Abnormal number recognition processing method and device

Abstract

The application discloses an abnormal number identification processing method and device. The method comprises the following steps: obtaining a target number to be processed, first characteristic data corresponding to the target number, and second characteristic data reflecting whether the target number will be complained after being processed as an abnormal number; analyzing the first characteristic data by using a pre-trained abnormal number classification model to obtain a first prediction probability of the target number being an abnormal number; analyzing the second characteristic data by using a pre-trained complaint prediction model to obtain a second prediction probability of the target number being complained after being processed as an abnormal number; comparing the first prediction probability with a first preset threshold, comparing the second prediction probability with a second preset threshold, and determining a target processing strategy for the target number according to the comparison results. The application solves the technical problem that users are prone to complaints due to inaccurate identification results in related abnormal number identification schemes.

Description

Technical Field

[0001] This application relates to the field of communication service technology, and more specifically, to a method and apparatus for identifying and processing abnormal numbers. Background Technology

[0002] With the rapid advancement of information technology and the widespread adoption of smartphones, people's lives, work, and social interactions have undergone profound changes. However, this technological leap has also been accompanied by an increasing number of criminals using communication technology for illegal activities. To address this challenge, operators have implemented a series of measures to manage abnormal numbers, focusing on identifying phone numbers that may be involved in misconduct or abnormal communication and taking appropriate action. However, current methods primarily focus on identifying the numbers themselves, often neglecting the potential user dissatisfaction or complaints that may result from handling these numbers, and lacking effective predictive tools for such anomalies. Furthermore, there is an urgent need for a more reliable and efficient control scheme to manage the number of phone numbers involved in abnormal behavior and the resulting number of user complaints over a given period.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides an abnormal number identification and processing method and apparatus to at least solve the technical problem that user complaints are easily caused by inaccurate identification results in related abnormal number identification schemes.

[0005] According to one aspect of the embodiments of this application, an abnormal number identification and processing method is provided, comprising: acquiring a target number to be processed, a first type of feature data and a second type of feature data corresponding to the target number, wherein the first type of feature data is used to reflect whether the target number is abnormal, and the second type of feature data is used to reflect whether a complaint will occur after the target number is processed as an abnormal number; analyzing the first type of feature data corresponding to the target number using a pre-trained abnormal number classification model to obtain a first predicted probability that the target number is an abnormal number; analyzing the second type of feature data corresponding to the target number using a pre-trained complaint prediction model to obtain a second predicted probability that a complaint will occur after the target number is processed as an abnormal number; comparing the first predicted probability with a first preset threshold, and comparing the second predicted probability with a second preset threshold, and determining a target processing strategy for the target number based on the comparison result.

[0006] Optionally, the first type of feature data includes at least one of the following: the number's location, whether it is roaming outside the province, the time of network access, the type of service plan, the number of calls made within a preset time period, the number of calls made by different users, the caller's dispersion, the number of calls made by different provinces, the number of calls made by different cities, the number of calls made to the operator's number, the number of calls received, the number of calls received by different provinces, and the number of calls received by different cities; the second type of feature data includes at least one of the following: the number owner's age, gender, the number's location, whether it is roaming outside the province, the time of network access, the type of service plan, whether it is classified as an abnormal number, the number of complaints within the province within a preset time period, the total number of complaints, the number of calls made, and the number of times the service is suspended.

[0007] Optionally, both the abnormal number classification model and the complaint prediction model are lightweight gradient boosting models. The training process of the abnormal number classification model includes: acquiring multiple first numbers within a historical time period, using abnormal numbers among the multiple first numbers as first positive samples, and using numbers that have been shut down and have been complained about among the multiple first numbers as first negative samples; acquiring first-class feature data and second-class feature data corresponding to each first positive sample and first negative sample, and training the abnormal number classification model based on the first-class feature data and second-class feature data corresponding to each first positive sample and first negative sample; the training process of the complaint prediction model includes: acquiring multiple shut-down second numbers within a historical time period, using numbers that have been complained about among the multiple second numbers as second positive samples, and using numbers that have not been complained about among the multiple second numbers as second negative samples; acquiring first-class feature data and second-class feature data corresponding to each second positive sample and second negative sample, and training the complaint prediction model based on the first-class feature data and second-class feature data corresponding to each second positive sample and second negative sample.

[0008] Optionally, when training the abnormal number classification model or complaint prediction model, the method further includes: determining the model's hyperparameters using a Bayesian optimization algorithm, and iteratively training the model using a K-fold cross-validation method, wherein the model's hyperparameters include: learning rate, maximum depth, subsampling ratio, and column sampling ratio; in each iteration batch, adjusting the model parameters using a target loss function, and evaluating the model trained in the current batch using an F1 score, wherein the target loss function formula is as follows: In the formula, N is the number of samples in the current training batch; F m-1 (x i A m-1 ) represents the first m-1 trees of the model with parameter A. m-1 Under the condition of input value x i The predicted value, A m-1 This includes the parameters a1, ..., a of the first m-1 trees. m-1 ;L(y i ,F m-1 (xi A m-1 )) describes the input value x i The corresponding true value y i Logarithmic loss function of the error between the current model prediction and the predicted value; coefficients Among them, y i =0 indicates a negative sample, y i =1 indicates a positive sample, c is a preset constant greater than 1; λ1 represents the L1 regularization coefficient, λ2 represents the L2 regularization coefficient, Where M represents the number of features, ω j The weight is the weight corresponding to the j-th feature.

