A telephone identification method, an electronic device and a storage medium

Abstract

The application discloses a telephone identification method, an electronic device and a storage medium, and relates to the technical field of communication security. The application comprises the following steps: acquiring dialing information of a calling number; then, respectively calculating feature similarities between the dialing information and feature information corresponding to a plurality of number categories, wherein the plurality of number categories are obtained by clustering a plurality of sample dialing information; then, determining a target number category corresponding to the calling number according to the plurality of feature similarities; labeling the calling number according to the target number category; and finally, outputting the labeling information to display the labeling information on a terminal corresponding to a called number. Thus, the current calling number can be identified and displayed in real time, the user can be reminded in real time, and the accuracy of telephone identification can be improved.

Description

Technical Field

[0001] This invention relates to the field of communication security technology, and in particular to a telephone identification method, electronic device, and storage medium. Background Technology

[0002] Currently, with the widespread use of convenient payment methods such as online payments, online fraud cases are frequent. In particular, users may not be able to promptly identify whether a call from an unknown number is a scam, thus potentially suffering financial losses.

[0003] In existing technologies, a common approach is to provide feedback after a user identifies an unknown call as a scam call, and then label the call based on this feedback. If the caller attempts to call again, the call is blocked or a warning message is sent to the user. However, this method of call identification relies too heavily on user feedback and suffers from significant delays in call identification. Summary of the Invention

[0004] In view of the above problems, the present invention is proposed to provide a telephone identification method, electronic device and storage medium that overcomes or at least partially solves the above problems.

[0005] Based on a first aspect of the present invention, a telephone identification method is provided, the method comprising:

[0006] Obtain the dialing information of the calling number;

[0007] Calculate the feature similarity between the dialing information and the feature information corresponding to multiple preset number categories, wherein the multiple number categories are obtained by clustering multiple sample dialing information;

[0008] Based on multiple feature similarities, the target number category corresponding to the calling number is determined;

[0009] The calling numbers are labeled according to the target number category;

[0010] The annotation information is output to be displayed on the terminal side corresponding to the called number.

[0011] Based on a second aspect of the present invention, an electronic device is also provided, comprising:

[0012] One or more processors;

[0013] Memory;

[0014] One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to perform any of the methods described above.

[0015] Based on a third aspect of the invention, a computer-readable storage medium is also provided for storing a computer program for use in conjunction with an electronic device, the computer program being executable by a processor to perform any of the methods described above.

[0016] Compared with existing technologies, this invention involves acquiring the dialing information of the calling number, then calculating the feature similarity between the dialing information and feature information corresponding to multiple preset number categories. These number categories are obtained by clustering multiple sample dialing information. Next, based on the multiple feature similarities, the target number category corresponding to the calling number is determined. The calling number is then labeled according to the target number category. Finally, the labeled information is output and displayed on the terminal corresponding to the called number. This allows for real-time identification and display of the current calling number, providing real-time reminders to the user, and improving the accuracy of call identification.

[0017] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings.

[0019] In the attached diagram:

[0020] Figure 1 This is a flowchart illustrating the steps of a telephone identification method provided in an embodiment of the present invention;

[0021] Figure 2 This is a schematic diagram of the steps for generating multiple number categories through clustering, provided in an embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of the decision tree structure provided in an embodiment of the present invention;

[0023] Figure 4 This is a schematic diagram of the structure of a telephone identification device provided in an embodiment of the present invention. Detailed Implementation

[0024] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0025] Reference Figure 1 This illustration shows a telephone identification method provided by an embodiment of the present invention, the method including:

[0026] S101. Obtain the dialing information of the calling number.

[0027] S102. Calculate the feature similarity between the dialing information and the feature information corresponding to multiple preset number categories, wherein the multiple number categories are obtained by clustering multiple sample dialing information.

[0028] S103. Based on multiple feature similarities, determine the target number category of the calling number.

[0029] S104. Label the calling number according to the target number category.

[0030] S105. Output the annotation information to display the annotation information on the terminal side corresponding to the called number.

