A broadband resource pre-judgment method and device, electronic equipment and storage medium

Abstract

The application discloses a broadband resource pre-judgment method and device, electronic equipment and a storage medium, relates to the technical field of communication, and is used for solving the problem that users cannot install a broadband because of incorrect judgment and omission of resource prediction when an operator installs a broadband. The method comprises the following steps: obtaining a target pre-installation address of a user who installs a broadband resource, and generating an address characteristic value of the target pre-installation address; obtaining an address classification result according to the address characteristic value and a pre-established resource pre-judgment method; determining a target address similarity between the target pre-installation address and an installed address; and adding the address classification result and the target address similarity after weighting, to obtain an installation pre-judgment result, which is used for determining whether the target pre-installation address can install a broadband.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a broadband resource prediction method, apparatus, electronic device and storage medium. Background Technology

[0002] Due to reasons related to property management and administration, there are situations in cities where home broadband cannot be installed. Therefore, operators need to conduct resource prediction before installing home broadband. However, the existing technical solutions simply determine whether home broadband resources can be installed based on the range of the access room's electronic fence. This may lead to misjudgment or omission of resource prediction, resulting in users being unable to install the service and affecting user experience. Summary of the Invention

[0003] This application provides a broadband resource prediction method, device, electronic device, and storage medium to solve the problem that users cannot install broadband due to misjudgment or omission of resource prediction during the installation of home broadband by operators.

[0004] To achieve the above objectives, this application adopts the following technical solution:

[0005] In a first aspect, this application provides a broadband resource prediction method, comprising: obtaining the target pre-installation address of a user installing broadband resources, and generating an address feature value of the target pre-installation address; obtaining an address classification result based on the address feature value and a pre-established resource prediction method; determining the target address similarity between the target pre-installation address and an already installed address; and weightedly adding the address classification result and the target address similarity to obtain an installation prediction result, wherein the installation prediction result is used to determine whether broadband can be installed at the target pre-installation address.

[0006] Optionally, obtain a historical data set of operator broadband installation orders; for at least one sample pre-installation address in the operator broadband installation order dataset, generate corresponding address feature values ​​for the sample pre-installation address, including whether the address is within an electronic fence, the address type, the address construction time, the type of the nearest access point, and the distance between the nearest access points; determine the address classification result of the address feature values ​​of the sample pre-installation address; the address classification result includes successful broadband installation and unsuccessful broadband installation; use the address feature values ​​of at least one sample pre-installation address and the corresponding address classification result to form a sample set; use the sample set to determine the resource prediction method.

[0007] Optionally, the above-mentioned determination of the target address similarity between the target pre-installed address and the installed address includes: performing word segmentation on the target pre-installed address to obtain a first address vector corresponding to the target pre-installed address; performing word segmentation on the installed address for a sample installed address in the installation address database to obtain a second address vector corresponding to the installed address; the dimensions of the first address vector and the second address vector respectively include at least one or more of the following: province, city, district, county, street, community, and house number; for a sample installed address, multiplying the first address vector and the second address vector by a preset score to obtain a first target address vector corresponding to the first address vector and a second target address vector corresponding to the second address vector; determining the first term frequency–inverse document frequency (TF-IDF) vector corresponding to the first target address vector and the second TF-IDF vector corresponding to the second target address vector; determining the cosine similarity using the first TF-IDF vector and the second TF-IDF vector; determining at least one cosine similarity for each installed address in the installation address database; and determining the average value of at least one cosine similarity to obtain the target address similarity between the target pre-installed address and the installed address.

[0008] Optionally, for a second address vector, determine the degree of influence of the values ​​of each dimension of the second address vector on the broadband installation result corresponding to the installed address of the second address vector; obtain a preset score based on at least one degree of influence corresponding to each second address vector.

[0009] Optionally, if the installation prediction result is greater than or equal to a preset threshold, a prompt message is output, which indicates that broadband resources can be installed at the target pre-installation address.

[0010] Optionally, the above-mentioned weighted summation of the address classification result and the target address similarity to obtain the installation prediction result includes: multiplying the address classification result by a first preset weight to obtain a first result, and multiplying the target address similarity by a second preset weight to obtain a second result; the first preset weight corresponding to the address classification result is less than the second preset weight corresponding to the target address similarity; and adding the first result and the second result to obtain the installation prediction result.