[0009] Optionally, the process of determining the first and second preset thresholds includes: obtaining an index of the number of abnormal numbers corresponding to the target time period, and determining the actual number of numbers that have been classified as abnormal numbers when the target number is obtained within the target time period; if the actual number of numbers is not greater than the index of the number of abnormal numbers, determining the pair of thresholds with the highest scores from a preset threshold scoring table as the first and second preset thresholds, wherein the threshold scoring table stores scores for multiple pairs of thresholds, each pair of thresholds including a threshold for judging whether a number is an abnormal number and a threshold for judging whether a complaint will occur after the number is treated as an abnormal number, and the score is the result of evaluating the classification effect of each pair of thresholds from the dimensions of abnormal number classification accuracy and user complaint rate; if the actual number of numbers is greater than the index of the number of abnormal numbers, determining the third preset threshold with the highest corresponding score for judging whether a number is an abnormal number from the threshold scoring table, and determining the first preset threshold that meets the target condition based on the third preset threshold, determining the threshold pair with the highest score including the first preset threshold from the threshold scoring table, and taking the other threshold in the threshold pair as the second preset threshold, wherein the target condition is: In the formula, n is the actual number of phone numbers, E is the abnormal number of phone numbers index, θ is the third preset threshold, and N is the number of phone numbers. θ θ′ represents the number of numbers that would be judged as abnormal when the third preset threshold is applied within the target time period, and θ′ represents the first preset threshold. N θ′ The number of numbers that will be judged as abnormal numbers when the first preset threshold is applied within the target time period.

[0010] Optionally, the target processing strategy includes a shutdown strategy and a reactivation strategy. The target processing strategy for the target number is determined based on the comparison results, including: if the first predicted probability is not less than the first preset threshold, the shutdown strategy is determined to be direct double shutdown, and the reactivation strategy is the preset number unblocking process; if the first predicted probability is less than the first preset threshold and the second predicted probability is less than the second preset threshold, the shutdown strategy is determined to be manual outbound call to verify the number, and the reactivation strategy is liveness authentication self-service reactivation; if the first predicted probability is less than the first preset threshold and the second predicted probability is not less than the second preset threshold, the target number is not shut down.

[0011] Optionally, the target number is a number that has been assigned an order and is judged to be a suspected abnormal number. The method further includes: if the first type of feature data and the second type of feature data corresponding to the target number are not successfully obtained for a preset number of consecutive attempts, a preset number processing strategy is executed; if the first prediction probability and the second prediction probability are not successfully obtained for a preset number of consecutive attempts, a preset number processing strategy is executed; wherein, the preset number processing strategy includes: if the target number is in use, it is directly suspended; if the target number has been protectively suspended, the target number is reactivated first and then suspended; if the target number has been suspended, no further processing is performed.

[0012] According to another aspect of the embodiments of this application, an abnormal number identification and processing device is also provided, comprising: an acquisition module, configured to acquire a target number to be processed, first type of feature data and second type of feature data corresponding to the target number, wherein the first type of feature data is used to reflect whether the target number is abnormal, and the second type of feature data is used to reflect whether a complaint will occur after the target number is processed as an abnormal number; a first analysis module, configured to analyze the first type of feature data corresponding to the target number using a pre-trained abnormal number classification model to obtain a first predicted probability that the target number is an abnormal number; a second analysis module, configured to analyze the second type of feature data corresponding to the target number using a pre-trained complaint prediction model to obtain a second predicted probability that a complaint will occur after the target number is processed as an abnormal number; and a determination module, configured to compare the first predicted probability with a first preset threshold and compare the second predicted probability with a second preset threshold, and determine a target processing strategy for the target number based on the comparison result.

[0013] According to another aspect of the embodiments of this application, a computer program product is also provided, the computer program product comprising: a computer program, wherein the computer program, when executed by a processor, implements the above-described abnormal number identification processing method.

[0014] According to another aspect of the embodiments of this application, an electronic device is also provided, the electronic device including: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the above-described abnormal number identification processing method through the computer program.

[0015] In this embodiment, the system first acquires the target number to be processed, along with first-type feature data reflecting whether the target number is abnormal and second-type feature data reflecting whether a complaint will arise after the target number is processed as an abnormal number. Then, a pre-trained abnormal number classification model is used to analyze the first-type feature data to obtain a first predicted probability that the target number is abnormal. Next, a pre-trained complaint prediction model is used to analyze the second-type feature data to obtain a second predicted probability that a complaint will arise after the target number is processed as an abnormal number. Finally, the first predicted probability is compared with a first preset threshold, and the second predicted probability is compared with a second preset threshold. Based on the comparison results, a target processing strategy for the target number is determined. This method not only improves the accuracy of abnormal number identification but also effectively reduces unnecessary user complaints by assessing potential complaint risks. This meticulous analysis and comprehensive judgment strategy ensures that the system can make more accurate and reasonable decisions when processing target numbers, balancing the identification of abnormal behavior with the maintenance of user satisfaction. In other words, this solution effectively solves the technical problem in related abnormal number identification schemes where inaccurate identification results easily lead to user complaints. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0017] Figure 1 This is a flowchart illustrating an optional abnormal number identification and processing method according to an embodiment of this application;

[0018] Figure 2 This is a schematic diagram of an optional actual business process for identifying abnormal numbers according to an embodiment of this application;

[0019] Figure 3 This is a schematic diagram of the structure of an optional abnormal number identification and processing device according to an embodiment of this application;

[0020] Figure 4 This is a schematic diagram of the structure of an optional electronic device according to an embodiment of this application. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0022] It should be noted that the terms "first," "second," etc., used in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] To better understand the embodiments of this application, the following is a translation and explanation of some nouns or terms that appear in the description of the embodiments of this application:

[0024] LightGBM (Light Gradient Boosting Machine) is an efficient gradient boosting decision tree model designed to address the slow training speed and high memory consumption of traditional gradient boosting tree algorithms on large datasets. LightGBM uses histograms to approximate the feature distribution, thus accelerating the process of finding split points. LightGBM employs a "Leaf-wise" splitting strategy instead of the traditional "Level-wise" splitting strategy, enabling the model to find the optimal splitting path faster. LightGBM also supports multi-threaded parallel training, leveraging the advantages of multi-core processors.

