example, from 10 pm to 5 am) can be analyzed through signaling data, and the most frequently visited one or more base station ranges are extracted as the residential range of the broadband user, and combined with the operator's base station information, the longitude and latitude of the base station are converted into the residential range of the broadband user. Furthermore, it can be determined whether the residential range of the broadband user is located in the demolished or newly built community recorded in the community dimension table. If so, it is marked as 1, otherwise it is marked as 0. It should be noted here that users who live in newly built communities or demolished communities at night have a tendency to move their phones, so they can both be marked as 1.

[0093] In addition, the structured query language can be used to obtain the broadband user's installation address in the operator. Furthermore, it can be determined whether the installation address is located in the demolished community recorded in the community dimension table. If so, it is marked as 1, otherwise it is marked as 0. It should be noted here that if the user's broadband installation address is in a demolished community, it means that there may be a tendency to move the machine, which can be marked as 1; for the user's broadband installation address, there is no need to consider the newly built community, because if the broadband installation address is in a newly built community, it means that the customer has completed the move and will not have a tendency to move in the short term.

[0094] It should be noted that the various contents included in the behavior data of the first broadband user and the various contents included in the behavior data of the second broadband user can be obtained in the manner described above.

[0095] In summary, as described in A-1 to A-7, in the embodiments of the present application, high-probability mobile behavior preference data (i.e., the above-mentioned behavior data) of broadband users in several dimensions such as call records, messages, and residential areas are obtained. After analyzing and sorting these behavior preference data, the basic attribute data of the user in the system generated by the network and the mobile behavior preference data (i.e., the above-mentioned behavior data) generated by the user's residence can be combined to form a high-quality feature data after multi-source data fusion processing, as shown in Table 1.

[0096] Table 1 Characteristic data examples

[0097]

[0098]

[0099]

[0100] It should be noted that, when the characteristic data includes the contents in Table 1, the similarity of the characteristic data of each second broadband user in the user population can be determined according to the specific values ​​of the contents included in the characteristic data. For example, it can be considered that the second broadband users whose broadband migration time is not empty are similar; or it can also be considered that: the second broadband users whose at least one of C-8 to C-20 is "yes" are similar.

[0101] Optionally, the process of constructing the user population includes the following steps B-1 to B-3:

[0102] Step B-1: selecting a plurality of broadband users that meet a first condition to construct a first seed user group, wherein the first condition includes that the broadband migration time is not empty;

[0103] Step B-2: selecting a plurality of broadband users meeting a second condition to construct a second seed user group, wherein the second condition includes that the broadband migration time is empty, and at least one indication information from the first indication information to the seventh indication information is a positive indication (i.e., at least one indication information indicates "yes");

[0104] Step B-3: Merge the first seed user group and the second seed user group to obtain the user population.

[0105] It can be seen from steps B-1 to B-3 that the user population in the embodiment of the present application includes two levels of seed user groups (ie, a first seed user group and a second seed user group). The reasons and process for constructing two levels of seed user groups are introduced below.

[0106] 1. Reasons

[0107] In the embodiment of the present application, the core idea of ​​population expansion is to construct personalized features based on the behavioral habits of user populations that are different from other users, and use the personalized features to calculate the similarity with the broadband users to be predicted, so as to predict that users with high similarity have the need for broadband migration. Among them, in the transition stage between the cold start and the hot start of the broadband migration marketing campaign, there are a small number of real customers, but the small number is not enough to support the prediction of whether broadband customers have the need for broadband migration by constructing the above user population.

[0108] In addition, through the analysis of the current existing customers who have moved their phones, the number of customers who have moved their phones is small and the proportion is low. The relevant samples are not enough to support the prediction of whether broadband customers have the need to move their phones by building the above user population. At the same time, users will initiate a move application in a relatively short period of time after they have the need to move their phones, so the timeliness of the move prediction is relatively high.