[0011] As can be seen from the above, the solution of this application provides a broadband resource prediction method. First, based on the address feature value and the pre-established resource prediction method, the address classification result is obtained. Then, the similarity between the target pre-installation address and the target address of the already installed address is determined. The address classification result and the target address similarity are weighted and added together to obtain the installation prediction result. This effectively improves the accuracy of broadband resource prediction, enhances the user experience, and solves the problem of users being unable to install due to misjudgment or omission of resource prediction.

[0012] Secondly, this application provides a broadband resource prediction device, comprising:

[0013] The acquisition module is used to obtain the target pre-installed address of the user who installed the broadband resource and generate the address feature value of the target pre-installed address.

[0014] The determination module is used to obtain address classification results based on address feature values ​​and a pre-established resource prediction method; and to determine the similarity between the target pre-installed address and the target address of the already installed address.

[0015] The calculation module is used to add the address classification results and the target address similarity in a weighted manner to obtain the installation prediction result. The installation prediction result is used to determine whether the target pre-installation address is suitable for broadband installation.

[0016] Optionally, the determination module is also used to obtain a historical data set of operator broadband installation orders; for at least one sample pre-installation address in the operator broadband installation order dataset, generate corresponding address feature values ​​for the sample pre-installation address, including whether the address is within an electronic fence, the address type, the address construction time, the type of the nearest access point, and the distance between the nearest access points; determine the address classification result of the address feature values ​​of the sample pre-installation address; the address classification result includes successful broadband installation and unsuccessful broadband installation; use the address feature values ​​of at least one sample pre-installation address and the corresponding address classification result to form a sample set; use the sample set to train the resource prediction method and determine the resource prediction method.

[0017] Thirdly, an electronic device is provided, comprising: a memory and a processor; the memory is used to store computer-executed instructions; when the electronic device is running, the processor executes the computer-executed instructions stored in the memory to cause the electronic device to execute the broadband resource prediction method as described in the first aspect and any of its implementations.

[0018] Fourthly, a computer-readable storage medium is provided, comprising: computer-executable instructions, which, when executed on a computer, cause the computer to perform the broadband resource prediction method as described in the first aspect and any of its implementations.

[0019] For a detailed description of the second to fourth aspects and their various implementations in this application, please refer to the detailed description in the first aspect and its various implementations; and for a detailed description of the beneficial effects of the second to fourth aspects and their various implementations, please refer to the beneficial effect analysis in the first aspect and its various implementations, which will not be repeated here.

[0020] These or other aspects of this application will become more readily apparent in the following description. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of a broadband resource prediction system provided in an embodiment of this application;

[0023] Figure 2 A flowchart illustrating a broadband resource prediction method provided in this application embodiment. Figure 1 ;

[0024] Figure 3 A flowchart illustrating a broadband resource prediction method provided in this application embodiment. Figure 2 ;

[0025] Figure 4 A schematic diagram of a broadband resource prediction device provided in an embodiment of this application;

[0026] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The words "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences. It should be noted that in this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or explanations.

[0029] It should be noted that any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0030] As residents' demands for material and cultural life increase, the internet has become a necessity, leading to a growing need for home broadband installations. However, due to property management and administrative reasons, some urban areas still lack suitable locations for home broadband installation. Therefore, operators need to conduct broadband resource prediction before installing broadband for users. Current technical solutions simply determine whether home broadband resources can be installed based on the electronic fence range of the access room. This can lead to misjudgments or omissions in resource prediction, resulting in users being unable to install the service and impacting user experience.

[0031] To address the aforementioned issues, this application provides a broadband resource prediction method. This method involves obtaining the target pre-installation address of a user intending to install broadband resources and generating an address feature value for that target pre-installation address. Based on the address feature value and a pre-established resource prediction method, an address classification result is obtained. The similarity between the target pre-installation address and an already installed address is determined. The address classification result and the target address similarity are weighted and summed to obtain an installation prediction result, which is used to determine whether broadband can be installed at the target pre-installation address.

[0032] The method provided in this application embodiment adopts a resource prediction method and a method of determining the similarity with the target address of the already installed address to comprehensively judge the installation prediction result. Therefore, it can effectively improve the accuracy of broadband resource prediction, enhance user experience, and solve the problem that users cannot install the equipment due to misjudgment or omission of resource prediction.