[0025] Dispatch: In this application, dispatch refers to the process by which related technologies, in the process of handling abnormal numbers, assign numbers to be processed to a specialized processing mechanism, such as a high-risk number prediction model, for processing.

[0026] Example 1

[0027] According to an embodiment of this application, an abnormal number identification processing method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0028] Figure 1 This is a flowchart illustrating an abnormal number identification and processing method according to an embodiment of this application, as shown below. Figure 1 As shown, the method includes the following steps:

[0029] Step S102: Obtain the target number to be processed, the first type of feature data and the second type of feature data corresponding to the target number, wherein the first type of feature data is used to reflect whether the target number is abnormal, and the second type of feature data is used to reflect whether a complaint will occur after the target number is processed as an abnormal number;

[0030] Step S104: Analyze the first type of feature data corresponding to the target number using the pre-trained abnormal number classification model to obtain the first predicted probability that the target number is an abnormal number;

[0031] Step S106: Analyze the second type of feature data corresponding to the target number using the pre-trained complaint prediction model to obtain the second predicted probability of a complaint occurring after the target number is treated as an abnormal number.

[0032] Step S108: Compare the first predicted probability with the first preset threshold, and compare the second predicted probability with the second preset threshold, and determine the target processing strategy for the target number based on the comparison results.

[0033] The following describes each step of the abnormal number identification and processing method in conjunction with the specific implementation process.

[0034] As an optional implementation, in order to improve the accuracy of abnormal number identification while reducing the user complaint rate, this application embodiment introduces an abnormal number classification model and a complaint prediction model, respectively, to analyze numbers from two dimensions: abnormal number identification and user complaint prediction. Both models can be trained using the LightGBM model.

[0035] Optionally, the training process of the abnormal number classification model can be carried out in the following way: First, obtain multiple first numbers within a historical time period. These first numbers may include data from new high-risk calls, high-risk call models, and surge high-risk call models under the management and operation platform. Abnormal numbers among the multiple first numbers are used as first positive samples, and numbers among the multiple first numbers that have been shut down and have been complained about are used as first negative samples. Then, obtain the first type of feature data corresponding to each first positive sample and first negative sample, and train the abnormal number classification model based on the first type of feature data corresponding to each first positive sample and first negative sample. The type of first type of feature data may include at least one of the following: the number's location, whether it is roaming outside the province, the time of network access, the package type, the number of calls within a preset time period, the number of calls from the calling user, the caller dispersion, the number of calls from the calling province, the number of calls from the calling city, the number of calls from the calling operator's number, the number of calls received, the number of calls from the called province, and the number of calls from the called city. The preset time period may be 1 day, 7 days, 1 month, 3 months, 6 months, 12 months, etc.

[0036] Optionally, the training process of the complaint prediction model can be carried out in the following way: First, obtain multiple blocked secondary numbers within a historical time period. Numbers with complaints received from these secondary numbers are designated as positive secondary samples, and numbers without complaints are designated as negative secondary samples. Then, obtain the second-class feature data corresponding to each positive and negative secondary sample, and train the complaint prediction model based on this data. The type of the second-class feature data can include at least one of the following: the number owner's age, gender, number's location, whether it is roaming outside the province, network access time, package type, whether it is classified as an abnormal number, the number of complaints within the province within a preset time period, the total number of complaints, the number of outgoing calls, and the number of times the service is suspended. This preset time period can also be 1 day, 7 days, 1 month, 3 months, 6 months, 12 months, etc.

[0037] To address the issue of relatively scarce abnormal number samples in the training data, and in order to improve the accuracy of abnormal number identification and reduce overfitting, this application improves the loss function of the conventional LightGBM model by adding L1 and L2 regularization terms and assigning higher weights to abnormal samples. The specific training method is as follows:

[0038] In each iteration batch, the model parameters are adjusted using the objective loss function, and the F1 score is used to evaluate the model trained in the current batch. The objective loss function formula is as follows:

[0039]

[0040] In the above formula, N is the number of samples in the current training batch; F m-1 (xi A m-1 ) represents the first m-1 trees of the model with parameter A. m-1 Under the condition of input value x i The predicted value, A m-1 This includes the parameters a1, ..., a of the first m-1 trees. m-1 ;L(y i ,F m-1 (x i A m-1 )) describes the input value x i The corresponding true value y i Logarithmic loss function of the error between the current model prediction and the predicted value; coefficients Among them, y i =0 indicates a negative sample, y i =1 indicates a positive sample, c is a preset constant greater than 1; λ1 represents the L1 regularization coefficient, λ2 represents the L2 regularization coefficient, Where M represents the number of features, ω j The weight is the weight corresponding to the j-th feature.