[0109] Based on the above analysis, in order to take into account both timeliness and accuracy requirements, on the basis of the expansion of traditional similar target populations, in the embodiment of the present application, a semi-supervised multi-level target population can be constructed to predict the migration of broadband customers and form a gradient prediction capability.

[0110] 2. Introduction to the construction process

[0111] First, the broadband users who actually moved their phones were taken as the first-level core seeds of the migration (i.e., as the first seed user group). The specific construction method is to select broadband users whose migration time is not empty to construct the first seed user group G based on the feature data after the multi-source data fusion processing constructed in the early stage (i.e., as shown in Table 1). base1 .

[0112] Secondly, based on the feature data after the multi-source data fusion processing constructed in the early stage, the behavior habits and behavior patterns of broadband users in different scenarios are analyzed, and broadband users with at least one item of C-8 to C-20 in Table 1 as "yes" are selected to construct the second sub-user group G base2 It is understandable that G base2 There will be noise in the data, that is, some users will not actually move their devices.

[0113] Finally, the number of broadband users who actually migrated (i.e., G base1 The number of users in the user group is small, and it is not easy to obtain reliable results by directly using it for machine migration prediction. In order to improve the prediction accuracy and generalization, in the embodiment of the present application, the semi-supervised learning idea is used to refer to the union of the first seed user group and the second seed user group to form a new user population G base , that is: G base =G base1 ∪G base2 .

[0114] In addition, the above user population G is obtained base After that, G base The user population as the population expansion is used to calculate the similarity with the broadband users to be predicted for migration, so as to realize the prediction of broadband migration customers based on the similarity.

[0115] It is understandable that G can be updated according to actual conditions. base, in order to keep the data fresh and thus ensure the accuracy of the machine relocation prediction.

[0116] It should also be noted that, unlike the traditional marketing-based population expansion model, the target results will appear repeatedly, such as repeated consumption and repeated purchases. Broadband relocation is basically a one-time behavior over a large time span, that is, the relocation will not be repeated in a short period of time after the relocation occurs. Therefore, the subsequent behavior of the customer after the relocation cannot be used as the feature data of the second broadband user in the user population. Therefore, for each data in Table 1, if the "broadband relocation time (i.e., item C-7)" of a broadband user is not empty, then the other features of the broadband user (i.e., C-1 to C-6, C-8 to C-20) need to be collected before the broadband relocation time to ensure the validity of the feature data, thereby improving the accuracy of the relocation prediction.

[0117] Optionally, in the above step 102, determining the target distance between the first broadband user and the user population according to the characteristic data of the first broadband user and the characteristic data of the second broadband user included in the pre-established user population comprises the following steps D-1 to D-2:

[0118] Step D-1: determining characteristic data of a center point of the user population according to characteristic data of each second broadband user in the user population;

[0119] Step D-2: Determine the distance between the characteristic data of the first broadband user and the characteristic data of the center point, and determine the obtained distance as the target distance.

[0120] Optionally, the target distance is a Euclidean distance. It is understandable that the target distance may also be other distances used to represent similarity, which are not listed here one by one.

[0121] It should be noted that if the first broadband user includes m broadband users to be predicted, the user population G base Including n second broadband users, then, in calculating the m broadband users to be predicted and G base When the target distance is large, if the number m is huge, a lot of computing resources will be consumed and the computational complexity O(nm) will be very high.

[0122] In order to reduce the computational complexity, improve efficiency and save computational resources, in the embodiment of the present application, G base The center point of G base All second broadband users in , calculate the distance between each broadband user to be predicted as G baseThe similarity measurement index between the broadband customers to be predicted reduces the computational complexity from O(nm) to O(m). It can be seen that the computational complexity is significantly reduced.