[0033] The methods described in this application embodiment can be applied to a broadband resource prediction system. See also... Figure 1 As shown, the system may include a broadband resource prediction device 101 and an operator server 102. The broadband resource prediction device 101 and the operator server 102 are communicatively connected.

[0034] On one hand, the broadband resource prediction device 101 is used to obtain the target pre-installation address of the user installing broadband resources and generate the address feature value of the target pre-installation address; based on the address feature value and a pre-established resource prediction method, an address classification result is obtained. On the other hand, the broadband resource prediction device 101 determines the target address similarity between the target pre-installation address and the already installed address; the address classification result is weighted and added to the target address similarity to obtain the installation prediction result, which is used to determine whether broadband can be installed at the target pre-installation address. The broadband resource prediction device 101 obtains information such as the target pre-installation address, historical broadband installation order datasets from the operator server 102, and a historical address database.

[0035] The operator server 102 is located on the operator network side and includes at least modules such as a database and a human-computer interaction module. The database includes at least a historical data set of operator broadband installation orders and a historical address database. The human-computer interaction module is used for inputting the target pre-installation address.

[0036] Optionally, the broadband resource prediction device 101 can be integrated into the operator server 102, or it can be directly connected to the operator server 102 via cable.

[0037] Alternatively, the aforementioned broadband resource prediction device 101 may be the aforementioned operator server 102, or it may be an electronic device, computing unit, processing device, etc., for performing broadband resource prediction.

[0038] When implementing the broadband resource prediction method in this application embodiment based on the above-mentioned broadband resource prediction system, the installation prediction result can also be comprehensively judged by the resource prediction method and the method of determining the similarity with the target address of the already installed address. This can effectively improve the accuracy of broadband resource prediction, enhance user experience, and solve the problem of users being unable to install due to misjudgment or omission of resource prediction.

[0039] The broadband resource prediction method provided in this application can be applied to electronic devices with computing and processing capabilities, such as operator servers and mainframes. Taking a broadband resource prediction device as an example, see [link to relevant documentation]. Figure 2 As shown, the method includes the following steps S201-S204.

[0040] S201. Obtain the target pre-installation address of the user installing broadband resources, and generate the address feature value of the target pre-installation address.

[0041] In some embodiments, the target pre-installation address is the detailed address of the user requiring home broadband installation, including at least the province, city, district, county, street, house number, community, building, and floor information. The broadband resource prediction device obtains the target pre-installation address of the user installing broadband resources and generates a corresponding address feature value for the target pre-installation address. The address feature value includes one or more of the following:

[0042] (1) Feature 1 “Whether the address is within the electronic fence”: Use Geographic Information System (GIS) service to resolve the latitude and longitude or planar coordinates of the target pre-installation address, and use the ray method to determine whether the target pre-installation address is within the electronic fence of the access room;

[0043] (2) Feature 2 "Address Type": The address type includes one of the following: open community, gated community, commercial building, university campus, industrial park, hotel, and professional market;

[0044] (3) Feature 3 "Address Construction Time";

[0045] (4) Feature 4 "Nearest Access Room Type" and Feature 5 "Nearest Access Room Distance": The latitude and longitude or plane coordinates of the target pre-installation address parsed by the GIS service are compared with the latitude and longitude of at least one existing access room of the operator. The distance between the current target pre-installation address and each existing access room is calculated using the latitude and longitude distance formula. The distance between the access room closest to the target pre-installation address is obtained, which is the value of Feature 5 "Nearest Access Room Distance". The type of the access room closest to the target pre-installation address is obtained, which is Feature 4 "Nearest Access Room Type", including "three-dimensional" or "self-built".

[0046] Access rooms are spaces deployed by operators to house telecommunications equipment, cable termination distribution equipment, and cable junctions, used to connect and aggregate networks inside and outside buildings. The "three-tiered" type of access room refers to access rooms located in end-point access network segments jointly funded and constructed by a third-party company and the operator, typically maintained by the third-party company; the "self-built" type refers to access rooms constructed by the operator itself, which are easier to maintain.

[0047] S202. Based on the address feature values ​​and the pre-established resource prediction method, the address classification results are obtained.