[0031] In this embodiment of the invention, the dialing information may include the following types: calling number, called number, and call time, etc. It can be obtained when the calling number is dialed. Therefore, based on the feature similarity between the dialing information and the feature information corresponding to multiple preset number categories, which are obtained by clustering multiple sample dialing information, the target number category corresponding to the feature similarity that meets the similarity condition can be determined as the target number category of the calling number.

[0032] In one example, after determining the corresponding target number category, the calling number can be labeled. For example, the target number category, or the target number category and its corresponding attribute characteristics can be labeled to generate and output labeling information. This labeling information is then displayed on the terminal corresponding to the called number. This allows for real-time identification and display of the current calling number and improves the accuracy of telephone identification.

[0033] In one optional embodiment of the invention, when calculating feature similarity, target feature information within each number category can be used. The cluster center can be understood as the dialing information within the same number category that has the highest cumulative sum of feature similarities with other sample dialing information.

[0034] Reference Figure 2 In an optional embodiment of the invention, the method may further include the step of clustering to generate multiple number categories:

[0035] S201. Obtain multiple sample dialing information, the sample dialing information including the calling number, the called number, and the call content.

[0036] In this embodiment of the invention, multiple sample dialing information entries are obtained. These sample dialing information entries may include the following types: caller ID, called ID, and call content. The sample dialing information is determined based on fraud-related call data, and the call content is obtained through call recording with user authorization.

[0037] S202. Perform the first clustering on multiple sample dialing information to determine multiple calling categories, and the calling numbers in the same calling category are consistent.

[0038] In this embodiment of the invention, the first clustering can be understood as classifying the multiple sample dialing information entries based on the calling number. Therefore, when performing the first clustering of the first sample dialing information entry, the sample dialing information entry can be pre-determined as the first calling category. When performing the first clustering of the second sample dialing information entry, its calling number can be compared with the calling numbers in the existing calling categories, and a determination can be made based on the comparison result to determine whether to classify it into a new category. If the calling numbers of the two entries are the same, the sample dialing information entry corresponding to the calling number is classified into the existing calling category with the same calling number. If the calling numbers of the two entries are different, the sample dialing information entry corresponding to the calling number is classified into a new category. This process continues until all multiple sample dialing information entries are subjected to the first clustering, resulting in multiple calling categories. Furthermore, the calling numbers within the same calling category are all the same, and different calling categories can be distinguished using different category identifiers.

[0039] S203. Based on multiple sample dialing information in each calling category, determine the dialing discrete value of the corresponding calling number. The dialing discrete value is used to characterize the call tightness among multiple sample call information corresponding to the calling number.

[0040] In this embodiment of the invention, the dialing discrete value of the corresponding calling number can be determined based on multiple sample dialing information from each calling category, such as the called number in the sample dialing information. The dialing discrete value characterizes the degree of connection between the calling number and each called number from the same calling number. For example, the dialing discrete value can range from 0 to 1. Correspondingly, the smaller the dialing discrete value, the lower the degree of connection between the calling number and each called number, and the higher the risk that the calling number is a fraudulent call.

[0041] In an optional embodiment of the invention, step S03 may further include:

[0042] A fourth clustering was performed on multiple sample dialing information in each calling category to identify multiple called subcategories, with the called numbers in the same called subcategory being consistent.

[0043] Based on the total number of all called subcategories and the corresponding multiple sample dialing information in each called subcategory, the dialing discrete value of the corresponding calling number is determined.

[0044] In this embodiment of the invention, the fourth clustering can be understood as classifying multiple sample dialing information in each calling category based on the called number. This determines multiple called subcategories, with the called number being the same within the same called subcategory. In other words, the calling and called numbers in multiple sample dialing information within the same called subcategory are identical. Therefore, based on the total number of all called subcategories and the corresponding multiple sample dialing information in each called subcategory, statistical calculations can be performed to determine the dialing discrete value of the corresponding calling number.