[0123] In addition, in step D-1, to facilitate the subsequent distance calculation, G base The characteristic data of each second broadband user in is mapped into a multidimensional vector space, thereby being converted into a vector representation form, such as G base The feature vector of the i-th second broadband user in is recorded as G base The characteristic vector of the center point is recorded as: Y = [y1, y2, ... y p ]; p represents the dimension of the feature included in the feature data of the broadband user; then according to the midpoint coordinate formula of the space vector, we can get:

[0124] Represents G base The value of the feature of the jth dimension in the feature vector of the i-th second broadband user, y j Represents G base The value of the feature of the j-th dimension in the feature vector of the center point.

[0125] Optionally, in step D-2, calculate G base The distance d between the feature vector Y of the center point and the kth broadband customer to be predicted k , k∈(0,m], as an indicator for subsequent judgment of whether there is a need for relocation. Among them, the Euclidean distance can be used as the distance calculation method for calculation, and the distance calculation formula is as follows:

[0126] Z k surface

[0127] represents the feature vector of the kth broadband user to be predicted, Represents the value of the feature of the jth dimension in the feature vector of the kth broadband user to be predicted.

[0128] Optionally, before the above step 102, the method further includes:

[0129] The characteristic data of the first broadband user and the characteristic data of each second broadband user in the user population are normalized.

[0130] It should be noted that there are different dimensions in the various contents included in the feature data. If not processed, when calculating the distance between broadband users, the calculated distance result will be biased towards the feature with a larger value range, causing the calculated distance to fail to fully consider all features, and ultimately leading to unreliable distance results. Therefore, in order to eliminate the dimensional relationship between the various feature dimensions of broadband mobile users and make the data comparable, in an embodiment of the present application, the feature data of the first broadband user and the feature data of each second broadband user in the user population can be normalized.

[0131] It can be understood that the characteristic data of each second broadband user in the user population can be normalized when the user population is constructed or updated, so that the characteristic data of each second broadband user in the saved user population is normalized. Then, when performing machine relocation prediction, the normalized data can be directly used, thereby shortening the machine relocation prediction time.

[0132] Among them, let the feature dimension included in the feature data be p, and f j represents the eigenvalue of the jth dimension of a broadband user, μ j represents the average value of the feature of the jth dimension, ρ j represents the standard deviation of the feature of the jth dimension, j∈(0,p]; then the normalization formula of the feature is as follows:

[0133] f j ′ represents f j The value after normalization.

[0134] Optionally, predicting whether the first broadband user has a broadband relocation demand according to the target distance includes the following steps E-1 to E-4:

[0135] Step E-1: determining characteristic data of a center point of the user population according to characteristic data of each second broadband user in the user population;

[0136] Step E-2: determining, according to the characteristic data of each second broadband user in the user population, a third broadband user farthest from the center point among each of the second broadband users, and determining the distance between the characteristic data of the third broadband user and the characteristic data of the center point as a first threshold;

[0137] Step E-3: When the target distance is less than or equal to the first threshold, determining that the first broadband user has a broadband migration demand;

[0138] Step E-4: when the target distance is greater than the first threshold, determine that the first broadband user does not have a broadband migration demand.

[0139] It can be seen from steps E-1 to E-4 that in this embodiment of the present application, the first threshold can be determined based on the characteristic data of each second broadband user in the user population, and then the target distance and the first threshold can be compared to determine whether the first broadband user has a need to move.

[0140] The third broadband user farthest from the center point in the user population is defined as L max , then L max The calculation formula is as follows:

[0141] Represents G base The value of the feature of the jth dimension in the feature vector of the i-th second broadband user, y j Represents G base The value of the feature of the j-th dimension in the feature vector of the center point.

[0142] If the distance d between the kth broadband user to be predicted and the center point k (Z k ,Y) is less than or equal to L max , it means that the kth broadband user to be predicted is a customer with a tendency to migrate broadband devices. Otherwise, the kth broadband user to be predicted has no tendency to migrate broadband devices.

[0143] Understandably, d k (Z k The smaller the ,Y), the higher the tendency of broadband migration is, which needs to be paid special attention during marketing.