[0048] In some embodiments, the broadband resource prediction device establishes a resource prediction method by: acquiring a historical data set of operator broadband installation orders; generating corresponding address feature values ​​for at least one sample pre-installation address in the operator broadband installation order dataset; determining the address classification result of the address feature values ​​of the sample pre-installation address; the address classification result includes successful broadband installation and unsuccessful broadband installation; forming a sample set using the address feature values ​​of at least one sample pre-installation address and the corresponding address classification result; and determining the resource prediction method using the sample set.

[0049] When determining the resource prediction method using the sample set described above, optionally, the resource prediction method can be trained and tested using a sample set composed of the address feature values ​​of at least one sample pre-installed address and the corresponding address classification results, thereby determining the resource prediction method. The core idea of ​​the AdaBoost iterative algorithm is to train different weak classifiers on the same training set, and then combine these weak classifiers to form a stronger final strong classifier, thus obtaining the final classification result as the address classification result output by the resource prediction method. See also... Figure 3 As shown, the method for generating an AdaBoost prediction classification model includes the following steps S301-S303:

[0050] S301. The sample set consisting of the address feature values ​​of at least one of the sample pre-installed addresses and the corresponding address classification results is taken as the AdaBoost sample set.

[0051] S302. Preprocess the AdaBoost sample set to generate a dataset: Convert the data in the AdaBoost sample set into a processable format, convert the address feature values ​​and the installation results of the sample pre-installed addresses into enumeration types, and encode them with labels respectively. For example, when the sample pre-installed address successfully installs broadband resources, the corresponding address classification result is encoded as 1, and when the sample pre-installed address fails to install broadband resources, the corresponding address classification result is encoded as 0.5.

[0052] S303. Training and testing datasets using the AdaBoost algorithm:

[0053] (1) Split the dataset: Divide the preprocessed AdaBoost sample set into a training set and a test set. The training set accounts for 80% of the original dataset, and the test set accounts for 20%.

[0054] (2) Selecting a base classifier: Select a basic algorithm that can be used as a base classifier, such as decision tree, K-Nearest Neighbor (KNN) algorithm, etc.

[0055] (3) Initialize weights: Take the address feature values ​​of a set of pre-installed addresses and the corresponding address classification results as a sample, and initialize the weights for each sample. Initially, all samples have the same weights.

[0056] (4) Iterative training of the model: The model is trained in multiple rounds of iteration. In each round of iteration, the model parameters are adjusted according to the weight of the misclassified samples in the previous round. During the iteration, the training of the base classifier aims to minimize the weighted error rate. Different weak classifiers are trained, and then the weight of correctly classified samples is reduced while the weight of misclassified samples is increased so that the next round of iteration can pay more attention to the misclassified samples.

[0057] (5) Integrate weak classifiers: Weight the classification results of all weak classifiers to obtain the final classification result of the AdaBoost model, which is used as the address classification result.

[0058] In some embodiments, the broadband resource prediction device uses the address feature value of the target pre-installation address and the above-described resource prediction method to output a classification result as an address classification result. Optionally, the address classification result may include two values: 0.5 and 1. 0.5 indicates that broadband cannot be installed on the target pre-installation address, and 1 indicates that broadband can be installed on the target pre-installation address.

[0059] S203. Determine the similarity between the target pre-installation address and the target address of the installed address.

[0060] In some embodiments, the target pre-installed address is segmented to obtain a first address vector corresponding to the target pre-installed address; for a sample installed address in the installation address database, the installed address is segmented to obtain a second address vector corresponding to the installed address; the dimensions of the first address vector and the second address vector each contain at least one or more of the following: province, city, district, county, street, community, and house number; for a sample installed address, the first address vector and the second address vector are multiplied by a preset score to obtain a first target address vector corresponding to the first address vector and a second target address vector corresponding to the second address vector; a first TF-IDF vector corresponding to the first target address vector and a second TF-IDF vector corresponding to the second target address vector are determined; a cosine similarity is determined using the first TF-IDF vector and the second TF-IDF vector; at least one cosine similarity is determined for each installed address in the installation address database; the average of the at least one cosine similarity is determined to obtain the target address similarity between the target pre-installed address and the installed address. The target address similarity is a floating-point number less than or equal to 1.

