CN115134805B - Method, device, equipment and storage medium for predicting potentially carried heterogeneous network number - Google Patents
Method, device, equipment and storage medium for predicting potentially carried heterogeneous network number Download PDFInfo
- Publication number
- CN115134805B CN115134805B CN202110336829.9A CN202110336829A CN115134805B CN 115134805 B CN115134805 B CN 115134805B CN 202110336829 A CN202110336829 A CN 202110336829A CN 115134805 B CN115134805 B CN 115134805B
- Authority
- CN
- China
- Prior art keywords
- target
- target number
- carried
- network
- potential
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 238000004891 communication Methods 0.000 claims abstract description 64
- 238000012549 training Methods 0.000 claims description 39
- 238000012360 testing method Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 13
- 238000012795 verification Methods 0.000 claims description 13
- 238000010200 validation analysis Methods 0.000 claims description 9
- 238000013135 deep learning Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W8/00—Network data management
- H04W8/26—Network addressing or numbering for mobility support
- H04W8/28—Number portability ; Network address portability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/06—Testing, supervising or monitoring using simulated traffic
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application discloses a method, a device, equipment and a storage medium for predicting a potentially carried-in different-network number, which relate to the field of communication and are used for solving the problem that the potentially carried-in different-network number cannot be predicted. The method for predicting the potentially carried heterogeneous network number comprises the following steps: obtaining a target number, wherein the target number is a different network number which is not carried into the home network and normally communicates with the home network number; obtaining a pre-trained target judgment model for predicting a potential carried-in foreign network number; and carrying out potential carry-in prediction on the target number by using the target judgment model. The method and the device are used for predicting the potential carried-in different network numbers.
Description
Technical Field
The present application relates to the field of communications, and in particular, to a method, an apparatus, a device, and a storage medium for predicting a potentially carried foreign network number.
Background
The number-carrying and network-transferring project is a welcome project with high social attention, and is also an important measure for better benefiting people in the development thought of people as a center practiced by the information communication industry. But keep the stability of a certain customer scale, which is a precondition for improving customer service perception and making the customer population form virtuous circle. In order to achieve the balance of carrying in and carrying out, besides the need of reducing carrying out, carrying in is increased, so that more convenient service is provided for carrying in clients.
Currently, there are only methods for predicting the home network number that is willing to be carried out from the home network. Because the telecom operator lacks knowledge of the users of the different networks, it is not possible to predict the different network numbers that will be carried into the home network from the different networks.
Disclosure of Invention
The embodiment of the application provides a method, a device, equipment and a storage medium for predicting a potentially carried-in different network number, which can solve the problem that the potentially carried-in different network number cannot be predicted.
In a first aspect, a method for predicting a potentially ported foreign network number is provided, the method comprising:
Obtaining a target number, wherein the target number is a different network number which is not carried into the home network and normally communicates with the home network number;
Obtaining a pre-trained target judgment model for predicting a potential carried-in foreign network number;
and carrying out potential carry-in prediction on the target number by using the target judgment model.
In a second aspect, there is provided an apparatus for predicting a potentially ported foreign network number, the apparatus comprising:
The acquisition module is used for acquiring a target number, wherein the target number is a different network number which is not carried into the local network and normally communicates with the local network number;
The acquisition module is also used for acquiring a pre-trained target judgment model for predicting the potential carried-in different network numbers;
and the prediction module is used for predicting the potential carry-in of the target number by utilizing the target judgment model.
In a third aspect, there is provided an apparatus for predicting a potentially ported foreign network number, the apparatus comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor performs the steps of the method as described in the first aspect above.
In a fourth aspect, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method as described in the first aspect above.
In the embodiment of the application, the target number, i.e. the different network number which is not carried into the home network and normally communicates with the home network number, can be acquired first, and then the potential carrying prediction is carried out on the target number through a pre-trained target judgment model, i.e. whether the target number is willing to carry into the home network or not is predicted. Thus, the potential carried-in foreign network number can be predicted.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
fig. 1 is a flowchart of a method for predicting a potentially ported foreign network number according to an embodiment of the present application.
Fig. 2 is a flowchart of a method for training a target decision model according to an embodiment of the present application.