[0041] As an optional implementation, an effective strategy in optimizing and training abnormal number classification models or complaint prediction models is to introduce Bayesian optimization algorithms to accurately determine the model's hyperparameters. These hyperparameters include, but are not limited to, learning rate, maximum depth, subsampling ratio, and column sampling ratio. To further verify the model's stability and generalization ability, iterative training is performed using K-fold cross-validation to ensure that the model performs well on different data subsets.

[0042] Optionally, to comprehensively evaluate the model's performance, recall, precision, and F1 score are selected as key evaluation metrics. These metrics are constructed based on a confusion matrix, which is an n x n matrix that visually displays the correspondence between the model's predictions and the actual results. In the context of binary classification problems, it is generally represented as shown in Table 1.

[0043] Table 1. Correspondence between prediction results and actual results in the context of binary classification problems.

[0044]

[0045] In Table 1, T (True) is generally used to represent correct, F (False) to represent incorrect, P (Positive) to represent 1, and N (Negative) to represent 0. Wherein:

[0046] TP: Predicted value was 1, actual value was 0, prediction was correct;

[0047] FP: Predicted value is 1, actual value is 0, prediction error;

[0048] FN: Predicted value was 0, actual value was 1, prediction error;

[0049] TN: Predicted to be 0, actually 0, prediction was correct.

[0050] Precision refers to the prediction result; it represents the probability that a given sample is actually positive out of all samples predicted as positive. The expression is:

[0051] Recall refers to the original sample; it represents the probability that a sample that is actually positive will be predicted as positive. The expression is:

[0052] The F1 score considers both precision and recall, aiming to maximize both while achieving a balance. The expression is:

[0053]

[0054] By calculating and comparing these evaluation metrics, a clear understanding of the performance of the abnormal number classification model and the complaint prediction model can be obtained. If the model's performance does not meet the expected standards, it can be further trained for optimization and improvement. If the model's performance evaluation is satisfactory, the final abnormal number classification model and complaint prediction model can be directly output.

[0055] After obtaining the above-mentioned abnormal number classification model and complaint prediction model, these two models are used to process the target numbers. The specific process of using the two models to process the target numbers is as follows:

[0056] First, obtain the target number to be processed. This target number can be a suspected abnormal number that has been judged through order dispatching, or it can be a non-suspected abnormal number. After obtaining the target number, obtain the first type of feature data and the second type of feature data corresponding to the target number.

[0057] The first type of feature data of the target number is imported into the abnormal number classification model to obtain the first predicted probability, and the second type of feature data of the target number is imported into the complaint prediction model to obtain the second predicted probability.

[0058] Optionally, after obtaining the predicted probability, a first preset threshold and a second preset threshold can be set for the first predicted probability and the second predicted probability according to the pre-stored threshold scoring table, thereby determining the processing strategy for the target number. The threshold scoring table stores multiple pairs of threshold scores. Each pair of thresholds includes a threshold for judging whether the number is an abnormal number and a threshold for judging whether a complaint will occur after the number is treated as an abnormal number. The score is the result of evaluating the classification effect of each pair of thresholds from the dimensions of abnormal number classification accuracy and user complaint rate.

[0059] As an alternative implementation, when setting the threshold score table, it is possible to analyze how the accuracy rates of the abnormal number classification model and the complaint prediction model change as the preset threshold is adjusted; at the same time, examine how the number of numbers processed changes as the first preset threshold and the second preset threshold change under different processing strategies; in addition, it is also possible to observe the changing trends of the number of unclassified abnormal numbers and the number of complaints as these two thresholds are adjusted; by comprehensively evaluating these dimensions, a comprehensive understanding of the impact of each threshold combination can be obtained, thereby obtaining the result of the threshold score and generating the threshold score table.

[0060] Optionally, in an actual scenario, for the identification of abnormal numbers and complaint prediction, it is often necessary to consider the abnormal number quantity index within the target time period, such as monthly indicators, etc. Due to the fluctuations in the abnormal number data, the number of abnormal numbers and complaints within the target time period may far exceed the abnormal number quantity index for that time period. The above process for determining the threshold is applicable when the actual number of numbers is not greater than the abnormal number quantity index. When the actual number of numbers is greater than the abnormal number quantity index, during the implementation of this application, a method is provided. This method adjusts the probability threshold set by the model according to the quantity index, thereby controlling the model prediction result. The design idea of the method is as follows:

[0061] Set the abnormal number quantity index within the target time period as E, and the actual number of numbers as n. If E < n, then the current excess quantity is: n - E, and the excess multiple is:

[0062] Determine the third preset threshold with the highest corresponding score in the threshold score table for judging whether a number is an abnormal number, and based on the third preset threshold, determine the first preset threshold that meets the target condition. Determine the threshold pair with the highest score including the first preset threshold from the threshold score table, and use the other threshold in the threshold pair as the second preset threshold. The target condition is:

[0063]

[0064] Simplify the above formula to get: where θ is the third preset threshold, N θ is the number of numbers that will be judged as abnormal numbers when applying the third preset threshold within the target time period, θ′ is the first preset threshold, and N θ′ is the number of numbers that will be judged as abnormal numbers when applying the first preset threshold within the target time period.