[0061] For example, the target pre-installation address is "Athletes' Village Area 4, No. 32, Xingya 2nd Road, Shilou Town, Panyu District, Guangzhou City". The historical address database contains the following:

[0062] South District of Media Village, No. 1 Xingya Third Road, Shilou Town, Panyu District, Guangzhou

[0063] Baodi Plaza, No. 89 Pazhou Avenue, Haizhu District, Guangzhou

[0064] China Unicom Guangzhou Branch, No. 32 Zhongshan Avenue, Tianhe District, Guangzhou.

[0065] First, select an installed address from the historical address database: "No. 1, Xingya Third Road, Shilou Town, Panyu District, Guangzhou City, Media Village South Area," and denote it as B. The specific calculation process for the similarity between addresses A and B is as follows:

[0066] (1) Segment the addresses A and B, remove stop words, and obtain the term list of A as: [Guangzhou City, Panyu District, Shilou Town, Xingya Second Road, No. 32, Athletes' Village, Fourth District], and the term list of B as: [Guangzhou City, Panyu District, Shilou Town, Xingya Third Road, No. 1, Media Village, South District];

[0067] (2) Since the corpus is a collection of duplicate terms from the target pre-installation address and the historical address database, the corpus is [Guangzhou City, Panyu District, Shilou Town, Xingya Second Road, No. 32, Athletes' Village, Fourth District, Xingya Third Road, No. 1, Media Village, South District, Haizhu District, Pazhou Avenue, No. 89, Baodi Plaza, Tianhe District, Zhongshan Avenue, No. 141, China Unicom Guangzhou Branch];

[0068] (3) For addresses A and B, each word appears 1 time, for a total of 7 times. Double weight is set for road names and community names. Therefore, the occurrence of “Xingya Second Road”, “Xingya Third Road”, “Media Village” and “Athletes Village” is counted as 2 times, and the total number of occurrences is changed to 9. Term Frequency (TF) is the number of times a word appears in an article / the total number of words in the article. Therefore, the TF of “Xingya Second Road”, “Xingya Third Road”, “Media Village” and “Athletes Village” is 2 / 9≈0.22, and the TF of other words is 1 / 9≈0.11.

[0069] (4) The Inverse Document Frequency (IDF) is log(total number of documents in the corpus / number of documents containing the word), where the total number of documents is 4 (the number of address sets). The IDF value for each word is:

[0070] "Guangzhou City":

[0071] Panyu District:

[0072] Shilou Town:

[0073] "Number 32":

[0074] "Xingya Second Road", "Xingya Third Road", "No. 1", "Athletes' Village", "Area Four", "Media Village", "South Area":

[0075] (5) For each word, the TF value of the word is multiplied by the IDF value to obtain the TF-IDF value, which constitutes the TF-IDF vector. The TF-IDF vectors corresponding to word list A and word list B are as follows:

[0076] First TF-IDF vector: [0, 0.0759, 0.0759, 0.31, 0.0759, 0.31, 0.15]

[0077] Second TF-IDF vector: [0,0.0759,0.0759,0.31,0.15,0.31,0.15],

[0078] Following the steps for calculating cosine similarity, calculate the modulus and dot product of A and B respectively:

[0079]

[0080] A·B=(0×0+0.0759×0.0759+0.0759×0.0759+0.31×0.31+0.0759×0.15+0.31×0.31+0.15×0.15=0.4476).

[0081] The cosine similarity between the first TF-IDF vector and the second TF-IDF vector is as follows:

[0082]

[0083] The address similarity between A and B is 0.9549.

[0084] The cosine similarity between the first TF-IDF vector corresponding to A and the second TF-IDF vector of any installed address in the historical address database can be calculated using the above method. The average of all cosine similarities is the target address similarity between the target pre-installed address and the installed address.

[0085] In the above embodiments, the preset score is used to adjust the influence of the values ​​of each dimension in the first address vector and the second address vector on the broadband installation result. The method for obtaining the preset score is as follows: for a second address vector, determine the influence of the values ​​of each dimension of the second address vector on the broadband installation result corresponding to the installed address of the second address vector, and represent it with a quantified number; obtain the preset score based on at least one degree of influence corresponding to each second address vector.