Fig. 3 is a flowchart of another method for predicting a potentially ported foreign network number according to an embodiment of the present application.
Fig. 4 is a block diagram of an apparatus for predicting a potentially ported foreign network number according to an embodiment of the present application.
Detailed Description
For the purpose of promoting an understanding of the principles of the application, reference will now be made in detail to specific embodiments of the application and the accompanying drawings, in which it is apparent that some, but not all embodiments of the application will be illustrated. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Aiming at solving the problem that the potentially carried-in different network numbers cannot be predicted, the application provides a solution and aims to provide a method for predicting the potentially carried-in different network numbers.
The following describes in detail the technical solutions provided by the embodiments of the present application with reference to the accompanying drawings.
In order to maintain the data security of the clients and the home network, the databases of the large operators are generally independent and encrypted, and the home network operators cannot generally know the data information of the users of the different networks. Therefore, the home network operator lacks knowledge of the users of the different networks, and cannot estimate the willingness of the users of the different networks to transfer to (hereinafter simply referred to as being carried into) the home network.
In order to solve the technical problems, the embodiment of the application provides a method for predicting a potentially ported foreign network number, which comprises the steps of firstly obtaining a target number, namely a foreign network number which is not ported to a home network and normally communicates with the home network number, and then predicting the target number to be potentially ported through a pre-trained target judgment model, namely predicting whether the target number is willing to be ported to the home network. The method is used for predicting the potentially carried heterogeneous network numbers. The execution body of the method may be a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, or the like.
Fig. 1 is a flowchart of a method for predicting a potentially ported foreign network number according to an embodiment of the present application. As shown in fig. 1, the method for predicting a potentially carried foreign network number according to the embodiment of the present application may include:
step 110, a target number is obtained, wherein the target number is a different network number which is not carried into the home network and normally communicates with the home network number.
It can be understood that the target number is a number that needs to be subjected to potential carry-in prediction, and the target number is a different network number that is not carried into the home network and normally communicates with the home network number. Because the home network operator cannot acquire most of information of the different network number, only the call information of the different network user and the home network user exists in the home network database, and therefore, the potential carry-in prediction can only be carried out on the different network number which is not carried into the home network and normally communicates with the home network number. The target number may be obtained from a traffic database, where the traffic database includes various data information of the home network number and partial information of a foreign network number that communicates with the home network number, such as home network number information that communicates with the home network number, service information that is handled by the home network number, and foreign network number information that communicates with the home network number (i.e., a foreign network number that normally communicates with the home network number and is not carried into the home network).
Step 120, a pre-trained target decision model for predicting potential ported-in foreign network numbers is obtained.
And 130, utilizing the target judgment model to carry out potential carry-in prediction on the target number.
The method for predicting the potential carried-in heterogeneous network number provided by the embodiment of the application can acquire the target number from the heterogeneous network number which is not carried in the local network and normally communicates with the local network number, and then carries out the potential carried-in prediction on the target number through the pre-trained target judgment model. Thus, whether the target number is willing to be carried into the home network or not can be predicted.
As described above, before the target number is potentially ported, a target determination model for predicting a potentially ported foreign network number needs to be trained in advance. Here, a training process of the target determination model will be described:
optionally, the training process of the target determination model may include:
obtaining a positive sample and a negative sample, wherein the positive sample comprises a different network number which is carried into the home network, and the negative sample comprises a different network number which is not carried into the home network and normally communicates with the home network number;
determining a target training set based on the positive and negative samples;
training a prediction model based on deep learning based on the target training set until the accuracy of the prediction model reaches a preset value to obtain the target judgment model.
The prediction model is a target judgment model in the training process, and the target judgment model is the prediction model after training. It can be understood that the positive sample may be obtained from the carried-in database, where the carried-in database is a database containing various information data of the foreign network number that has been carried in the home network and normally communicates with the home network number, such as a business situation and a communication situation handled after the foreign network number that has been carried in the home network and normally communicates with the home network number is transferred into the home network, and an internet passing communication situation between the foreign network number that has been carried in the home network and normally communicates with the home network number before the foreign network number that has been carried in the home network and normally communicates with the home network number is transferred into the home network. The network communication condition represents the communication condition between the different network number and the local network number. Meanwhile, the negative sample can also be obtained from the telephone traffic database.