[0065] Optionally, the first predicted probability is compared with a first preset threshold, and the second predicted probability is compared with a second preset threshold. Based on the comparison results, a target processing strategy for the target number is determined. This target strategy includes a shutdown strategy and a reactivation strategy. Specifically: if the first predicted probability is not less than the first preset threshold, the shutdown strategy is a direct double shutdown, and the reactivation strategy is a preset number unblocking process; if the first predicted probability is less than the first preset threshold and the second predicted probability is less than the second preset threshold, the shutdown strategy is a manual outbound call to verify the number, and the reactivation strategy is a liveness detection self-service reactivation; if the first predicted probability is less than the first preset threshold and the second predicted probability is not less than the second preset threshold, the target number is not shut down. Assuming Pa and Pb are the first and second preset thresholds respectively, P1 is the first predicted probability, and P2 is the second predicted probability, as shown in Table 2, this target strategy is divided into three levels: strict, relatively strict, and lenient, corresponding to the three situations mentioned above.

[0066] Table 2. Target Strategy Illustration

[0067]

[0068] As an optional implementation method, the method for determining whether a number is an abnormal number in this application embodiment can be used alone or in combination with existing abnormal number processing procedures.

[0069] In the existing abnormal number handling process, the most stringent suspension or reinstatement strategy is adopted for suspected abnormal numbers generated by existing high-risk models. This means that numbers in use or under protective single suspension are upgraded to double suspension and reinstatement materials are required for review. In this application, however, for suspected abnormal numbers generated by high-risk models, feature extraction can be performed. The corresponding features are input into the abnormal number classification model and the complaint prediction model to obtain the probability that the number is judged as an abnormal number and the probability of a complaint. The suspension strategy is determined based on the magnitude of these two probability values, including strict level (double suspension), medium level (manual outbound verification of the number), and lenient level (no suspension). Among them, when manually verifying the number, if the verification fails, the number is under protective single suspension; if the verification passes, the number is not suspended.

[0070] As an optional implementation, if the target number is a number that has been assigned an order and is judged to be a suspected abnormal number, it can also be processed in the following way: if the first type of feature data and the second type of feature data corresponding to the target number are not successfully obtained for a preset number of consecutive attempts, a preset number processing strategy is executed; if the first prediction probability and the second prediction probability are not successfully obtained for a preset number of consecutive attempts, a preset number processing strategy is executed; wherein, the preset number processing strategy is the existing abnormal number processing method, for example: if the target number is in use, it is directly suspended; if the target number has been protectively suspended, the target number is first reactivated and then suspended; if the target number has been suspended, no further processing is performed.

[0071] Figure 2 This application provides a business processing flow for numbers, such as... Figure 2 As shown, B represents numbers identified as in use after automatic order dispatch, C represents numbers that have been suspended for protection after automatic order dispatch, and E represents numbers not subject to automatic order dispatch. E can be a number automatically assigned or provided by a service or communication platform and directly enters the abnormal number comprehensive order dispatch process. The probability model in this process refers to the abnormal number classification model and complaint prediction model in this application. Figure 2 In this process, suspected abnormal numbers, after automatic order dispatch, are assessed by the model to determine the call type. If it is not a high-risk call, the current status of the number is then assessed. If the number is not in use, no action is taken; if it is in use, a comprehensive abnormal number order is dispatched. If it is a high-risk call, the current status of the number is assessed again. If it is a suspected abnormal number subject to single / double suspension, the abnormal number classification model and complaint prediction model are used to determine whether to implement a strategy of not shutting down, protective single suspension, or double suspension. For numbers in categories B, C, and E, a comprehensive abnormal number order is dispatched. After the model determines the call type, if it is not a high-risk call, a manual outbound call is made to verify the number. If the verification is successful, the number is not shut down; if the verification fails, the number is protectively suspended. If it is a high-risk call, after being processed by the abnormal number classification model and complaint prediction model of this application, it will be divided into three categories: not shut down, protective single suspension, or double suspension.

[0072] In this embodiment, the system first acquires the target number to be processed, along with first-type feature data reflecting whether the target number is abnormal and second-type feature data reflecting whether a complaint will arise after the target number is processed as an abnormal number. Then, a pre-trained abnormal number classification model is used to analyze the first-type feature data to obtain a first predicted probability that the target number is abnormal. Next, a pre-trained complaint prediction model is used to analyze the second-type feature data to obtain a second predicted probability that a complaint will arise after the target number is processed as an abnormal number. Finally, the first predicted probability is compared with a first preset threshold, and the second predicted probability is compared with a second preset threshold. Based on the comparison results, a target processing strategy for the target number is determined. This method not only improves the accuracy of abnormal number identification but also effectively reduces unnecessary user complaints by assessing potential complaint risks. This meticulous analysis and comprehensive judgment strategy ensures that the system can make more accurate and reasonable decisions when processing target numbers, balancing the identification of abnormal behavior with the maintenance of user satisfaction. In other words, this solution effectively solves the technical problem in related abnormal number identification schemes where inaccurate identification results easily lead to user complaints.

[0073] Example 2

[0074] According to an embodiment of this application, an abnormal number identification and processing device for implementing the abnormal number identification and processing method in Embodiment 1 is also provided, such as... Figure 3 As shown, the abnormal number identification and processing device includes at least: an acquisition module 31, a first analysis module 32, a second analysis module 33, and a determination module 34, wherein:

[0075] The acquisition module 31 is used to acquire the target number to be processed, the first type of feature data and the second type of feature data corresponding to the target number, wherein the first type of feature data is used to reflect whether the target number is abnormal, and the second type of feature data is used to reflect whether a complaint will occur after the target number is processed as an abnormal number.