[0086] For example, the gradient boosting algorithm can be used to calculate the preset score, and the specific steps are as follows:

[0087] (1) Prepare the dataset: Process the address dataset into corresponding feature vectors, where each feature vector contains information such as province, city, district, county, street, house number, and community. At the same time, some discrete data also need to be processed by one-hot encoding. One-hot encoding is the encoding of a set of discrete features as binary vectors. This encoding extends the values ​​to Euclidean space, which makes the distance calculation between features more reasonable.

[0088] (2) Divide the dataset into training and test sets: Divide the dataset containing several feature vectors into training and test sets. The proportion of the training set should be set to 70%, and the proportion of the test set should be 30%.

[0089] (3) Training the Gradient Boosting Decision Tree (GBDT) model: Train the GBDT model using the training set and record the preset score of each feature vector in the model.

[0090] (4) Feature selection and adjustment of preset scores: Construct a target variable, which is a feature vector multiplied by a preset score. The initial value of the target variable is the feature vector. Each time the preset score is updated, the feature vector is also updated. Use feature selection algorithms, such as ANOVA, chi-square test, mutual information, recursive feature elimination, etc., to evaluate the correlation between each feature vector and the target variable on the training set, and select the feature vector with high correlation. At the same time, combine the preset scores recorded in step (3), sort all feature vectors, and select feature vectors in descending order of preset scores. Use the test set to evaluate the selected feature vectors, that is, multiply the feature vectors in the test set by the preset score, and determine whether the broadband installation results match the installation results corresponding to each feature vector in the test set. If the prediction accuracy and generalization ability of the model for the broadband installation results meet the requirements, the final preset score can be obtained.

[0091] S204. The address classification result is weighted and added to the target address similarity to obtain the installation prediction result. The installation prediction result is used to determine whether the target pre-installation address can be installed with broadband.

[0092] In some embodiments, the address classification result is multiplied by a first preset weight to obtain a first result, and the target address similarity is multiplied by a second preset weight to obtain a second result; the first preset weight corresponding to the address classification result is less than the second preset weight corresponding to the target address similarity; the first result and the second result are added together to obtain the installation prediction result. Furthermore, both the first preset weight and the second preset weight are less than 1, and the sum of the first preset weight and the second preset weight is greater than or equal to 1.

[0093] For example, the values ​​of the first preset weight and the second preset weight are determined by the grid search method. The first preset weight can be 0.4, while the second preset weight is 0.8. The second preset weight is much different from the first preset weight, indicating that the installation results of the historical address database are given more importance in terms of the installation prediction results.

[0094] In some embodiments, if the installation prediction result is greater than or equal to a preset threshold, a prompt message is output, which is used to indicate that broadband resources can be installed at the target pre-installation address.

[0095] For example, the preset threshold is related to the first preset weight and the second preset weight. When the first preset weight is 0.4 and the second preset weight is 0.8, since the address classification result is 0.5 or 1, the target address similarity is a floating-point number less than 1. At this time, the preset threshold can be set to 0.8.

[0096] As can be seen from the above, in the broadband installation prediction method of this application embodiment, the address classification result is first obtained based on the address feature value and the pre-established resource prediction method. Then, the similarity between the target pre-installation address and the target address of the installed address is determined. The address classification result and the target address similarity are weighted and added together to obtain the installation prediction result. This effectively improves the accuracy of broadband resource prediction, enhances the user experience, and solves the problem that users cannot install due to misjudgment or omission of resource prediction.

[0097] The foregoing mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0098] In this embodiment, the broadband resource prediction device can be divided into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0099] This application also provides a broadband resource prediction device, such as... Figure 4 As shown, the device includes an acquisition module 401, a determination module 402, and a calculation module 403.

[0100] The acquisition module 401 acquires the target pre-installation address of the user installing broadband resources and generates the address feature value of the target pre-installation address.

[0101] The determination module 402 obtains the address classification result based on the address feature value and the pre-established resource prediction method; and determines the target address similarity between the target pre-installed address and the installed address.

[0102] The calculation module 403 adds the address classification result and the target address similarity in a weighted manner to obtain the installation prediction result, which is used to determine whether the target pre-installation address is suitable for broadband installation.