The positive sample includes a foreign network number that has been carried into the home network, the negative sample includes a foreign network number that has not been carried into the home network and is in normal communication with the home network number, which may be understood as taking the foreign network number that has been carried into the home network as a carried sample, and taking the foreign network number that has not been carried into the home network and is in normal communication with the home network number as a non-carried sample, where the carried sample and the non-carried sample have references. Therefore, whether the target number is willing to be carried into the home network or not can be judged according to the carried-in sample and the non-carried-in sample. And then training according to the positive sample and the negative sample to obtain the target judgment model, and carrying out potential carry-in prediction on the target number through the target judgment model, namely judging whether the target number is willing to carry in the home network or not, and simultaneously, adopting the target judgment model to carry out potential carry-in prediction on the target number can reduce the prediction cost and save the prediction time.
Optionally, the method for predicting the potentially carried foreign network number provided by the embodiment of the present application may further include:
acquiring potential carry-in characteristics of the target number;
The utilizing the target determination model to make the potential carry-in prediction for the target number comprises:
and inputting the target number and the acquired potential carry-in characteristics of the target number into the target judgment model to obtain the probability of carrying the target number into the local network.
The potential carry-in feature of the target number is feature information which can indicate that the target number is willing to carry in the local network, and the potential carry-in feature of the target number can be obtained from the telephone traffic database. The target number needs to be subjected to potential carry-in prediction according to the target judgment model, namely the target number needs to be analyzed according to the target judgment model. For ease of analysis, the potential carry-in characteristics of the target number and the target number may be input into the target decision model for predictive analysis. Meanwhile, when the potential carry-in prediction is performed on the target number, the probability that the target number is carried in the local network can be represented by the probability that the target number is carried in the local network, and the target judgment model can calculate the probability that the target number is carried in the local network through a deep learning algorithm.
Therefore, the probability of the target number being carried into the home network can be obtained directly through the target judgment model, so that the calculated amount is greatly reduced, and the method for predicting the potential carried-in different network numbers is more convenient.
Because the information of the target number (the different network number which is not carried into the home network and normally communicates with the home network number) which can be obtained by the home network operator is limited, the probability of the target number being carried into the home network is not accurate enough only according to the information of the target number. The home network operator can obtain a large amount of information of the home network number, so that in order to make the obtained probability that the target number is carried into the home network more accurate, the information of the home network number closely contacted with the target number can be considered when determining the probability that the target number is carried into the home network.
It can be understood that if a user of a different network number is closely associated with a user of a home network number, the dependence or preference of the user of the home network number on the home network may affect the preference of the user of the different network number on the home network, thereby affecting the willingness of the user of the different network number to carry into the home network.
Thus, in order to improve the accuracy of the probability prediction of the target number being ported to the home network, optionally, the acquiring the potential ported feature of the target number may include:
acquiring a local network number with the communication frequency greater than a preset threshold value as a target number association number;
And taking the internet passing communication condition of the target number and the business handling condition of the target number associated number as potential carry-in characteristics of the target number.
When the potential carry-in characteristics of the target number are obtained, the communication condition of the target number and the own network number passing through the network and some characteristics of the own network number (the target association number) closely related with the target number are also considered, so that the potential carry-in characteristics are enriched, and the accuracy of probability prediction of carrying the target number into the own network is improved.
If a user of a different network number is closely related to a user of a home network number, the dependence or preference of the user of the home network number on the home network may affect the preference of the user of the different network number on the home network, thereby affecting the willingness of the user of the different network number to carry into the home network. Thus, the potential carry-in feature of the target number may also include a feature that may represent the home network's dependency or preference of users of home network numbers that are closely related to the foreign network number. The compactness of the connection between the different network number and the local network number can also be used as the basis for judging the willingness of the user of the different network number to carry in the local network.
Optionally, the network-passing communication condition of the target number may include: at least one of communication duration, communication times and communication frequency of the target number and the target number associated number;
The business handling situation of the target number associated number comprises the following steps: and the target number is associated with at least one of the types, the quantity and the duration of the number transacted business.