[0076] The first analysis module 32 is used to analyze the first type of feature data corresponding to the target number using a pre-trained abnormal number classification model to obtain the first predicted probability that the target number is an abnormal number.

[0077] The second analysis module 33 is used to analyze the second type of feature data corresponding to the target number using a pre-trained complaint prediction model to obtain the second predicted probability of a complaint occurring after the target number is treated as an abnormal number.

[0078] The determination module 34 is used to compare the first predicted probability with the first preset threshold and the second predicted probability with the second preset threshold, and determine the target processing strategy for the target number based on the comparison result.

[0079] The following describes the functions of each module of the abnormal number identification and processing device in conjunction with the specific implementation process.

[0080] First, the acquisition module obtains the target number to be processed. This target number can be a suspected abnormal number that has been judged by the existing model dispatching process, or it can be a non-suspected abnormal number. After obtaining the target number, the first type of feature data and the second type of feature data corresponding to the target number are then obtained.

[0081] The first analysis module imports the first type of feature data of the target number into the abnormal number classification model to obtain the first predicted probability, and the second analysis module imports the second type of feature data of the target number into the complaint prediction model to obtain the second predicted probability.

[0082] Optionally, after obtaining the predicted probability, a first preset threshold and a second preset threshold can be set for the first predicted probability and the second predicted probability according to the pre-stored threshold scoring table, thereby determining the processing strategy for the target number. The threshold scoring table stores multiple pairs of threshold scores. Each pair of thresholds includes a threshold for judging whether the number is an abnormal number and a threshold for judging whether a complaint will occur after the number is treated as an abnormal number. The score is the result of evaluating the classification effect of each pair of thresholds from the dimensions of abnormal number classification accuracy and user complaint rate.

[0083] As an optional implementation, when setting the threshold scoring table, one can analyze how the accuracy of the abnormal number classification model and the complaint prediction model changes with the adjustment of preset thresholds; at the same time, one can examine how the number of numbers processed changes with the changes of the first preset threshold and the second preset threshold under different processing strategies; in addition, one can observe the changing trends of the number of unclassified abnormal numbers and the number of complaints as these two thresholds are adjusted; by comprehensively evaluating these dimensions, one can have a comprehensive understanding of the impact of each threshold combination, thereby obtaining the threshold scoring results and generating a threshold scoring table.

[0084] Optionally, in real-world scenarios, identifying abnormal phone numbers and predicting complaints often requires considering indicators such as the number of abnormal phone numbers within a target time period, such as monthly indicators. Due to fluctuations in abnormal phone number data, the number of abnormal phone numbers and complaints within a target time period may far exceed the indicator for the number of abnormal phone numbers within that time period. The threshold determination process described above applies when the actual number of phone numbers is not greater than the indicator for the number of abnormal phone numbers. When the actual number of phone numbers exceeds the indicator for the number of abnormal phone numbers, this application provides a method in its implementation that adjusts the probability threshold set by the model based on the quantity indicator, thereby controlling the model's prediction results. The design idea of ​​this method is as follows:

[0085] Set the abnormal number quantity index within the target time period as E, and the actual number of numbers as n. If E < n, then the current excess quantity is: n - E, and the excess multiple is:

[0086] Determine the third preset threshold with the highest corresponding score in the threshold score table for judging whether a number is an abnormal number, and determine the first preset threshold that meets the target condition based on the third preset threshold. Determine the threshold pair with the highest score including the first preset threshold from the threshold score table, and use the other threshold in the threshold pair as the second preset threshold. The target condition is:

[0087]

[0088] Simplify the above formula to get: where θ is the third preset threshold, N θ is the number of numbers that will be judged as abnormal numbers when applying the third preset threshold within the target time period, θ′ is the first preset threshold, and N θ′ is the number of numbers that will be judged as abnormal numbers when applying the first preset threshold within the target time period.

[0089] Optionally, after the determination module compares the above first prediction probability with the first preset threshold and compares the above second prediction probability with the second preset threshold, it determines the target processing strategy for the target number according to the comparison result. The target strategy includes a shutdown strategy and a resume strategy, where: if the first prediction probability is not less than the first preset threshold, determine the shutdown strategy as direct dual shutdown, and the resume strategy as a preset number解封流程; if the first prediction probability is less than the first preset threshold and the second prediction probability is less than the second preset threshold, determine the shutdown strategy as manual outbound call to verify the number, and the resume strategy as live authentication self-resume; if the first prediction probability is less than the first preset threshold and the second prediction probability is not less than the second preset threshold, do not shut down the target number.

[0090] As an optional implementation manner, if the target number is a number that has been assigned a task and is judged as a suspected abnormal number, the following method can also be used for processing: if the first type of characteristic data and the second type of characteristic data corresponding to the target number cannot be successfully obtained for a continuous preset number of times, execute the preset number processing strategy; if the first prediction probability and the second prediction probability cannot be successfully obtained for a continuous preset number of times, execute the preset number processing strategy; where the preset number processing strategy is the existing abnormal number processing method, for example: if the target number is in use, directly perform dual shutdown, if the target number has been protected by a single shutdown, first resume the target number and then perform dual shutdown, if the target number has been dual shutdown, do not process it anymore.

[0091] It should be noted that each module in the abnormal number identification processing device in this application embodiment corresponds one-to-one with each implementation step of the abnormal number identification processing method in embodiment 1. Since embodiment 1 has been described in detail, some details not shown in this embodiment can be referred to embodiment 1, and will not be elaborated further here.