[0045] In one alternative embodiment, the total number of called subcategories and the maximum number of samples included in a single called subcategory can be determined. For example, the dialing discrete value can be obtained by the following formula (1):

[0046]

[0047] In formula (1) above, Di refers to the dialing discrete value of the i-th calling number; N i (1) refers to the average number of samples corresponding to all called subcategories of the i-th calling number, which is the ratio of the total number of sample dialing information in all called subcategories to the total number of categories in all called subcategories; This refers to the maximum number of samples included in a single called subcategory corresponding to the i-th calling number. Therefore, the dialing dispersion value can be represented as the ratio between the average sample size and the maximum sample size. In other words, the closer the dialing dispersion value is to 1, the closer the average sample size is to the maximum sample size, meaning the data differences between called subcategories are not significant. This can be understood as the calling number being closely associated with different called numbers. The smaller the dialing dispersion value, the further the average sample size is from the maximum sample size, meaning the data differences between called subcategories are significant. This is equivalent to the calling number being closely associated with some called numbers and sparsely associated with others.

[0048] In another optional embodiment of the invention, the sample dialing information further includes the following types: call time and call base station identifier. The step of determining the dialing discrete value of the corresponding caller number based on multiple sample dialing information entries in each caller category may further include the following steps:

[0049] A fourth clustering was performed on multiple sample dialing information in each calling category to identify multiple called subcategories, with the called numbers in the same called subcategory being consistent.

[0050] In this embodiment of the invention, the fourth clustering can be understood as classifying multiple sample dialing information entries in each calling category based on the called number. This determines multiple called subcategories, with the called number being the same within the same called subcategory. In other words, the calling number and called number are identical in multiple sample dialing information entries within the same called subcategory.

[0051] A fifth clustering was performed on multiple sample dialing information from each calling category to identify multiple base station subcategories, with the calling base station identifier being consistent within the same base station subcategory.

[0052] In this embodiment of the invention, the fifth clustering can be understood as classifying multiple sample dialing information in each calling category based on the calling base station identifier, thereby determining multiple base station subcategories, with the calling base station identifier being consistent within the same base station subcategory. The calling base station identifier is used to distinguish different base station devices from which the calling information of the calling number originates.

[0053] The call frequency corresponding to the target category is determined. The target category includes at least the calling category, the called subcategory, and the base station subcategory.

[0054] In this embodiment of the invention, after determining multiple caller categories, multiple sample dialing messages from each caller category are arranged in order according to call time, and the time difference between two adjacent sample dialing messages within the same caller category is calculated. The average time value corresponding to all time differences is then used as the call frequency of the caller category.

[0055] After identifying multiple called subcategories, the sample dialing information in each called subcategory is arranged in order according to the call time. The time difference between two adjacent sample dialing information in the same called subcategory is calculated, and the time average in the corresponding called subcategory is determined. The minimum time average in the called subcategory is taken as the call frequency of the called subcategory.

[0056] After identifying multiple base station subcategories, the dialing information samples in each base station subcategory are arranged in order according to the call time. The time difference between two adjacent dialing information samples in the same base station subcategory is calculated, and the time mean in the corresponding base station subcategory is determined. The minimum time mean of the base station subcategory is taken as the call frequency of the base station subcategory.

[0057] Based on the total number of all called subcategories, the multiple sample dialing information corresponding to each called subcategory, the total number of all base station subcategories, the multiple sample dialing information corresponding to each base station subcategory, and the call frequency corresponding to the target category, the dialing discrete value of the corresponding calling number is determined.