Optionally, the network communication condition of the target number may further include whether the target number actively contacts the home network customer service, or may also include the number of times of communication between the target number and the home network customer service.
Thus, the contact tightness between the target number and the target number associated number can be judged according to the communication duration, the communication times and the communication frequency of the target number and the target number associated number; and judging the dependence or preference of the target number associated number on the local network according to the type, the number and the duration of the service handled by the target number associated number.
Optionally, the services handled by the target number associated number include a sticky service, where the sticky service is a service that can represent the dependency of the user on the home network.
Alternatively, the sticky service may include home network, broadband, etc. services.
Thus, the favorites and the dependences of the target number associated number on the local network can be analyzed according to the conditions of the sticky service transacted by the target number associated number.
As described above, the probability of the target number being carried into the home network can be more conveniently obtained by using the target determination model, which is trained based on the target training set determined from the positive and negative samples.
Optionally, the target training set includes at least a portion of the positive samples and at least a portion of the negative samples, and the method provided by the embodiment of the present application may further include:
Determining a target validation set based on the positive sample and the negative sample; wherein the target validation set includes at least a portion of the positive samples and at least a portion of the negative samples;
The method may further comprise:
and optimizing parameters of the prediction model according to the target verification set.
In this way, the prediction model trained by the target training set can be further perfected through the optimization of the parameters of the prediction model by the target verification set.
Optionally, the method provided by the embodiment of the application may further include: determining a target test set based on the positive and negative samples; wherein the target test set includes at least a portion of the positive samples and at least a portion of the negative samples;
The method may further comprise: and checking the accuracy of the prediction model according to the target test set.
Optionally, the method provided by the embodiment of the application may further include: and forming a sample set according to the positive sample and the negative sample, wherein the sample set can comprise the target training set, the target verification set and the target test set.
Meanwhile, after the optimization of the parameters of the prediction model according to the verification set, the following steps may be performed: checking the accuracy of the prediction model according to the target test set;
and if the accuracy of the prediction model reaches a preset value, obtaining the target training model.
The target training set, target validation set, and target test set each comprise a plurality of samples in the sample set.
In deep learning, the sample is typically divided into three separate parts: training set, validation set and test set. The training set is used for building a model, the verification set is used for determining parameters of network structure or control model complexity, and the test set is used for checking whether the optimal model is finally selected.
Fig. 2 is a flowchart of a method for training a target decision model according to an embodiment of the present application.
Optionally, as shown in fig. 2, training the target decision model includes the steps of:
Step 210, dividing the sample set into a target training set, a target verification set and a target test set;
Step 220, training a prediction model by using the target training set;
step 230, optimizing parameters of the prediction model by using the verification set;
step 240, checking the accuracy of the prediction model by using the test set;
and step 250, if the accuracy of the prediction model reaches a preset value, obtaining the target training model.
Optionally, the dividing the sample set into the target training set, the target verification set and the target test set may include: the target training set comprises 50% of the sample set, and the target validation set and the target test set each comprise 25% of the sample set.
Optionally, the method provided by the embodiment of the application may further include:
obtaining a potential carry-in characteristic index of the target number according to the potential carry-in characteristic of the target number;
Inputting the target number and the acquired potential carry-in characteristics of the target number into the target judgment model;
And the target judgment model calculates the probability of the target number being carried into the local network according to the weight of the potential carried-in characteristic of the target number and the potential carried-in characteristic index.
The potential carried-in feature of the target number may be a parameter change value of the potential carried-in feature or directly a parameter value of the potential carried-in feature.
How to obtain the index of the potential carry-in feature of the target number according to the potential carry-in feature of the target number is exemplified herein:
If the number of the communication times between the target number and the target number associated number is determined, determining the traffic change index of the target number; determining a customer service contact index of the target number according to the communication times of the target user and the local network customer service; and determining the business change index of the target number associated number according to the number of businesses handled by the target number associated number. The traffic change index can be a traffic increase and decrease value, the customer service contact index can be the communication times between a target user and the local network customer service, and the service change index can be a service number increase and decrease value of the target user associated user.