[0092] Example 3

[0093] According to an embodiment of this application, a computer program product is also provided, which includes a computer program, wherein when the computer program is executed by a processor, it implements the abnormal number identification processing method in Embodiment 1.

[0094] According to an embodiment of this application, a non-volatile storage medium is also provided, which includes a stored computer program, wherein the device containing the non-volatile storage medium executes the abnormal number identification processing method in Embodiment 1 by running the computer program.

[0095] According to an embodiment of this application, a processor is also provided for running a computer program, wherein the computer program executes the abnormal number identification processing method in Embodiment 1 during runtime.

[0096] According to an embodiment of this application, an electronic device is also provided, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the abnormal number identification processing method of Embodiment 1 through the computer program.

[0097] Specifically, the computer program executes the following steps during runtime: acquiring the target number to be processed, the first type of feature data and the second type of feature data corresponding to the target number, wherein the first type of feature data is used to reflect whether the target number is abnormal, and the second type of feature data is used to reflect whether a complaint will occur after the target number is processed as an abnormal number; analyzing the first type of feature data corresponding to the target number using a pre-trained abnormal number classification model to obtain a first predicted probability that the target number is an abnormal number; analyzing the second type of feature data corresponding to the target number using a pre-trained complaint prediction model to obtain a second predicted probability that a complaint will occur after the target number is processed as an abnormal number; comparing the first predicted probability with a first preset threshold, and comparing the second predicted probability with a second preset threshold, and determining the target processing strategy for the target number based on the comparison results.

[0098] As an alternative implementation, the above-mentioned electronic device may exist in the form of a mobile terminal, a computer terminal, or a similar computing device. Figure 4 A hardware block diagram of an electronic device for implementing an abnormal number identification and processing method is shown. Figure 4As shown, the electronic device 40 may include one or more processors 402 (shown as 402a, 402b, ..., 402n in the figure) 402 (processor 402 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 404 for storing data, and a transmission device 406 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 4 The structure shown is for illustrative purposes only and does not limit the structure of the electronic device described above. For example, electronic device 40 may also include... Figure 4 The more or fewer components shown, or having the same Figure 4 The different configurations shown.

[0099] It should be noted that the aforementioned one or more processors 402 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element of the electronic device 40. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0100] The memory 404 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the abnormal number identification processing method in this embodiment. The processor 402 executes various functional applications and data processing by running the software programs and modules stored in the memory 404, thereby implementing the above-mentioned application vulnerability detection method. The memory 404 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 404 may further include memory remotely located relative to the processor 402, and these remote memories can be connected to the electronic device 40 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0101] The transmission device 406 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the electronic device 40. In one example, the transmission device 406 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 406 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0102] The display can be, for example, a touchscreen liquid crystal display (LCD), which allows the user to interact with the user interface of the electronic device 40.

[0103] The sequence numbers of the above embodiments are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0104] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0105] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0106] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0107] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0108] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0109] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for identifying and processing abnormal phone numbers, characterized in that, include: The target number to be processed, the first type of feature data and the second type of feature data corresponding to the target number are obtained, wherein the first type of feature data is used to reflect whether the target number is abnormal, and the second type of feature data is used to reflect whether a complaint will occur after the target number is processed as an abnormal number; The first type of feature data corresponding to the target number is analyzed using a pre-trained abnormal number classification model to obtain the first predicted probability that the target number is an abnormal number; The second type of feature data corresponding to the target number is analyzed using a pre-trained complaint prediction model to obtain the second predicted probability of a complaint occurring after the target number is treated as an abnormal number. The first predicted probability is compared with a first preset threshold, and the second predicted probability is compared with a second preset threshold. Based on the comparison results, a target processing strategy for the target number is determined. Both the abnormal number classification model and the complaint prediction model are lightweight gradient boosting models. The training process of the abnormal number classification model includes: acquiring multiple first numbers within a historical time period, taking the abnormal numbers among the multiple first numbers as first positive samples, and taking the numbers among the multiple first numbers that have been shut down and have been complained about as first negative samples; acquiring the first type of feature data corresponding to each first positive sample and the first negative sample, and training the abnormal number classification model based on the first type of feature data corresponding to each first positive sample and the first negative sample. The training process of the complaint prediction model includes: acquiring multiple shut-down second numbers within a historical time period; using the numbers among the multiple second numbers that have been complained about as second positive samples and the numbers among the multiple second numbers that have not been complained about as second negative samples; acquiring the second type of feature data corresponding to each second positive sample and the second negative sample; and training the complaint prediction model based on the second type of feature data corresponding to each second positive sample and the second negative sample.

2. The method according to claim 1, characterized in that, The first type of feature data includes at least one of the following: the location of the number, whether it is roaming outside the province, the time of network access, the type of package, the number of calls within a preset time period, the number of calls from users, the caller dispersion, the number of calls from provinces, the number of calls from cities, the number of calls from operator numbers, the number of calls received, the number of calls from provinces received, and the number of calls from cities received. The second type of feature data includes at least one of the following: the number owner's age, gender, number's location, whether it is roaming outside the province, network access time, package type, whether it is classified as an abnormal number, number of complaints within the province within a preset time period, total number of complaints, number of outgoing calls, and number of times the service is suspended.