[0058] In this embodiment of the invention, the dialing discrete value of the calling number can be obtained by the following formula (2):

[0059]

[0060] In the above formula (2), Di refers to the dialing discrete value of the i-th calling number; It refers to the average number of samples corresponding to all called subcategories of the i-th calling number, which is the ratio of the total number of sample dialing information in all called subcategories to the total number of categories in all called subcategories; This refers to the maximum number of samples included in a single called subcategory corresponding to the i-th calling number; N i N refers to the number of sample dialing information corresponding to the i-th calling number; i (2) refers to the average number of samples corresponding to all base station subcategories of the i-th calling number, which is the ratio of the total number of sample dialing information in all base station subcategories to the total number of categories in all base station subcategories; This refers to the maximum number of samples included in a single base station subcategory corresponding to the i-th calling number; This refers to the calling frequency of the called subcategory corresponding to the i-th calling number; This refers to the call frequency of the base station subclass corresponding to the i-th calling number; This refers to the calling frequency of the calling category corresponding to the i-th calling number.

[0061] In the above formula, It represents the proportion of the average number of samples corresponding to the same calling number and called number to the maximum number of samples. The closer the proportion is to 1, the closer the average number of samples is to the maximum number of samples. In other words, the data difference between each called sub-category is not large, which is equivalent to the caller number being evenly connected with different called numbers. It is proportional to the dialing dispersion value. This represents the proportion of call frequencies for the same calling and called numbers relative to the call frequencies for the corresponding calling category. The smaller the proportion, the closer the connection between the calling and called numbers; therefore, it is inversely proportional to the dialing dispersion value. And so on. It represents the proportion of the average number of samples corresponding to the same calling number and the same calling base station to the largest number of samples. The closer the proportion is to 1, the more likely the calling number is to dial from multiple fixed calling base stations. It is proportional to the dialing discrete value. This represents the proportion of call frequencies from the same caller ID and the same calling base station relative to the call frequencies corresponding to the caller category. The smaller the proportion, the higher the frequency of dialing through the calling base station. Therefore, it is inversely proportional to the dialing discrete value. This represents the product of the ratio of the total number of called subcategories to the total number of samples corresponding to the calling category, and the ratio of the total number of base station subcategories to the total number of samples corresponding to the calling category. The closer the product is to 1, the greater the differences between each called number and the calling base station within the same calling category, and the more discrete their distribution. This is inversely proportional to the dialing discrete value. Therefore, according to the above formula (2), the dialing discrete value can be calculated, where the dialing discrete value is used to characterize the call tightness among multiple sample call information corresponding to the calling number.

[0062] S204. Perform a second clustering on multiple sample dialing information to determine multiple called categories, with the called numbers in the same called category being consistent.

[0063] In this embodiment of the invention, the second clustering can be understood as classifying the multiple sample dialing information based on the called number. Therefore, when performing the second clustering of the first sample dialing information, the sample dialing information can be pre-determined as the first called category. When performing the second clustering of the second sample dialing information, the called number can be compared with the called numbers in the existing called categories, and a determination can be made based on the comparison result to determine whether to classify it into a new category. If the called numbers of the two are the same, the sample dialing information corresponding to the called number is classified into the existing called category with the same called number. If the called numbers of the two are different, the sample dialing information corresponding to the called number is classified into a new category. And so on, after performing the second clustering on all multiple sample dialing information, multiple called categories are determined, and the called numbers in the same called category are all the same, while different called categories can be distinguished by different category identifiers.

[0064] S205. Based on multiple sample dialing information in each called category, determine the association value between multiple calling numbers associated with the corresponding called number.

[0065] In this embodiment of the invention, considering the numerous methods employed in fraud crimes and the prevalence of professional, team-based operations—for example, online fraud can be divided into different roles such as technical support, call center promotion, and cash withdrawal—the associated calling numbers can be identified based on the aforementioned second clustering operation. The association value is used to characterize the number association between multiple calling numbers dialing the called number.

[0066] In an optional embodiment of the invention, the sample dialing information may further include the following type: call time, and step S205 may further include the following steps:

[0067] A sixth clustering was performed on multiple sample dialing information in each called category to identify multiple calling subcategories, with the calling numbers in the same calling subcategory being consistent.

[0068] For multiple sample dialing messages in each calling subcategory, arrange them in order according to the call time, and determine the start call time and the last call time in each calling subcategory.

[0069] Based on the start call time and the last call time, the association value between multiple calling numbers associated with the corresponding called number is determined.