The potential portability feature index may represent a change in favorability of the association number of the target number to the home network and a change in compactness of contact between the target number and the association number of the target number, that is, indirectly may represent a change in willingness of the target number to carry into the home network.
Optionally, the determining the target training set based on the positive and negative samples may include: determining the target training set based on the carried-in characteristics of the positive and negative samples;
The positive sample carry-in feature includes: the network communication condition of the positive sample and the business handling condition of the positive sample associated number;
potential carry-in characteristics of the negative sample include: the network communication condition of the negative sample and the business handling condition of the negative sample associated number;
Wherein, the network communication condition of the positive sample includes: at least one of communication duration, communication times and communication frequency of the positive sample and the positive sample associated number;
The business handling situation of the positive sample associated number comprises the following steps: at least one of the types, the numbers and the duration of the services transacted by the positive sample associated number;
The negative-sample internet communication condition comprises: at least one of communication duration, communication times and communication frequency of the negative sample and the negative sample associated number;
the business handling situation of the negative sample associated number comprises the following steps: at least one of the category, the number and the duration of the business handled by the negative sample associated number.
The positive sample association number is a local network number with the communication frequency of the positive sample being greater than a preset threshold value, and the communication frequency of the negative sample association number and the negative sample is greater than the local network number with the preset threshold value.
When the target judgment model is trained, the target judgment model can determine the weight of each carried-in feature according to the distinguishing degree of each carried-in feature of the positive sample and each carried-in feature of the negative sample.
The distinction degree between each carried-in feature of the positive sample and each carried-in feature of the negative sample may be understood as the importance degree of each carried-in feature when distinguishing the positive sample from the negative sample, for example, the positive sample includes 10 different network numbers that have been carried into the home network, the negative sample includes 10 different network numbers that have not been carried into the home network, the number of communications between 9 different network numbers and their associated numbers (the number of the home network with the different network numbers having a communication frequency greater than a preset threshold) in the positive sample is greater than 100, and the number of communications between only 1 different network number and its associated number (the number of the home network with the different network number having a communication frequency greater than a preset threshold) in the negative sample is greater than 100. Therefore, when judging whether the different network number which is not carried into the home network is carried into the home network, the communication times of the different network number and the associated number can be considered.
For example, the positive sample includes 10 different network numbers that have been carried into the home network, the negative sample includes 10 different network numbers that have not been carried into the home network, the number of services handled by the associated number of 5 different network numbers in the positive sample is greater than 4, and the number of services handled by the associated number of 4 different network numbers in the negative sample is greater than 4. At this time, when the different network number which is not carried into the home network is carried into the home network, the service quantity handled by the associated number of the different network number does not need to be considered.
It should be noted that, each potential carrying-in feature of the target number corresponds to each carrying-in feature of the positive sample and each carrying-in feature of the negative sample, and the weight of each potential carrying-in feature of the target number is equal to the weight of each carrying-in feature of the positive sample and the weight of each carrying-in feature of the negative sample.
Specifically, the process of calculating the probability of the target number being ported into the home network by the target determination model according to the weight of the potential ported-in feature of the target number and the potential ported-in feature index is as follows:
For example, let the probability that the target number is carried into the home network be P, where the carried-in characteristics of the positive sample and the negative sample include a communication duration and a communication number of times with the associated number, and the number of services handled by the associated number; the potential carry-in characteristics of the target number comprise the communication duration and the communication times of the negative sample and the negative sample associated number, and the number of the business transacted by the associated number; the communication duration characteristic index in the potential carry-in characteristic index is x, the communication frequency characteristic index is y, and the associated number business quantity characteristic index is z; obtaining a communication duration weight of the carried-in feature as a, a communication frequency weight as b and an associated number business quantity weight as c according to the target training model; the probability of the target number being carried into the home network is p=ax+by+cz, and the a, b, c, x, y, z and P are constants.
Alternatively, the target decision model may be a random forest model.