[0144] Optionally, predicting whether the first broadband user has a broadband relocation demand according to the target distance includes:

[0145] When the target distance is less than or equal to a preset third threshold, determining that the first broadband user has a broadband relocation demand;

[0146] When the target distance is greater than the third threshold, it is determined that the first broadband user has no broadband migration demand.

[0147] It can be understood that the third threshold can be set and adjusted according to actual conditions.

[0148] In addition, it should be noted that all actions to obtain data in this application (for example, the above-mentioned basic attribute data and behavioral data) are carried out in compliance with the relevant data protection laws and policies of the country where the application is located, and with the authorization given by the owner of the corresponding device.

[0149] In summary, the specific implementation of the broadband user migration prediction method of the embodiment of the present application can be described as follows:

[0150] like Figure 2 As shown, by taking advantage of the operator's data, through data mining and artificial intelligence technology, the basic attribute data of broadband users in the system and the behavior data related to broadband migration generated by the network are fused and processed from multiple sources to obtain the characteristic data of broadband users, thereby constructing the overall characteristic dimensions contained in the characteristic data of each broadband user. It should be noted that the specific content and acquisition method of the basic attribute data and the behavior data can be found in the previous article and will not be repeated here.

[0151] Then, based on the characteristic data of broadband users, broadband users that meet the first condition are selected to construct a first seed user group, and broadband users that meet the second condition are selected to construct a second seed user group; and then the first seed user group and the second seed user group are merged to obtain a user population. It should be noted that the specific contents of the first condition and the second condition can be found in the above description, and will not be repeated here.

[0152] Secondly, for any broadband user to be predicted, its characteristic data is obtained, so as to perform the following prediction process based on the above user population:

[0153] Determine the characteristic data of the center point of the user population;

[0154] Calculating a target distance between characteristic data of a broadband user to be predicted and characteristic data of a center point;

[0155] Determine a user in the user population that is farthest from the center point, and determine the distance between the user and the center point as a first threshold;

[0156] If the target distance is less than or equal to the first threshold, it is determined that the broadband user to be predicted has a broadband migration demand;

[0157] If the target distance is greater than the first threshold, it is determined that the broadband user to be predicted does not have a broadband migration demand.

[0158] For example, if the broadband user to be predicted is user u1, the first threshold = 1, and the target distance = 0.8, it means that user u1 is a customer with a tendency to migrate to broadband.

[0159] If the broadband user to be predicted is user u2, the first threshold = 1, and the target distance = 0.2, then user u2 is a customer with a tendency to migrate broadband devices, and the probability of the tendency to migrate broadband devices is very high;

[0160] If the broadband user to be predicted is user u3, the first threshold=1, and the target distance=1.5, it means that user u3 has no tendency to migrate broadband.

[0161] In summary, the embodiments of the present application have the following advantages:

[0162] 1. The demand for broadband user migration is huge. The existing technology directly predicts the loss of broadband users based on the data generated within the operator, but there is no targeted research on user migration. The embodiment of the present application targets the broadband migration problem and conducts targeted analysis after multi-source data analysis and fusion processing. That is, the potential rules of broadband migration users are mined through massive data, and the migration prediction of broadband users is realized by using the rules.

[0163] 2. Make full use of the operator's unique data advantages to obtain data on broadband users in several dimensions such as call records, messages, and residential areas generated on this network, and use relevant technologies to batch obtain broadband relocation-related behavioral data such as moving, renting, buying, demolishing communities, and building new communities in the user's place of residence. Save the information in the database, solidify it into an automatically running model through script code, and perform regular maintenance and updates. Combine the basic attribute data in the system generated by users on this network with the above-mentioned behavioral data to form a high-quality feature data after multi-source data fusion processing, maintain the freshness of the data, and thus ensure the effectiveness of broadband user relocation prediction.