[0070] In this embodiment of the invention, a sixth clustering is performed on multiple sample dialing messages in each called category. This sixth clustering can be understood as classifying the multiple sample dialing messages within the called category based on the calling number. This results in multiple calling subcategories, where the calling numbers within the same calling subcategory are identical. The multiple sample dialing messages in each calling subcategory are then arranged in order according to the call time, and the start and end call times for each calling subcategory are determined. Where there is only one sample dialing message in a calling subcategory, the call time in that sample dialing message can be considered either the start or end call time.

[0071] Based on the start and last call times, the association value between multiple calling numbers that call the same called number can be determined. In one example, the association value can be calculated using the following formula (3).

[0072]

[0073] In the above formula (3), Aab(k) refers to the correlation value between the a-th and b-th calling numbers corresponding to the k-th called number; ta0(k) refers to the start call time in the a-th calling subcategory corresponding to the k-th called number; tb1(k) refers to the last call time in the b-th calling subcategory corresponding to the k-th called number; δ refers to the time difference threshold, which is preset according to the fraud type of gang-related fraud in the sample dialing information, the corresponding call content and call time. Therefore, the closer the correlation value is to 1, the greater the correlation between multiple calling numbers, that is, the higher the probability of fraud in a gang-related manner.

[0074] S206. Perform semantic recognition on each call content and determine the probability of the call content belonging to different number categories.

[0075] In this embodiment of the invention, semantic recognition can be understood as extracting keywords from call content in various text formats, thereby determining whether the call content is a fraudulent call and the corresponding fraud type based on the extracted keywords. In one embodiment, the fraud type may include, but is not limited to, credit card fraud, contract fraud, loan fraud, financial fraud, and insurance fraud. Correspondingly, multiple keywords matching the corresponding fraudulent scripts can be preset for different fraud types, and these keywords can be weighted. Thus, when a corresponding keyword is extracted, a weighting operation is performed on the keyword. Finally, after performing semantic recognition on all call content, the weighted probability values ​​corresponding to all keywords are accumulated to obtain the category probability of the corresponding call content belonging to different number categories.

[0076] S207. Based on the dialing discrete value, correlation value, and category probability value, perform a third clustering on all sample dialing information to determine multiple number categories.

[0077] In this embodiment of the invention, a third clustering can be performed on all sample dialing information to generate a decision tree, which can then be referenced. Figure 3 As shown, the decision tree may include internal nodes (node ​​boxes are rounded boxes), leaf nodes (node ​​boxes are rectangles), and directed edges. The internal nodes represent the classification features, the leaf nodes represent the number categories, and the directed edges represent the category partitioning rules. These rules are determined based on the dialing discrete value and the category probability value. In one example, the category partitioning rule may be that the dialing discrete value is greater than or equal to a first threshold, or that the category probability value is greater than or equal to a second threshold, etc., without further limitation.

[0078] Then, after binary classification, the Gini coefficient of the decision tree is calculated. The Gini coefficient can be obtained according to the following formula (4):

[0079]

[0080]

[0081]

[0082] In the above formulas (3), (4) and (5), K refers to the number of pre-set number categories, and k refers to the kth number category, which can be used to distinguish different number categories; and These refer to grouping based on classification features and the proportion of samples in each group to the total number of samples. This refers to the proportion of the first sample group D1 that belongs to the k-th class; This refers to the proportion of samples D2 belonging to the k-th class. Therefore, the first and second thresholds (which can also be understood as feature thresholds set for classification features) can be updated to find the optimal feature threshold. This ensures that the Gini coefficient meets a preset partitioning condition. The preset partitioning condition can be that the Gini coefficient reaches its minimum; when the Gini coefficient reaches its minimum, the corresponding feature threshold is the optimal feature threshold.

[0083] If the Gini coefficient meets the preset division conditions, new internal nodes and leaf nodes are divided according to the new classification features. During this process, the Gini coefficient is calculated and the corresponding optimal feature threshold is determined repeatedly until all sample dialing information is assigned to the leaf nodes belonging to their true categories, thus obtaining the target decision tree.