Fig. 3 is a flowchart of another method for predicting a potentially ported foreign network number according to an embodiment of the present application. As shown in fig. 3, the method for predicting a potentially carried foreign network number according to the embodiment of the present application may include the following steps:
Step 310, the different network number which is already carried into the local network is obtained from the carried-in database as a positive sample, and the different network number which is not already carried into the local network and normally communicates with the local network number is obtained from the telephone traffic database as a negative sample.
A target training set is determined based on the positive and negative sample carried-in characteristics, step 320.
And step 330, training according to the target training set to obtain a target judgment model and weights of all the carried-in characteristics.
Step 340, obtaining a target number and a potential carry-in feature of the target number.
Step 350, obtaining the index of the potential carry-in feature of the target number according to the potential carry-in feature of the target number.
Step 360, inputting the target number and the potential carry-in feature of the target number into the target determination model.
In step 370, the target decision model calculates the probability of the target number being ported into the home network according to the weights of the potential ported features of the target number and the indexes of the potential ported features.
According to the method for predicting the potential carried-in heterogeneous network number, provided by the embodiment of the application, the heterogeneous network number which is carried in the local network can be obtained from the carried-in database to serve as a positive sample, the heterogeneous network number which is not carried in the local network and normally communicates with the local network number can be obtained from the telephone traffic database to serve as a negative sample, a target judgment model is obtained according to the training of the positive sample and the negative sample, and then the probability that the heterogeneous network number which is not carried in the local network and normally communicates with the local network number is predicted by utilizing the target judgment model.
Fig. 4 is a block diagram of an apparatus for predicting a potentially ported foreign network number according to an embodiment of the present application. As shown in fig. 4, an embodiment of the present application provides an apparatus 400 for predicting a potentially carried foreign network number, which includes an acquisition module 401 and a prediction module 402.
The obtaining module 401 is configured to obtain a target number, where the target number is a different network number that is not carried into the home network and normally communicates with the home network number.
The obtaining module 401 is further configured to obtain a pre-trained target decision model for predicting a potential carried-in foreign network number.
The prediction module 402 is configured to perform a potential carry-in prediction on the target number by using the target determination model.
Optionally, the obtaining module 401 is further configured to obtain a positive sample and a negative sample, where the positive sample includes a foreign network number that has been carried into the home network, and the negative sample includes a foreign network number that has not been carried into the home network and is in normal communication with the home network number.
Optionally, an apparatus for predicting a potentially carried foreign network number according to an embodiment of the present application may further include a determining module 403. The determining module 403 may be configured to determine a target training set based on the positive sample and the negative sample, and train a prediction model based on deep learning based on the target training set until an accuracy of the prediction model reaches a preset value, to obtain the target determination model.
Optionally, the obtaining module 401 is further configured to obtain a potential carry-in feature of the target number.
Optionally, when the target decision model is used to predict the potential carry-in of the target number, the prediction module 402 is specifically configured to input the target number and the obtained potential carry-in feature of the target number into the target decision model, so as to obtain a probability that the target number carries in the home network.
Optionally, the obtaining module 401 is further configured to obtain, as the target number association number, a home network number with a communication frequency with the target number greater than a preset threshold. The determining module 403 is further configured to use the internet communication situation of the target number and the business handling situation of the target number associated number as the potential carry-in feature of the target number.
Optionally, the internet communication condition of the target number includes: at least one of communication duration, communication times and communication frequency of the target number and the target number associated number; the business handling situation of the target number associated number comprises the following steps: and the target number is associated with at least one of the types, the quantity and the duration of the number transacted business.
Optionally, the target training set comprises at least a portion of the positive samples and at least a portion of the negative samples. The determining module 403 is further configured to determine a target verification set based on the positive sample and the negative sample and optimize parameters of the prediction model according to the target verification set; wherein the target validation set includes at least a portion of the positive samples and at least a portion of the negative samples.
Optionally, the determining module 403 is further configured to determine a target verification set based on the positive sample and the negative sample and optimize parameters of the prediction model according to the target verification set; wherein the target validation set includes at least a portion of the positive samples and at least a portion of the negative samples.
It should be appreciated that the method for predicting a potentially ported heterogeneous network number described above may be applied to the apparatus for predicting a potentially ported heterogeneous network number provided in the embodiments of the present application, so the content of the apparatus for predicting a potentially ported heterogeneous network number may be referred to in the description of the method section above.