[0164] 3. In order to take into account both timeliness and accuracy requirements, the embodiment of the present application draws on the idea of ​​semi-supervised learning, analyzes the behavior of users before applying for broadband mobile phone services, expands the number of samples based on the behavioral characteristics of the users' places of residence, and constructs the above-mentioned user population.

[0165] 4. In order to reduce the computational complexity, the center point of the user population is used to replace all broadband users in the user population, the user farthest from the center point in the user population is determined, and the distance between the user and the center point is defined as the boundary for judging whether the customer has a tendency to move. Thus, the tendency to move is judged by calculating the distance between the center point of the user population and the broadband customer to be predicted, thereby quantifying the similarity between the user population and the broadband user to be predicted, and finally outputting a list of broadband users with a high probability of moving.

[0166] The above describes the broadband user relocation prediction method provided by the embodiment of the present application. The following will describe the broadband user relocation prediction device provided by the embodiment of the present application in conjunction with the accompanying drawings.

[0167] See also Figure 3The embodiment of the present application further provides a broadband user migration prediction device, the device comprising:

[0168] A first acquisition module 301 is used to acquire characteristic data of a first broadband user in a first time period, wherein the characteristic data includes basic attribute data of the broadband user and behavior data related to broadband migration;

[0169] A first determination module 302 is configured to determine a target distance between the first broadband user and the user population according to the characteristic data of the first broadband user and the characteristic data of second broadband users included in a pre-established user population, wherein the user population includes a plurality of second broadband users with similar characteristic data, at least some of the second broadband users have a broadband migration requirement, and each second broadband user has characteristic data within a second time period;

[0170] The prediction module 303 is used to predict whether the first broadband user has a broadband migration demand according to the target distance.

[0171] Optionally, the basic attribute data includes at least one of the following:

[0172] Broadband account, target communication number bound to the broadband account, age of broadband user, broadband size, broadband online time, broadband package balance, and broadband migration time.

[0173] Optionally, the behavior data includes at least one of the following:

[0174] The first indication information is used to indicate whether the target communication number bound to the broadband account has dialed the communication number of the broadband manual customer service;

[0175] The second indication information is used to indicate whether the broadband user has visited the mobile interface of the broadband application APP;

[0176] The third indication information is used to indicate whether there is a call record data between the target communication number and a specific communication number at the residence of the broadband user, wherein the specific communication number includes at least one of a moving communication number, a renting communication number, and a purchasing communication number;

[0177] The fourth indication information is used to indicate whether the broadband user has visited a specific website address at the place where the broadband user lives, wherein the specific network address includes at least one of a moving website address, a renting website address, and a buying website address;

[0178] The fifth indication information is used to indicate whether the broadband user has visited a page related to the residence of the broadband user in a specific APP, wherein the specific APP includes at least one of a moving APP, a rental APP, and a house purchase APP;

[0179] The sixth indication information is used to indicate whether the residential area of ​​the broadband user is in a demolished community or a newly built community where the broadband user lives;

[0180] The seventh indication information is used to indicate whether the broadband installation address of the broadband user is located in a demolished community where the broadband user lives.

[0181] Optionally, the device further comprises: a second determination module, configured to determine a residential area of ​​the broadband user;

[0182] The second determination module determines the residential area of ​​the broadband user, including:

[0183] Acquiring base station access information of the target communication number within a third time period;

[0184] Determining, according to the base station access information, a target base station whose access frequency exceeds a second threshold;

[0185] The residential area of ​​the broadband user is determined according to the coverage area of ​​the target base station.

[0186] Optionally, the device further comprises a user population building module, which is used to:

[0187] Selecting a plurality of broadband users that meet a first condition to construct a first seed user group, wherein the first condition includes that the broadband migration time is not empty;

[0188] Select multiple broadband users that meet the second condition to construct a second sub-user group, wherein the second condition includes that the broadband migration time is empty, and at least one indication information from the first indication information to the seventh indication information is a positive indication;

[0189] The first seed user group and the second seed user group are combined to obtain the user population.