[0084] In one embodiment, after obtaining the target decision tree, the sample dialing information located in different leaf nodes of the target decision tree can be categorized based on the association value to determine multiple number categories of the target decision tree. For example, a preset association threshold is set, and the association value between each calling number is obtained. If the corresponding association value is greater than or equal to the association threshold, it is determined whether the associated calling numbers belong to the same category. If they are in the same number category, no adjustment is made. If they are not in the same number category, adjustment is made based on the dialing dispersion of the two calling numbers. That is, the sample dialing information corresponding to the calling number with higher dialing dispersion is classified into the number category corresponding to the calling number with lower dispersion. If the dialing dispersion of the two calling numbers is the same, no adjustment operation is performed. Thus, by using number association, calling numbers with high fraud probability and small dispersion values ​​can be clustered into one category. This ensures the accuracy and real-time nature of fraud call identification, enabling timely alerts to users and protecting their rights.

[0085] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of this application.

[0086] Reference Figure 4 This illustration shows a telephone identification device provided by an embodiment of the present invention, the device may include:

[0087] The information acquisition module 401 is used to acquire the dialing information of the calling number.

[0088] The similarity determination module 402 is used to calculate the feature similarity between the dialing information and the feature information corresponding to multiple preset number categories, wherein the multiple number categories are obtained by clustering multiple sample dialing information.

[0089] The category determination module 403 is used to determine the target number category corresponding to the calling number based on multiple feature similarities.

[0090] The number labeling module 404 is used to label the calling number according to the target number category.

[0091] The information output module 405 is used to output the annotation information so as to display the annotation information on the terminal side corresponding to the called number.

[0092] In an optional embodiment of the invention, the category determination module 403 may also be used for:

[0093] If one of the multiple feature similarities meets the similarity condition, the number category corresponding to the feature similarity is determined as the target number category corresponding to the calling number.

[0094] In an optional embodiment of the invention, the similarity determination module 402 may include:

[0095] The similarity between the dialing information and target feature information in multiple preset number categories is calculated, wherein the target feature information is the cluster center of the number category.

[0096] In an optional embodiment of the invention, the apparatus further includes a module for clustering to generate multiple number categories:

[0097] The sample acquisition module is used to acquire multiple sample dialing information, which includes the calling number, the called number, and the call content.

[0098] The caller category determination module is used to perform the first clustering on multiple sample dialing information to determine multiple caller categories, and the caller numbers in the same caller category are consistent.

[0099] The dialing discrete value determination module is used to determine the dialing discrete value of the corresponding calling number based on multiple sample dialing information in each calling category. The dialing discrete value is used to characterize the call tightness among multiple sample call information corresponding to the calling number.

[0100] The called party category determination module is used to perform a second clustering on multiple sample dialing information to determine multiple called party categories, and the called party numbers in the same called party category are consistent.

[0101] The association value determination module is used to determine the association value between multiple calling numbers associated with the corresponding called number based on multiple sample dialing information in each called category.

[0102] The probability determination module is used to perform semantic recognition on each call content and determine the category probability of the call content belonging to different number categories.

[0103] The number category determination module is used to perform a third clustering on all sample dialing information based on the dialing discrete value, correlation value, and category probability value to determine multiple number categories.

[0104] In one optional embodiment of the invention, the dialing discrete value determination module may include:

[0105] The called party sub-category determination submodule is used to perform a fourth clustering on multiple sample dialing information in each calling party category to determine multiple called party sub-categories, and the called party numbers in the same called party sub-category are consistent.

[0106] The dialing discrete value determination submodule is used to determine the dialing discrete value of the corresponding calling number based on the total number of all called subcategories and the corresponding multiple sample dialing information in each called subcategory.

[0107] In an optional embodiment of the invention, the sample dialing information further includes the following types: call time and call base station identifier, and the dialing discrete value determination module may further include:

[0108] The called sub-category determination submodule is also used to perform a fourth clustering on multiple sample dialing information in each calling category to determine multiple called sub-categories, with the called numbers in the same called sub-category being consistent.