The embodiment of the application also provides a device for predicting the potential carried-in heterogeneous network number, which comprises: a memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor implements any of the methods as described above.
Embodiments of the present application also provide a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements any of the methods as described above.
From the description of the embodiments above, those skilled in the art will appreciate that embodiments of the invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The embodiments of the present application have been described above with reference to the accompanying drawings, but the present application is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present application and the scope of the claims, which are to be protected by the present application.
Claims (8)
1. A method of predicting a potentially ported foreign network number, the method comprising:
Obtaining a target number, wherein the target number is a different network number which is not carried into the home network and normally communicates with the home network number;
acquiring a local network number with the communication frequency greater than a preset threshold value as a target number association number;
taking the internet communication condition of the target number and the business handling condition of the target number associated number as potential carry-in characteristics of the target number;
Obtaining a pre-trained target judgment model for predicting a potential carried-in foreign network number;
Inputting the target number and the acquired potential carry-in characteristics of the target number into the target judgment model to obtain the probability of carrying the target number into the local network;
the target judgment model calculates the probability of the target number being carried into the local network according to the weight of the potential carried-in characteristic of the target number and the potential carried-in characteristic index;
the potential carry-in characteristic index of the target number is obtained according to the potential carry-in characteristic of the target number; the potential carry-in feature of the target number is a parameter change value of the potential carry-in feature or is directly a parameter value of the potential carry-in feature, and the potential carry-in feature index represents a change of favorites of the target number associated number on the local network and a change of contact compactness of the target number and the target number associated number.
2. The method of claim 1, wherein the training process of the target decision model comprises:
obtaining a positive sample and a negative sample, wherein the positive sample comprises a different network number which is carried into the home network, and the negative sample comprises a different network number which is not carried into the home network and normally communicates with the home network number;
determining a target training set based on the positive and negative samples;
training a prediction model based on deep learning based on the target training set until the accuracy of the prediction model reaches a preset value to obtain the target judgment model.
3. The method of claim 1, wherein the network-passing communication condition of the destination number comprises: at least one of communication duration, communication times and communication frequency of the target number and the target number associated number;
The business handling situation of the target number associated number comprises the following steps: and the target number is associated with at least one of the types, the quantity and the duration of the number transacted business.
4. The method according to claim 2, wherein the method further comprises:
Determining a target validation set based on the positive sample and the negative sample; wherein the target validation set includes at least a portion of the positive samples and at least a portion of the negative samples;
The method further comprises the steps of:
and optimizing parameters of the prediction model according to the target verification set.
5. The method according to claim 2, wherein the method further comprises:
Determining a target test set based on the positive and negative samples; wherein the target test set includes at least a portion of the positive samples and at least a portion of the negative samples;
The method further comprises the steps of: and checking the accuracy of the prediction model according to the target test set.
6. An apparatus for predicting a potentially ported foreign network number, the apparatus comprising:
The acquisition module is used for acquiring a target number, wherein the target number is a different network number which is not carried into the local network and normally communicates with the local network number;
the acquisition module is further used for acquiring a local network number with the communication frequency of the target number being greater than a preset threshold value as a target number association number; taking the internet communication condition of the target number and the business handling condition of the target number associated number as potential carry-in characteristics of the target number;
The acquisition module is also used for acquiring a pre-trained target judgment model for predicting the potential carried-in different network numbers;
The prediction module is used for inputting the target number and the acquired potential carry-in characteristics of the target number into the target judgment model to obtain the probability of carrying the target number into the local network; the target judgment model calculates the probability of the target number being carried into the local network according to the weight of the potential carried-in characteristic of the target number and the potential carried-in characteristic index; the potential carry-in characteristic index of the target number is obtained according to the potential carry-in characteristic of the target number; the potential carry-in feature of the target number is a parameter change value of the potential carry-in feature or is directly a parameter value of the potential carry-in feature, and the potential carry-in feature index represents a change of favorites of the target number associated number on the local network and a change of contact compactness of the target number and the target number associated number.