[0190] Optionally, the first determining module 302 is specifically configured to:

[0191] Determining characteristic data of a center point of the user population according to characteristic data of each second broadband user in the user population;

[0192] The distance between the characteristic data of the first broadband user and the characteristic data of the center point is determined, and the obtained distance is determined as the target distance.

[0193] Optionally, the prediction module 303 is specifically used for:

[0194] Determining characteristic data of a center point of the user population according to characteristic data of each second broadband user in the user population;

[0195] Determine, according to the characteristic data of each second broadband user in the user population, a third broadband user farthest from the center point among each second broadband user, and determine the distance between the characteristic data of the third broadband user and the characteristic data of the center point as a first threshold;

[0196] When the target distance is less than or equal to the first threshold, determining that the first broadband user has a broadband relocation demand;

[0197] When the target distance is greater than the first threshold, it is determined that the first broadband user has no broadband migration demand.

[0198] It should be noted that the division of units in the embodiments of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional unit in each embodiment of the present application may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0199] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0200] It should be noted here that the above-mentioned device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those in the method embodiment will not be described in detail here.

[0201] The embodiment of the present application also provides an electronic device, such as Figure 4 As shown, the electronic device includes a memory 420, a transceiver 410, and a processor 400;

[0202] Memory 420, used for storing computer programs;

[0203] The transceiver 410 is used to receive and send data under the control of the processor 400;

[0204] The processor 400 is configured to read the computer program in the memory 420 and execute the broadband user migration prediction method described in the first aspect.

[0205] Among them, Figure 4 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically linking together various circuits of one or more processors represented by processor 400 and memory represented by memory 420. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The transceiver 410 may be a plurality of components, namely, a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium, which transmission medium may include a wireless channel, a wired channel, an optical cable, and the like. The processor 400 is responsible for managing the bus architecture and general processing, and the memory 420 may store data used by the processor 400 when performing operations.

[0206] The processor 400 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or a complex programmable logic device (CPLD). The processor 400 may also adopt a multi-core architecture.

[0207] It should be noted here that the above-mentioned device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those in the method embodiment will not be described in detail here.

[0208] An embodiment of the present application further provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the broadband user migration prediction method described in the first aspect is implemented.

[0209] The computer-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (Magnet-Optica, MO)), etc.), optical storage (such as Compact Disk (CD), Digital Versatile Disc (DVD), Blu-ray Disc (BD), Holographic Versatile Disc (HVD), etc.), and semiconductor memory (such as Read-Only Memory (ROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read Only Memory (EEPROM), non-volatile memory (NAND FLASH), solid-state drive (Solid State Disk or Solid State Drive, SSD)), etc.

[0210] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) that contain computer-usable program code.

[0211] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer executable instructions. These computer executable instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0212] These processor executable instructions may also be stored in a processor readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the processor readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0213] These processor-executable instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0214] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A broadband user migration prediction method, characterized in that: The method comprises: Acquiring characteristic data of a first broadband user in a first time period, wherein the characteristic data includes basic attribute data of the broadband user and behavior data related to broadband migration; Determining a target distance between the first broadband user and the user population according to the characteristic data of the first broadband user and the characteristic data of the second broadband users included in the pre-established user population, wherein the user population includes a plurality of second broadband users with similar characteristic data, at least some of the second broadband users have a broadband migration demand, and each second broadband user has characteristic data within a second time period; It is predicted whether the first broadband user has a broadband relocation demand according to the target distance.

2. The method according to claim 1, characterized in that The basic attribute data includes at least one of the following: Broadband account, target communication number bound to the broadband account, age of broadband user, broadband size, broadband online time, broadband package balance, and broadband migration time.

3. The method according to claim 2, characterized in that The behavior data includes at least one of the following: The first indication information is used to indicate whether the target communication number bound to the broadband account has dialed the communication number of the broadband manual customer service; The second indication information is used to indicate whether the broadband user has visited the mobile interface of the broadband application APP; The third indication information is used to indicate whether there is a call record data between the target communication number and a specific communication number at the residence of the broadband user, wherein the specific communication number includes at least one of a moving communication number, a renting communication number, and a purchasing communication number; The fourth indication information is used to indicate whether the broadband user has visited a specific website address at the place where the broadband user lives, wherein the specific network address includes at least one of a moving website address, a renting website address, and a buying website address; The fifth indication information is used to indicate whether the broadband user has visited a page related to the residence of the broadband user in a specific APP, wherein the specific APP includes at least one of a moving APP, a rental APP, and a house purchase APP; The sixth indication information is used to indicate whether the residential area of ​​the broadband user is in a demolished community or a newly built community where the broadband user lives; The seventh indication information is used to indicate whether the broadband installation address of the broadband user is located in a demolished community where the broadband user lives.

4. The method according to claim 3, characterized in that The process of determining the residential area of ​​the broadband user includes: Acquiring base station access information of the target communication number within a third time period; Determining, according to the base station access information, a target base station whose access frequency exceeds a second threshold; The residential area of ​​the broadband user is determined according to the coverage area of ​​the target base station.

5. The method according to claim 3, characterized in that: The process of constructing the user population includes: Selecting a plurality of broadband users that meet a first condition to construct a first seed user group, wherein the first condition includes that the broadband migration time is not empty; Select multiple broadband users that meet the second condition to construct a second sub-user group, wherein the second condition includes that the broadband migration time is empty, and at least one indication information from the first indication information to the seventh indication information is a positive indication; The first seed user group and the second seed user group are combined to obtain the user population.

6. The method according to any one of claims 1 to 5, characterized in that: The determining, based on the characteristic data of the first broadband user and the characteristic data of the second broadband user included in the pre-established user population, a target distance between the first broadband user and the user population includes: Determining characteristic data of a center point of the user population according to characteristic data of each second broadband user in the user population; The distance between the characteristic data of the first broadband user and the characteristic data of the center point is determined, and the obtained distance is determined as the target distance.

7. The method according to any one of claims 1 to 5, characterized in that: The predicting, according to the target distance, whether the first broadband user has a broadband relocation demand includes: Determining characteristic data of a center point of the user population according to characteristic data of each second broadband user in the user population; Determine, according to the characteristic data of each second broadband user in the user population, a third broadband user farthest from the center point among each second broadband user, and determine the distance between the characteristic data of the third broadband user and the characteristic data of the center point as a first threshold; When the target distance is less than or equal to the first threshold, determining that the first broadband user has a broadband relocation demand; When the target distance is greater than the first threshold, it is determined that the first broadband user has no broadband migration demand.

8. A broadband user migration prediction device, characterized in that: The device comprises: A first acquisition module, configured to acquire characteristic data of a first broadband user in a first time period, wherein the characteristic data includes basic attribute data of the broadband user and behavior data related to broadband migration; a first determination module, configured to determine a target distance between the first broadband user and the user population according to feature data of the first broadband user and feature data of second broadband users included in a pre-established user population, wherein the user population includes a plurality of second broadband users with similar feature data, at least some of the second broadband users have a need to migrate broadband devices, and each second broadband user has feature data within a second time period; The prediction module is used to predict whether the first broadband user has a broadband migration demand according to the target distance.

9. An electronic device, characterized in that: Including memory, transceiver, processor: Memory for storing computer programs; A transceiver, used for sending and receiving data under the control of the processor; a processor, used for reading the computer program in the memory and executing the broadband user migration prediction method according to any one of claims 1 to 7.

10. A readable storage medium, characterized in that: The readable storage medium stores a program or an instruction, and when the program or the instruction is executed by the processor, the method for predicting the migration of broadband users according to any one of claims 1 to 7 is implemented.

Images