[0109] The base station sub-category determination submodule is used to perform a fifth clustering on multiple sample dialing information in each calling category to determine multiple base station sub-categories. Calling base station identifiers are consistent within the same base station sub-category.

[0110] The call frequency determination submodule is used to determine the call frequency corresponding to the target category. The target category includes at least the calling category, the called subcategory, and the base station subcategory.

[0111] The dialing discrete value determination submodule is further used to determine the dialing discrete value of the corresponding calling number based on the total number of all called subcategories, the multiple sample dialing information corresponding to each called subcategory, the total number of all base station subcategories, the multiple sample dialing information corresponding to each base station subcategory, and the call frequency corresponding to the target category.

[0112] In an optional embodiment of the invention, the association value determination module may include:

[0113] The calling subcategory determination submodule is used to perform a sixth clustering on multiple sample dialing information in each called category to determine multiple calling subcategories, with the calling numbers in the same calling subcategory being consistent.

[0114] The time determination submodule is used to arrange multiple sample dialing information in each calling subcategory in an orderly manner according to the call time, and determine the start call time and the last call time in each calling subcategory.

[0115] The association value determination submodule is used to determine the association value between multiple calling numbers associated with the corresponding called number based on the start call time and the last call time.

[0116] In an optional embodiment of the invention, the number category determination module may include:

[0117] The decision tree generation submodule is used to perform a third clustering on all sample dialing information to generate a decision tree. The decision tree includes internal nodes, leaf nodes, and directed edges. The internal nodes are used to represent the classification features, the leaf nodes are used to represent the number categories, and the directed edges are used to represent the category partitioning rules. The category partitioning rules are determined based on the dialing discrete value and the category probability value.

[0118] The coefficient calculation submodule is used to calculate the Gini coefficient of the decision tree, and determine the target decision tree when the Gini coefficient meets the preset partitioning conditions.

[0119] The number category determination submodule is used to perform category association on sample dialing information located in different leaf nodes of the target decision tree based on the association value, and determine multiple number categories of the target decision tree.

[0120] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0121] It will be readily apparent to those skilled in the art that any combination of the above embodiments is feasible, and therefore any combination of the above embodiments is an implementation scheme of the present invention. However, due to space limitations, this specification will not describe them in detail here.

[0122] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0123] Similarly, it should be understood that, in order to simplify the invention and aid in understanding one or more of the various inventive aspects, features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the above description of exemplary embodiments of the invention. However, this disclosure should not be construed as reflecting an intention that the claimed invention requires more features than expressly recited in each claim. Rather, as reflected in the claims, inventive aspects lie in fewer than all features of the single foregoing disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0124] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0125] An electronic device, comprising:

[0126] One or more processors;

[0127] Memory;

[0128] One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to perform the methods described in the above embodiments.

[0129] A computer-readable storage medium stores a computer program for use in conjunction with an electronic device, the computer program being executable by a processor to perform the methods described in the embodiments above.