[0103] The acquisition module 401 is further configured to acquire a historical data set of operator broadband installation orders; for at least one sample pre-installation address in the operator broadband installation order dataset, generate corresponding address feature values ​​for the sample pre-installation address, the address feature values ​​including whether the address is within an electronic fence, the address type, the address construction time, the type of the nearest access point, and the distance between the nearest access points; determine the address classification result of the address feature values ​​of the sample pre-installation address; the address classification result includes successful broadband installation and unsuccessful broadband installation; form a sample set using the address feature values ​​of at least one sample pre-installation address and the corresponding address classification result; and train the resource prediction method using the sample set to determine the resource prediction method.

[0104] The determining module 402 is further configured to perform word segmentation on the target pre-installation address to obtain a first address vector corresponding to the target pre-installation address; for a sample installed address in the installation address database, perform word segmentation on the installed address to obtain a second address vector corresponding to the installed address; the dimensions of the first address vector and the second address vector respectively include at least one or more of province, city, district, county, street, community, and house number; for a sample installed address, multiply the first address vector and the second address vector by a preset score to obtain a first target address vector corresponding to the first address vector and a second target address vector corresponding to the second address vector; determine a first TF-IDF vector corresponding to the first target address vector and a second TF-IDF vector corresponding to the second target address vector; use the first TF-IDF vector and the second TF-IDF vector to determine a cosine similarity; for each installed address in the installation address database, determine at least one cosine similarity; determine the average value of the at least one cosine similarity to obtain the target address similarity between the target pre-installation address and the installed address.

[0105] The determining module 402 is further configured to, for a second address vector, determine the degree of influence of the values ​​of each dimension of the second address vector on the broadband installation result corresponding to the installed address corresponding to the second address vector; and obtain the preset score based on the at least one degree of influence corresponding to each second address vector.

[0106] The calculation module 403 is also used to output a prompt message if the installation prediction result is greater than or equal to a preset threshold, the prompt message being used to indicate that broadband resources can be installed at the target pre-installation address.

[0107] The calculation module 403 is further configured to multiply the address classification result by a first preset weight to obtain a first result, and multiply the target address similarity by a second preset weight to obtain a second result; the first preset weight corresponding to the address classification result is less than the second preset weight corresponding to the target address similarity; the first result and the second result are added together to obtain the installation prediction result.

[0108] When implementing the broadband resource prediction method in this application embodiment, the resource prediction method is trained based on the historical installation order dataset to obtain the address classification result of the pre-installation address. Then, the similarity between the pre-installation address and the target address is calculated. The address classification result and the target address similarity are weighted and summed to obtain the installation prediction result. Based on the operator's historical installation orders and historical address database, factors such as whether it is within an electronic fence, the type of address, the construction time of the address, the type of the nearest access point, and the distance between the nearest access points are comprehensively considered. This effectively improves the accuracy of broadband resource prediction, enhances user experience, and avoids users being unable to install due to misjudgment or omission of resource prediction.

[0109] This application embodiment can divide the communication device into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0110] Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application, such as... Figure 5As shown, the electronic device includes a processor 501 and a bus 504. Optionally, the electronic device may also include a memory 502; alternatively, the electronic device may also include a communication interface 503. The processor 501 may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 501 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field-programmable gate array, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 501 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc. The communication interface 503 is used to connect with other devices through a communication network. The communication network may be Ethernet, a wireless access network, a wireless local area network (WLAN), etc. The memory 502 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. As one possible implementation, the memory 502 may exist independently of the processor 501. The memory 502 may be connected to the processor 501 via a bus 504 for storing instructions or program code. When the processor 501 calls and executes the instructions or program code stored in the memory 502, it can implement the broadband resource prediction method provided in this application embodiment. In another possible implementation, the memory 502 can also be integrated with the processor 501. The bus 504 can be an extended industry standard architecture (EISA) bus, etc. The bus 504 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0111] Some embodiments of this application provide a computer-readable storage medium (e.g., a non-transitory computer-readable storage medium) storing computer program instructions that, when executed on a computer, cause the computer to perform the method as described in any of the above embodiments.

[0112] Exemplary examples of computer-readable storage media may include, but are not limited to: magnetic storage devices (e.g., hard disks, floppy disks, or magnetic tapes), optical discs (e.g., compact disks (23CDs), digital versatile disks (DVDs), etc.), smart cards, and flash memory devices (e.g., electrically erasable programmable read-only memory (EEPROM), cards, sticks, or key drives, etc.). The various computer-readable storage media described in this disclosure may represent one or more devices and / or other machine-readable storage media for storing information. The term "machine-readable storage medium" may include, but is not limited to, wireless channels and various other media capable of storing, containing, and / or carrying instructions and / or data.