3. The method according to claim 1, characterized in that, When training the abnormal number classification model or the complaint prediction model, the method further includes: The hyperparameters of the model are determined using the Bayesian optimization algorithm, and the model is iteratively trained using the K-fold cross-validation method. The hyperparameters of the model include: learning rate, maximum depth, subsampling ratio, and column sampling ratio. In each iteration batch, the model parameters are adjusted using the objective loss function, and the F1 score is used to evaluate the model trained in the current batch. The objective loss function is formulated as follows: ' In the formula, This represents the number of samples in the current training batch. This indicates that the first m-1 trees in the model have parameters... Under the condition of input value The predicted value, It includes the parameters of the first m-1 trees. ; To describe the input value Corresponding true value Logarithmic loss function of the error between the current model prediction and the predicted value; coefficients ,in, Indicates a negative sample. Indicates a positive sample. It is a preset constant greater than 1; Represents the L1 regularization coefficient. Represents the L2 regularization coefficient. ,in, Indicates the number of features. For the first The weights corresponding to each feature.

4. The method according to claim 1, characterized in that, The process of determining the first preset threshold and the second preset threshold includes: Obtain the abnormal number quantity index corresponding to the target time period, and determine the actual number of numbers that have been classified as abnormal numbers when the target number is obtained within the target time period; If the actual number of numbers is not greater than the number of abnormal numbers indicator, the pair of thresholds with the highest scores are determined from the preset threshold scoring table as the first preset threshold and the second preset threshold. The threshold scoring table stores scores for multiple pairs of thresholds. Each pair of thresholds includes a threshold for determining whether a number is an abnormal number and a threshold for determining whether a complaint will occur after the number is treated as an abnormal number. The score is the result of evaluating the classification effect of each pair of thresholds from the dimensions of abnormal number classification accuracy and user complaint rate. If the actual number of phone numbers is greater than the number of abnormal phone numbers, a third preset threshold with the highest corresponding score for determining whether a number is an abnormal number is determined from the threshold scoring table. Based on the third preset threshold, a first preset threshold that satisfies the target condition is determined. A threshold pair with the highest score, including the first preset threshold, is determined from the threshold scoring table, and the other threshold in the threshold pair is used as the second preset threshold. The target condition is: ' In the formula, The actual number of numbers, This refers to the number of abnormal phone numbers. The third preset threshold, The number of phone numbers that would be judged as abnormal when the third preset threshold is applied within the target time period. The first preset threshold, The number of numbers that would be judged as abnormal numbers when the first preset threshold is applied within the target time period.

5. The method according to claim 1, characterized in that, The target processing strategy includes: a shutdown strategy and a reactivation strategy. The target processing strategy for the target number is determined based on the comparison results, including: If the first predicted probability is not less than the first preset threshold, the shutdown strategy is determined to be direct double shutdown, and the reactivation strategy is the preset number unblocking process; If the first predicted probability is less than the first preset threshold and the second predicted probability is less than the second preset threshold, the shutdown strategy is determined to be manual outbound call to verify the number, and the reactivation strategy is self-service reactivation with liveness authentication. If the first predicted probability is less than the first preset threshold and the second predicted probability is not less than the second preset threshold, the target number will not be shut down.

6. The method according to claim 1, characterized in that, The target number is a number that has already been assigned an order and has been identified as a suspected abnormal number. The method further includes: If the first type of feature data and the second type of feature data corresponding to the target number are not successfully obtained after a preset number of consecutive attempts, the preset number processing strategy is executed. If the first prediction probability and the second prediction probability are not successfully obtained after a preset number of consecutive attempts, a preset number processing strategy is executed. The preset number processing strategy includes: if the target number is in use, it will be directly suspended; if the target number has been protected by a single suspension, the target number will be reactivated first and then suspended; if the target number has been suspended, no further processing will be performed.

7. An abnormal number identification and processing device, characterized in that, include: The acquisition module is used to acquire the target number to be processed, the first type of feature data and the second type of feature data corresponding to the target number, wherein the first type of feature data is used to reflect whether the target number is abnormal, and the second type of feature data is used to reflect whether a complaint will occur after the target number is processed as an abnormal number; The first analysis module is used to analyze the first type of feature data corresponding to the target number using a pre-trained abnormal number classification model to obtain a first predicted probability that the target number is an abnormal number. The abnormal number classification model is a lightweight gradient boosting model. The training process of the abnormal number classification model includes: acquiring multiple first numbers within a historical time period; using abnormal numbers among the multiple first numbers as first positive samples; and using numbers among the multiple first numbers that have been shut down and have been complained about as first negative samples; acquiring the first type of feature data corresponding to each first positive sample and the first negative sample; and training the abnormal number classification model based on the first type of feature data corresponding to each first positive sample and the first negative sample. The second analysis module is used to analyze the second type of feature data corresponding to the target number using a pre-trained complaint prediction model to obtain a second predicted probability of a complaint occurring after the target number is treated as an abnormal number. The complaint prediction model is a lightweight gradient boosting model, and its training process includes: acquiring multiple blocked second numbers within a historical time period; using numbers that have been complained about as second positive samples and numbers that have not been complained about as second negative samples; acquiring the second type of feature data corresponding to each second positive and second negative sample; and training the complaint prediction model based on the second type of feature data corresponding to each second positive and second negative sample. The determination module is used to compare the first predicted probability with a first preset threshold and the second predicted probability with a second preset threshold, and determine the target processing strategy for the target number based on the comparison results.

8. A computer program product, characterized in that, include: A computer program, wherein when executed by a processor, the computer program implements the abnormal number identification processing method according to any one of claims 1 to 6.

9. An electronic device, characterized in that, include: A memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the abnormal number identification processing method according to any one of claims 1 to 6 through the computer program.

Images