[0130] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0131] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0132] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0133] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0134] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0135] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0136] The present invention has provided a detailed description of a telephone identification method and a telephone identification device. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A telephone identification method, characterized in that, The method includes: Obtain the dialing information of the calling number; Calculate the feature similarity between the dialing information and the feature information corresponding to multiple preset number categories, wherein the multiple number categories are obtained by clustering multiple sample dialing information; Based on multiple feature similarities, the target number category corresponding to the calling number is determined; The calling numbers are labeled according to the target number category; Output annotation information to display the annotation information on the terminal side corresponding to the called number; The method also includes a step of clustering to generate multiple number categories: Obtain multiple sample dialing information entries, including the calling number, the called number, and the call content; The first clustering of multiple sample dialing information is performed to determine multiple calling categories, and the calling numbers in the same calling category are consistent. Based on multiple sample dialing information in each calling category, the dialing discrete value of the corresponding calling number is determined. The dialing discrete value is used to characterize the call tightness among multiple sample call information corresponding to the calling number. A second clustering was performed on multiple sample dialing information to determine multiple called party categories, and the called party numbers in the same called party category were consistent. Based on multiple sample dialing information in each called party category, determine the association value between multiple calling numbers associated with the corresponding called party number; Semantic recognition is performed on each call content, and the probability of the call content belonging to different number categories is determined; Based on the dialing discrete value, correlation value, and category probability value, a third clustering is performed on all sample dialing information to determine multiple number categories; Based on the dialing discrete value, correlation value, and category probability value, a third clustering is performed on all sample dialing information to determine multiple number categories, including: A third clustering is performed on all sample dialing information to generate a decision tree. The decision tree includes internal nodes, leaf nodes, and directed edges. The internal nodes are used to represent classification features, the leaf nodes are used to represent number categories, and the directed edges are used to represent category partitioning rules. The category partitioning rules are determined based on the dialing discrete value and the category probability value. Calculate the Gini coefficient of the decision tree, and determine the target decision tree if the Gini coefficient meets the preset splitting conditions; Based on the association value, the dialing information of samples located in different leaf nodes of the target decision tree is classified and associated to determine multiple number categories of the target decision tree.

2. The telephone identification method according to claim 1, characterized in that, The determination of the target number category corresponding to the calling number based on multiple feature similarities includes: If one of the multiple feature similarities meets the similarity condition, the number category corresponding to the feature similarity is determined as the target number category corresponding to the calling number.

3. The telephone identification method according to claim 2, characterized in that, The step of calculating the feature similarity between the dialing information and feature information corresponding to multiple preset number categories includes: The similarity between the dialing information and target feature information in multiple preset number categories is calculated, wherein the target feature information is the cluster center of the number category.

4. The telephone identification method according to claim 3, characterized in that, The step of determining the dialing discrete value of the corresponding calling number based on multiple sample dialing information from each calling category includes: A fourth clustering was performed on multiple sample dialing information in each calling category to determine multiple called subcategories, and the called numbers in the same called subcategory were consistent. Based on the total number of all called subcategories and the corresponding multiple sample dialing information in each called subcategory, the dialing discrete value of the corresponding calling number is determined.

5. The telephone identification method according to claim 3, characterized in that, The sample dialing information also includes the following types: Call time and call base station identifier, the step of determining the dialing discrete value of the corresponding caller number based on multiple sample dialing information in each caller category includes: A fourth clustering was performed on multiple sample dialing information in each calling category to determine multiple called subcategories, and the called numbers in the same called subcategory were consistent. A fifth clustering was performed on multiple sample dialing information in each calling category to determine multiple base station subcategories. The calling base station identifiers in the same base station subcategory are consistent. The call frequency corresponding to the target category is determined, and the target category includes at least: the calling category, the called subcategory, and the base station subcategory; Based on the total number of all called subcategories, the multiple sample dialing information corresponding to each called subcategory, the total number of all base station subcategories, the multiple sample dialing information corresponding to each base station subcategory, and the call frequency corresponding to the target category, the dialing discrete value of the corresponding calling number is determined.

6. The telephone identification method according to claim 5, characterized in that, The step of determining the association value between multiple calling numbers associated with the corresponding called number based on multiple sample dialing information in each called category includes: A sixth clustering was performed on multiple sample dialing information in each called category to determine multiple calling subcategories, with the calling numbers in the same calling subcategory being consistent; For multiple sample dialing messages in each calling subcategory, arrange them in order according to the call time, and determine the start call time and the last call time in each calling subcategory; Based on the start call time and the last call time, the association value between multiple calling numbers associated with the corresponding called number is determined.

7. An electronic device, comprising: One or more processors; Memory; One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to perform the method of any one of claims 1-5.

8. A computer-readable storage medium storing a computer program for use in conjunction with an electronic device, said computer program being executable by a processor to perform the method of any one of claims 1-5.

Images