[0113] This application provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the method of any of the above embodiments.

[0114] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for wideband resource pre-judgment, characterized in that, The method comprises the following steps: obtaining historical time operator broadband installation order data set; generating corresponding address feature values for at least one sample pre-installation address in the operator broadband installation order data set, the address feature values including whether the address is within an electronic fence, the type of the address, the construction time of the address, the type of the nearest access interval of the address, and the distance of the nearest access interval of the address; determining the address classification result of the address feature values of the sample pre-installation address; the address classification result includes successful broadband installation and unsuccessful broadband installation; using at least one of the address feature values of the sample pre-installation address and the corresponding address classification result to form a sample set; using the sample set to determine a resource pre-judgment method; obtaining a target pre-installation address of a user installing broadband resources, and generating address feature values of the target pre-installation address; obtaining an address classification result according to the address feature values and the resource pre-judgment method; determining the target address similarity between the target pre-installation address and an installed address; multiplying the address classification result by a first preset weight to obtain a first result, and multiplying the target address similarity by a second preset weight to obtain a second result; the first preset weight corresponding to the address classification result is less than the second preset weight corresponding to the target address similarity; adding the first result and the second result to obtain a broadband installation pre-judgment result, which is used to determine whether the target pre-installation address can install broadband.

2. The method of claim 1, wherein, The determination of the target address similarity between the target pre-installation address and the installed address comprises: performing word segmentation on the target pre-installation address to obtain a first address vector corresponding to the target pre-installation address; performing word segmentation on a sample installed address in a broadband installation address library to obtain a second address vector corresponding to the installed address; the dimensions of the first address vector and the second address vector at least include one or more of province, city, district, county, street, community, and house number; multiplying the first address vector and the second address vector by a preset score respectively to obtain a first target address vector corresponding to the first address vector and a second target address vector corresponding to the second address vector; determining a first TF-IDF vector corresponding to the first target address vector and a second TF-IDF vector corresponding to the second target address vector; determining a cosine similarity using the first TF-IDF vector and the second TF-IDF vector; determining at least one cosine similarity for each installed address in the broadband installation address library; determining the average value of the at least one cosine similarity to obtain the target address similarity between the target pre-installation address and the installed address.

3. The method of claim 2, wherein, Further comprising: determining the influence degree of the value of each dimension of a second address vector on the broadband installation result of the installed address corresponding to the second address vector for a second address vector; obtaining the preset score according to the at least one influence degree corresponding to each second address vector.

4. The method of claim 1, wherein, Further comprising: If the installation prediction result is greater than or equal to a preset threshold, output prompt information, the prompt information is used to indicate that the target pre-installation address can install broadband resources.

5. A broadband resource pre-judgment apparatus, characterized by, The device further comprises: A determination module is configured to acquire operator broadband installation order data sets of historical time; generate corresponding address feature values for at least one sample pre-installation address in the operator broadband installation order data sets; the address feature values include whether the address is within an electronic fence, the type of the address, the construction time of the address, the type of the nearest access interval of the address, and the distance of the nearest access interval of the address; determine the address classification result of the address feature values of the sample pre-installation address; the address classification result includes successful installation of broadband and unsuccessful installation of broadband; use the address feature values of at least one sample pre-installation address and the corresponding address classification result to form a sample set; use the sample set to determine a resource prediction method; An acquisition module is configured to acquire a target pre-installation address of a user installing broadband resources, and generate address feature values of the target pre-installation address; The determination module is further configured to obtain an address classification result according to the address feature values and a pre-established resource prediction method, and determine a target address similarity between the target pre-installation address and an installed address. A calculation module is configured to multiply the address classification result by a first preset weight to obtain a first result, multiply the target address similarity by a second preset weight to obtain a second result, add the first result and the second result to obtain an installation prediction result, and use the installation prediction result to determine whether the target pre-installation address can install broadband.

6. An electronic device, comprising: The electronic device comprises a memory and a processor; the memory is configured to store computer execution instructions; when the electronic device is running, the processor executes the computer execution instructions stored in the memory to enable the electronic device to perform the broadband resource prediction method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the broadband resource prediction method according to any one of claims 1-4.

Images