7. An apparatus for predicting a potentially ported foreign network number, the apparatus comprising: memory, a processor and a computer program stored on the memory and executable on the processor, which when executed by the processor, performs the steps of the method according to any one of claims 1 to 5.
8. A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110336829.9A CN115134805B (en) | 2021-03-29 | 2021-03-29 | Method, device, equipment and storage medium for predicting potentially carried heterogeneous network number |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110336829.9A CN115134805B (en) | 2021-03-29 | 2021-03-29 | Method, device, equipment and storage medium for predicting potentially carried heterogeneous network number |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115134805A CN115134805A (en) | 2022-09-30 |
CN115134805B true CN115134805B (en) | 2024-07-16 |
Family
ID=83375591
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110336829.9A Active CN115134805B (en) | 2021-03-29 | 2021-03-29 | Method, device, equipment and storage medium for predicting potentially carried heterogeneous network number |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115134805B (en) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111092762A (en) * | 2019-12-19 | 2020-05-01 | 深圳市博瑞得科技有限公司 | Prediction method, device and storage medium for number portability potential user |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10069862B2 (en) * | 2013-03-15 | 2018-09-04 | Symantec Corporation | Techniques for predicting and protecting spearphishing targets |
US20180247379A1 (en) * | 2017-02-24 | 2018-08-30 | Facebook, Inc. | Evaluating potential connections based on instrumental variables |
CN112200375B (en) * | 2020-10-15 | 2023-08-29 | 中国联合网络通信集团有限公司 | Prediction model generation method, prediction model generation device, and computer-readable medium |
CN112153636A (en) * | 2020-10-29 | 2020-12-29 | 浙江鸿程计算机系统有限公司 | Method for predicting number portability and roll-out of telecommunication industry user based on machine learning |
-
2021
- 2021-03-29 CN CN202110336829.9A patent/CN115134805B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111092762A (en) * | 2019-12-19 | 2020-05-01 | 深圳市博瑞得科技有限公司 | Prediction method, device and storage medium for number portability potential user |
Also Published As
Publication number | Publication date |
---|---|
CN115134805A (en) | 2022-09-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106682906B (en) | Risk identification and service processing method and equipment | |
CN109379240B (en) | Internet of vehicles flow prediction model construction method and device and electronic equipment | |
CN114610773A (en) | Data application scene recognition method based on big data and cloud computing server | |
CN115310762A (en) | Target service determination method and device based on heterogeneous graph neural network | |
CN106095688A (en) | A kind of software performance testing method and device | |
CN111294227A (en) | Method, apparatus, device and medium for neural network-based traffic prediction | |
CN114548118A (en) | Service conversation detection method and system | |
CN115134805B (en) | Method, device, equipment and storage medium for predicting potentially carried heterogeneous network number | |
CN111181957B (en) | Internet of things equipment security verification method and system and central control equipment | |
CN116032652B (en) | Gateway authentication method and system based on intelligent interactive touch panel | |
US20230121564A1 (en) | Bias detection and reduction in machine-learning techniques | |
CN110782276B (en) | Access shunting policy interference judgment method and device and electronic equipment | |
CN115134399B (en) | User identification method and device | |
US20180285077A1 (en) | Sample-based multidimensional data cloning | |
CN109857726B (en) | Application feature library maintenance method and device, electronic equipment and storage medium | |
CN111241376B (en) | Multistage information matching method and device and cloud service platform | |
CN114969073A (en) | Wind control method and device based on block chain | |
CN114722723B (en) | Emotion tendency prediction method and equipment based on kernel extreme learning machine optimization | |
CN118051626A (en) | Business recommendation method and device based on knowledge graph and electronic equipment | |
CN117648638A (en) | Method and device for determining monitoring strategy and electronic equipment | |
CN114240255A (en) | Incidence relation-based power overhaul scheme recommendation method and device | |
CN118214783A (en) | Data processing method and device, storage medium and electronic equipment | |
CN117132169A (en) | Electric power artificial intelligent model calling evaluation method and system | |
CN117950891A (en) | Business exception processing method and device, electronic equipment and storage medium | |
CN116401243A (en) | Engine implementation method, device, terminal and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant |