Bad phone number identification method, device, storage medium and computer equipment

Abstract

The embodiment of the present invention provides a method, device, storage medium and computer equipment for identifying bad phone numbers. By parsing the generated media stream information, multiple keyword information and part-of-speech information corresponding to each keyword information are generated; calculations are performed based on the multiple keyword information and the multiple part-of-speech information to generate a significant intensity coefficient variance; it is determined whether the significant intensity coefficient variance is located in a closed interval; if it is determined that the significant intensity coefficient variance is located in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number. In the embodiment of the present invention, by calculating the acquired media signaling message to generate a significant intensity coefficient variance, if it is determined that the significant intensity coefficient variance is located in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number, thereby improving the accuracy of identifying bad phone numbers.

Description

[Technical field]

[0001] The present invention relates to the field of security technology, and in particular to a method, device, storage medium and computer equipment for identifying bad telephone numbers. [Background technology]

[0002] In the related technologies, the main way to manage bad calls is for telecom operators to collect massive call records on the network side, and combine big data risk mining technology to perform feature analysis on high-frequency, high-discrete calls to identify various types of bad phone numbers. However, in recent years, various types of bad phone numbers have added a large number of caller numbers to evade supervision, and implemented widespread telephone harassment by controlling the frequency of outbound calls, the number of outbound numbers, and the duration of outbound calls of the caller numbers, which has led to a decline in the control effect of the technical solutions in the related technologies and reduced the accuracy of identifying bad phone numbers. [Summary of the invention]

[0003] In view of this, embodiments of the present invention provide a method, apparatus, storage medium and computer device for identifying bad telephone numbers, so as to improve the accuracy of identifying bad telephone numbers.

[0004] On the one hand, an embodiment of the present invention provides a method for identifying bad phone numbers, which is applied to a working honeypot server in a honeypot pool. The method includes:

[0005] Parsing the generated media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each of the keyword information;

[0006] Calculate based on the plurality of keyword information and the plurality of part-of-speech information to generate a significant intensity coefficient variance;

[0007] Determine whether the significant intensity coefficient variance is in a closed interval, where the closed interval includes a numerical range between a first set threshold and a second set threshold;

[0008] If it is determined that the significant intensity coefficient variance is located in the closed interval, the calling number corresponding to the media signaling message is set as a bad phone number.

[0009] Optionally, before parsing the acquired media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information, the method includes:

[0010] Receiving a media signaling message sent by a distribution unit in the honeypot pool, wherein the media signaling message is sent by a terminal device to the distribution unit;

[0011] Extracting multiple media content information of the calling direction from the media signaling message;

[0012] Arrange the plurality of media content information in chronological order to generate media stream information.

[0013] Optionally, the parsing the media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information includes:

[0014] Calculating the media stream information through speech recognition technology to generate media text information;

[0015] The media text information is parsed by word segmentation technology to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information.

[0016] Optionally, the calculating based on the plurality of keyword information and the plurality of part-of-speech information to generate a significant intensity coefficient variance includes:

[0017] Calculate the multiple keyword information by using a text ranking algorithm to generate multiple sensitive word strength coefficients;

[0018] Calculating the plurality of part-of-speech information by an expert algorithm to generate a plurality of part-of-speech strength coefficients;

[0019] Generating a plurality of significant intensity coefficients according to the plurality of sensitive word intensity coefficients and the plurality of part-of-speech intensity coefficients;

[0020] Calculating an average value of a plurality of the significant intensity coefficients to generate an average significant intensity coefficient;

[0021] A significant intensity coefficient variance is generated based on the plurality of significant intensity coefficients and the average significant intensity coefficient.

[0022] Optionally, it also includes:

[0023] If it is determined that the significant intensity coefficient variance is not located in the closed interval, the calling number corresponding to the media signaling message is set as a good telephone number.

[0024] Optionally, the step of setting the calling number corresponding to the media signaling message as a bad phone number includes:

[0025] Stop receiving a media signaling message sent by a distribution unit in the honeypot pool, where the media signaling message is sent by a terminal device;

[0026] A blocking instruction is sent to the server so that the server can block the terminal device corresponding to the media signaling message.

[0027] Optionally, generating a plurality of significant intensity coefficients according to a plurality of sensitive word intensity coefficients and a plurality of part-of-speech intensity coefficients comprises:

[0028] Calculate the multiple sensitive word strength coefficients and the multiple part-of-speech strength coefficients by formula θ=α*β to generate multiple significant strength coefficients, where θ is the significant strength coefficient, α is the part-of-speech strength coefficient, and β is the sensitive word strength coefficient;

[0029] The generating a significant intensity coefficient variance according to the plurality of significant intensity coefficients and the average significant intensity coefficient comprises:

[0030] By formula The significant intensity coefficients and the average significant intensity coefficient are calculated to generate the significant intensity coefficient variance, where θ i is the significant intensity coefficient, The average significant intensity coefficient, n is the total number of the significant intensity coefficients.

[0031] On the other hand, an embodiment of the present invention provides a device for identifying bad phone numbers, including:

[0032] A first generating module is used to parse the acquired media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each of the keyword information;

[0033] A second generating module, configured to generate a significant intensity coefficient variance by performing calculations based on the plurality of keyword information and the plurality of part-of-speech information;

[0034] A judging module, used to judge whether the significant intensity coefficient variance is in a closed interval, wherein the closed interval includes a numerical range between a first set threshold and a second set threshold;

[0035] The first setting module is configured to set the calling number corresponding to the media signaling message as a bad phone number if the judging module judges that the significant intensity coefficient variance is located in the closed interval.

[0036] On the other hand, an embodiment of the present invention provides a storage medium, including: the storage medium includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the above-mentioned bad telephone number identification method.

[0037] On the other hand, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions, and is characterized in that the program instructions implement the steps of the above-mentioned bad telephone number identification method when loaded and executed by the processor.

[0038] In a technical solution of a method for identifying bad phone numbers provided in an embodiment of the present invention, multiple keyword information and part-of-speech information corresponding to each keyword information are generated by parsing the generated media stream information; a significant intensity coefficient variance is generated based on the multiple keyword information and the multiple part-of-speech information; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number. In an embodiment of the present invention, a significant intensity coefficient variance is generated by calculating the acquired media signaling message; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number, thereby improving the accuracy of identifying bad phone numbers.

Brief Description of the Drawings

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 A schematic diagram of the structure of a bad telephone number identification system provided by an embodiment of the present invention;

[0041] Figure 2 A flowchart of a method for identifying bad phone numbers provided by an embodiment of the present invention;

[0042] Figure 3 A flowchart of another bad phone number identification method provided by an embodiment of the present invention;

[0043] Figure 4 for Figure 3 The working honeypot server in the middle parses the generated media stream information to generate a flowchart of multiple keyword information and part-of-speech information corresponding to each keyword information;

[0044] Figure 5 for Figure 3 The working honeypot server in the middle calculates based on multiple keyword information and multiple part-of-speech information to generate a flow chart of the significant intensity coefficient variance;

[0045] Figure 6 A schematic diagram of a multi-domain feature library organization provided by an embodiment of the present invention;

[0046] Figure 7 A schematic diagram of the structure of a bad telephone number identification device provided by an embodiment of the present invention;

[0047] Figure 8 A schematic diagram of a computer device provided by an embodiment of the present invention. [Specific implementation method]

[0048] In order to better understand the technical solution of the present invention, the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0049] It should be clear that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0051] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0052] In a related technology, a single caller number is limited by obtaining the scope of the call content and the scope of the risk warning object, so that the scope of identifying bad phone numbers is limited, reducing the accuracy of identifying bad phone numbers. In addition, the technical solution in the related technology also mainly calculates the call content keywords, semantics, intonation and other features around a single caller number, and cannot identify bad phone numbers for massive calls, reducing the accuracy of identifying bad phone numbers. There are harassment risks that constitute high-frequency outbound calls in many fields, such as real estate, financial management, loans, fraud, education and training, etc., and the related technical methods cannot accurately identify bad phone numbers in multiple different fields.

[0053] In another related technology, a call behavior feature analysis method is used to predict bad phone numbers by applying big data analysis and mining technology to massive call record data. However, this method fails to meet the specified thresholds in terms of outbound call frequency, outbound call discreteness, etc., which may result in missing bad calls and reduce the accuracy of identifying bad phone numbers.

[0054] In another related technology, various types of bad phone numbers marked and reported by users are collected through an Internet security APP. The marking results in the related technology are prone to marking errors, which can easily lead to the situation where bad calls are mistakenly blocked, reducing the accuracy of identifying bad phone numbers.

[0055] In order to solve the technical problems in the related art, an embodiment of the present invention provides a bad telephone number identification system. Figure 1 A schematic diagram of a bad phone number identification system provided by an embodiment of the present invention is shown in FIG. Figure 1 As shown, the system includes: a server 1 and a terminal device 2.

[0056] In the embodiment of the present invention, the server 1 includes a honeypot pool.

[0057] In the embodiment of the present invention, the honeypot pool includes a distribution unit 11 and at least one honeypot server 12. As an optional solution, the honeypot pool 1 includes multiple honeypot servers 12.

[0058] In the embodiment of the present invention, the distribution unit 11 is connected to each honeypot server 12 .

[0059] In the embodiment of the present invention, the terminal device 2 includes: a mobile phone, a car computer, a tablet computer, a computer or a wearable device.

[0060] In the embodiment of the present invention, the server 1 is connected to the terminal device 2. As an optional solution, the server 1 is connected to the terminal device 2 via mobile communication technology. The mobile communication technology includes the third generation mobile communication technology (3rd-Generation, referred to as 3G), the fourth generation mobile communication technology (the 4th generation mobile communication technology, referred to as 4G) or the fifth generation mobile communication technology (the 5th generation mobile communication technology, referred to as 5G).

[0061] In an embodiment of the present invention, the terminal device 2 sends a media signaling message to the distribution unit 11 in the honeypot pool. Each honeypot server 12 in the honeypot pool has two states: a working state and an idle state. By default, the initial state of each honeypot server 12 is an idle state. When the honeypot server 12 receives the media signaling message, the distribution unit 11 selects an idle honeypot server 12 from multiple honeypot servers 12, and the state of the honeypot server 12 changes from the idle state to the working state, and the idle honeypot server 12 serves as the working honeypot server. The distribution unit 11 sends a media signaling message to the working honeypot server. When the working honeypot server finishes working, the state of the working honeypot server changes from the working state to the idle state, and the working honeypot server serves as the honeypot server 12.

[0062] In an embodiment of the present invention, a working honeypot server is used to parse the generated media stream information, generate multiple keyword information and part-of-speech information corresponding to each keyword information; perform calculations based on the multiple keyword information and the multiple part-of-speech information to generate a significant intensity coefficient variance; determine whether the significant intensity coefficient variance is located in a closed interval, the closed interval includes a numerical range between a first set threshold and a second set threshold, and the first set threshold and the second set threshold are endpoints of the closed interval, and the first set threshold is less than the second set threshold; if it is determined that the significant intensity coefficient variance is located in the closed interval, the calling number corresponding to the media signaling message is set as a bad phone number.

[0063] In an embodiment of the present invention, the working honeypot server is also used to receive a media signaling message sent by the distribution unit 11 in the honeypot pool, where the media signaling message is sent by the terminal device 2 to the distribution unit 11; extract multiple media content information of the calling direction in the media signaling message; arrange the multiple media content information in chronological order to generate media stream information.

[0064] In the embodiment of the present invention, the working honeypot server is specifically used to calculate the media stream information through speech recognition technology to generate media text information; and to parse the media text information through word segmentation technology to generate multiple keyword information and part-of-speech information corresponding to each keyword information.

[0065] In an embodiment of the present invention, the working honeypot server is specifically used to calculate multiple keyword information through a text ranking algorithm to generate multiple sensitive word strength coefficients; calculate multiple part-of-speech information through an expert algorithm to generate multiple part-of-speech strength coefficients; generate multiple significant strength coefficients based on multiple sensitive word strength coefficients and multiple part-of-speech strength coefficients; calculate the average of multiple significant strength coefficients to generate an average significant strength coefficient; generate a significant strength coefficient variance based on multiple significant strength coefficients and the average significant strength coefficient.

[0066] In the embodiment of the present invention, the working honeypot server is further configured to set the calling number corresponding to the media signaling message as a good telephone number if it is determined that the significant intensity coefficient variance is not located in a closed interval.

[0067] In an embodiment of the present invention, the working honeypot server is also used to stop receiving media signaling messages sent by the distribution unit 11 in the honeypot pool 1, where the media signaling messages are sent by the terminal device 2 to the distribution unit 11; and send a blocking instruction to the server 1 so that the server 1 can block the terminal device 2 corresponding to the media signaling message.

[0068] In the embodiment of the present invention, the working honeypot server is specifically used to calculate multiple sensitive word strength coefficients and multiple part-of-speech strength coefficients through the formula θ=α*β to generate multiple significant strength coefficients, where θ is the significant strength coefficient, α is the part-of-speech strength coefficient, and β is the sensitive word strength coefficient. Multiple significant intensity coefficients and average significant intensity coefficients are calculated to generate significant intensity coefficient variance, where θ i is the significant strength coefficient, is the average significant intensity coefficient, and n is the total number of significant intensity coefficients.

[0069] In a technical solution of a method for identifying bad phone numbers provided in an embodiment of the present invention, multiple keyword information and part-of-speech information corresponding to each keyword information are generated by parsing the generated media stream information; a significant intensity coefficient variance is generated based on the multiple keyword information and the multiple part-of-speech information; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number. In an embodiment of the present invention, a significant intensity coefficient variance is generated by calculating the acquired media signaling message; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number, thereby improving the accuracy of identifying bad phone numbers.

[0070] The embodiment of the present invention provides a method for identifying bad telephone numbers. Figure 2 A flowchart of a method for identifying bad phone numbers provided by an embodiment of the present invention is shown in FIG. Figure 2 As shown, the method includes:

[0071] Step 101: parse the generated media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information.

[0072] In the embodiment of the present invention, each step is executed by a working honeypot server.

[0073] Specifically, the media stream information is calculated by using speech recognition technology to generate media text information; the media text information is parsed by using word segmentation technology to generate multiple keyword information and part-of-speech information corresponding to each keyword information.

[0074] In the embodiment of the present invention, the speech recognition technology includes automatic speech recognition technology (Automatic Speech Recognition, referred to as ASR).

[0075] In the embodiment of the present invention, the word segmentation technology includes the Jieba word segmentation technology.

[0076] As an optional solution, the media text information is parsed through word segmentation technology to generate a set of multiple keyword information and the part-of-speech information corresponding to each keyword information. The set is written as P = {(W0, C0), (W1, C1), ... (Wi, Ci)}, where Wi is the keyword information and Ci is the part-of-speech information corresponding to the keyword information.

[0077] Step 102: Calculate based on multiple keyword information and multiple part-of-speech information to generate a significant intensity coefficient variance.

[0078] Step 103: determine whether the significant intensity coefficient variance is in a closed interval, where the closed interval includes a numerical range between a first set threshold and a second set threshold.

[0079] Specifically, it is determined whether the significant intensity coefficient variance is greater than or equal to a first set threshold and less than or equal to a second set threshold.

[0080] In the embodiment of the present invention, the first set threshold and the second set threshold are endpoints of a closed interval, and the first set threshold is smaller than the second set threshold.

[0081] In an embodiment of the present invention, if it is determined that the significant intensity coefficient variance is located in a closed interval, it indicates that the significant intensity coefficient variance is greater than or equal to the first set threshold and less than or equal to the second set threshold, and the calling number corresponding to the significant intensity coefficient variance is a bad telephone number; if it is determined that the significant intensity coefficient variance is not located in a closed interval, it indicates that the significant intensity coefficient variance is less than the first set threshold and greater than the second set threshold, and the calling number corresponding to the significant intensity coefficient variance is a good telephone number.

[0082] In the embodiment of the present invention, the bad telephone number includes a nuisance telephone number, a fraudulent telephone number or a toll-charging telephone number.

[0083] Step 104: If it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number.

[0084] In a technical solution of a method for identifying bad phone numbers provided in an embodiment of the present invention, multiple keyword information and part-of-speech information corresponding to each keyword information are generated by parsing the generated media stream information; a significant intensity coefficient variance is generated based on the multiple keyword information and the multiple part-of-speech information; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number. In an embodiment of the present invention, a significant intensity coefficient variance is generated by calculating the acquired media signaling message; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number, thereby improving the accuracy of identifying bad phone numbers.

[0085] The embodiment of the present invention provides another method for identifying bad telephone numbers. Figure 3 A flowchart of another bad phone number identification method provided by an embodiment of the present invention, such as Figure 3 As shown, the method includes:

[0086] Step 201: The terminal device sends a media signaling message to a distribution unit in the honeypot pool.

[0087] Step 202: The working honeypot server receives a media signaling message sent by a distribution unit in the honeypot pool.

[0088] In the embodiment of the present invention, the distribution unit sends the media signaling message to the idle honeypot server through the elastic contraction distribution management technology.

[0089] Step 203: The working honeypot server extracts multiple media content information of the calling direction from the media signaling message.

[0090] In the embodiment of the present invention, the media signaling message includes a sending and receiving flag, and the sending and receiving flag is used to indicate a calling direction or a called direction of the media signaling message.

[0091] In the embodiment of the present invention, specifically, the working honeypot server extracts multiple media content information of the calling direction in the media signaling message according to the sending and receiving identifiers in the media signaling message.

[0092] Step 204: The working honeypot server arranges the multiple media content information in chronological order to generate media stream information.

[0093] In the embodiment of the present invention, the time sequence includes a time sequence from first to last or a time sequence from last to first.

[0094] As an optional solution, the working honeypot server arranges multiple media content information in a time-ordered order to generate media stream information.

[0095] Step 205: The working honeypot server parses the generated media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information.

[0096] In the embodiment of the present invention, the keyword information includes keywords, such as: "buy", "suitable" or "product".

[0097] In the embodiment of the present invention, the part-of-speech information includes part-of-speech, such as verb, adjective or noun.

[0098] For example, the part-of-speech information corresponding to "purchase" is a verb, the part-of-speech information corresponding to "suitable" is an adjective, and the part-of-speech information corresponding to "product" is a noun.

[0099] In the embodiment of the present invention, Figure 4 for Figure 3 The working honeypot server parses the generated media stream information and generates a flowchart of multiple keyword information and part-of-speech information corresponding to each keyword information, as shown in FIG. Figure 4 As shown, step 205 includes:

[0100] Step 2051: The working honeypot server calculates the media stream information through speech recognition technology to generate media text information.

[0101] In the embodiment of the present invention, the speech recognition technology includes automatic speech recognition technology (Automatic Speech Recognition, referred to as ASR).

[0102] Step 2052: The working honeypot server parses the media text information through word segmentation technology to generate multiple keyword information and part-of-speech information corresponding to each keyword information.

[0103] In the embodiment of the present invention, the word segmentation technology includes the Jieba word segmentation technology.

[0104] As an optional solution, the working honeypot server parses the media text information through word segmentation technology to generate a set of multiple keyword information and the part-of-speech information corresponding to each keyword information. The set is written as P = {(W0, C0), (W1, C1), ... (Wi, Ci)}, where Wi is the keyword information and Ci is the part-of-speech information corresponding to the keyword information.

[0105] Step 206: The working honeypot server performs calculations based on the multiple keyword information and the multiple part-of-speech information to generate a significant intensity coefficient variance.

[0106] In the embodiment of the present invention, Figure 5 for Figure 3 The working honeypot server calculates based on multiple keyword information and multiple part-of-speech information to generate a flowchart of the significant intensity coefficient variance, as shown in Figure 5 As shown, step 206 includes:

[0107] Step 2061: The working honeypot server calculates multiple keyword information using a text ranking algorithm to generate multiple sensitive word strength coefficients.

[0108] In the embodiment of the present invention, specifically, the working honeypot server calculates multiple keyword information through a text ranking algorithm to generate a sensitive word strength coefficient corresponding to each keyword information. The keyword information corresponds to the sensitive word strength coefficient one by one.

[0109] Step 2062: The working honeypot server calculates multiple part-of-speech information using an expert algorithm to generate multiple part-of-speech strength coefficients.

[0110] In the embodiment of the present invention, specifically, the working honeypot server calculates multiple part-of-speech information through an expert algorithm to generate a part-of-speech strength coefficient corresponding to each part-of-speech information. The part-of-speech information corresponds to the part-of-speech strength coefficient one by one.

[0111] Step 2063: The working honeypot server generates a plurality of significant strength coefficients according to a plurality of sensitive word strength coefficients and a plurality of part-of-speech strength coefficients.

[0112] Specifically, the working honeypot server calculates multiple sensitive word strength coefficients and multiple part-of-speech strength coefficients through the formula θ=α*β to generate multiple significant strength coefficients, where θ is the significant strength coefficient, α is the part-of-speech strength coefficient, and β is the sensitive word strength coefficient.

[0113] As an optional solution, the embodiment of the present invention provides a multi-domain feature library organization. Figure 6 A schematic diagram of a multi-domain feature library organization provided by an embodiment of the present invention is shown in FIG. Figure 6 As shown, the multi-domain feature library organization includes multiple domains, for example: domain A, domain B...domain N. Each domain includes multiple matrices, for example: domain A includes matrix F1, matrix F2...matrix Fx. Each matrix includes multiple keyword information, part-of-speech information corresponding to each keyword information, and a significant intensity coefficient calculated based on each keyword information and its corresponding part-of-speech information, for example: matrix F1 includes {(W1, C1, θ1), (W2, C2, θ2)...(Wn, Cn, θn)}, where Wn is keyword information, Cn is part-of-speech information, and θn is a significant intensity coefficient.

[0114] In an embodiment of the present invention, as an optional scheme, keyword information, part-of-speech information and significant intensity coefficient with the same sensitive word attributes constitute a row. For example, the keyword information W1, part-of-speech information C1 and significant intensity coefficient θ1 in the first row (W1, C1, θ1) in the matrix F1 have the same sensitive word attributes.

[0115] In an embodiment of the present invention, as an optional solution, the working honeypot server can determine a set of sensitive word attributes in a certain field as a feature cluster of the field, for example: F1 matrix, F2 matrix...Fx matrix in field A are the feature clusters of field A.

[0116] Step 2064: The working honeypot server calculates the average value of multiple significant intensity coefficients to generate an average significant intensity coefficient.

[0117] In the embodiment of the present invention, as an optional solution, the working honeypot server can calculate the average value of multiple significant intensity coefficients corresponding to each feature cluster to generate the average significant intensity coefficient corresponding to each feature cluster.

[0118] Step 2065: The working honeypot server generates a significant intensity coefficient variance according to the multiple significant intensity coefficients and the average significant intensity coefficient.

[0119] Specifically, the working honeypot server is calculated by formula Multiple significant intensity coefficients and average significant intensity coefficients are calculated to generate significant intensity coefficient variance, where θ i is the significant strength coefficient, is the average significant intensity coefficient, and n is the total number of significant intensity coefficients.

[0120] In the embodiment of the present invention, as an optional solution, the working honeypot server can generate the significant intensity coefficient variance corresponding to each feature cluster according to multiple significant intensity coefficients corresponding to each feature cluster and the average significant intensity coefficient corresponding to each feature cluster.

[0121] Step 207 , the working honeypot server determines whether the significant intensity coefficient variance is in a closed interval, where the closed interval includes a numerical range between a first set threshold and a second set threshold. If so, execute step 209 ; if not, execute step 208 .

[0122] Specifically, it is determined whether the significant intensity coefficient variance is greater than or equal to a first set threshold and less than or equal to a second set threshold.

[0123] In the embodiment of the present invention, the first set threshold is smaller than the second set threshold, and the first set threshold and the second set threshold are endpoints of a closed interval.

[0124] In the embodiment of the present invention, as an optional solution, the working honeypot server determines whether the variance of the significant intensity coefficient corresponding to each feature cluster is located in a closed interval.

[0125] In an embodiment of the present invention, as an optional scheme, if the working honeypot server determines that the significant intensity coefficient variance corresponding to each feature cluster is located in a closed interval, it indicates that the media signaling message corresponding to the significant intensity coefficient variance hits the feature cluster, and the calling number corresponding to the media signaling message is a bad phone number; if the working honeypot server determines that the significant intensity coefficient variance corresponding to each feature cluster is not located in a closed interval, it indicates that the media signaling message corresponding to the significant intensity coefficient variance hits the feature cluster, and the calling number corresponding to the media signaling message is a good phone number.

[0126] Step 208: The working honeypot server sets the calling number corresponding to the media signaling message as a good phone number, and the process ends.

[0127] Step 209: The working honeypot server sets the calling number corresponding to the media signaling message as a bad phone number.

[0128] In the embodiment of the present invention, the bad telephone number includes a nuisance telephone number, a fraudulent telephone number or a toll-charging telephone number.

[0129] In the embodiment of the present invention, as an optional solution, step 209 specifically includes: the working honeypot server identifies the calling number corresponding to the media signaling message according to the feature cluster name corresponding to the hit feature cluster, and sets the calling number as a bad phone number.

[0130] Step 210: The working honeypot server stops receiving media signaling messages sent by the distribution units in the honeypot pool.

[0131] Step 211: The working honeypot server sends a blocking instruction to the server so that the server can block the terminal device corresponding to the media signaling message.

[0132] In the embodiment of the present invention, the server blocks the terminal device corresponding to the media signaling message to block the call of the current calling number.

[0133] In a technical solution of a method for identifying bad phone numbers provided in an embodiment of the present invention, multiple keyword information and part-of-speech information corresponding to each keyword information are generated by parsing the generated media stream information; a significant intensity coefficient variance is generated based on the multiple keyword information and the multiple part-of-speech information; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number. In an embodiment of the present invention, a significant intensity coefficient variance is generated by calculating the acquired media signaling message; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number, thereby improving the accuracy of identifying bad phone numbers.

[0134] In the technical solution provided by the embodiment of the present invention, it is possible to identify based on the media signaling messages in the call, and the timeliness of handling bad phone numbers is high. It is also applicable to the identification of all medium-high frequency and medium-low frequency bad phone numbers, and the features of the media content information are calculated and algorithm matching is performed in combination with the domain feature library to block bad calls, thereby improving the timeliness of handling bad phone numbers.

[0135] In the technical solution provided by the embodiment of the present invention, the multi-field feature library content organization design method can achieve refined classification, can widely cover the personalized call content features of various types of outbound bad calls in different fields, improve recognition accuracy, and thus reduce the risk of false interception.

[0136] The technical solution provided by the embodiment of the present invention does not need to consume terminal device resources and can be applied to all mobile phone numbers.

[0137] The embodiment of the present invention provides a device for identifying bad telephone numbers. Figure 7 A schematic diagram of a bad phone number identification device provided by an embodiment of the present invention is shown in FIG. Figure 7 As shown, the device includes: a first generating module 31, a second generating module 32, a judging module 33 and a first setting module 34.

[0138] The first generating module 31 is used to parse the acquired media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information.

[0139] The second generating module 32 is used to perform calculations based on multiple keyword information and multiple part-of-speech information to generate a significant intensity coefficient variance.

[0140] The judgment module 33 is used to judge whether the significant intensity coefficient variance is located in a closed interval, and the closed interval includes a numerical range between a first set threshold and a second set threshold.

[0141] The first setting module 34 is used for setting the calling number corresponding to the media signaling message as a bad phone number if the judging module 33 judges that the significant intensity coefficient variance is in a closed interval.

[0142] In the embodiment of the present invention, the device further includes: a receiving module 35 , an extracting module 36 and a third generating module 37 .

[0143] The receiving module 35 is used to receive a media signaling message sent by a distribution unit in the honeypot pool. The media signaling message is sent by a terminal device to the distribution unit.

[0144] The extraction module 36 is used to extract multiple media content information of the calling direction from the media signaling message.

[0145] The third generating module 37 is used to arrange the multiple media content information in chronological order to generate media stream information.

[0146] In the embodiment of the present invention, the first generating module 31 includes: a first generating submodule 311 and a second generating submodule 312 .

[0147] The first generating submodule 311 is used to calculate the media stream information through the speech recognition technology to generate media text information.

[0148] The second generating submodule 312 is used to parse the media text information through word segmentation technology to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information.

[0149] In the embodiment of the present invention, the second generation module 32 includes: a third generation submodule 321 , a fourth generation submodule 322 , a fifth generation submodule 323 , a sixth generation submodule 324 and a seventh generation submodule 325 .

[0150] The third generation submodule 321 is used to calculate multiple keyword information through a text ranking algorithm to generate multiple sensitive word strength coefficients.

[0151] The fourth generation submodule 322 is used to calculate multiple part-of-speech information through an expert algorithm to generate multiple part-of-speech strength coefficients.

[0152] The fifth generating submodule 323 is used to generate a plurality of significant strength coefficients according to a plurality of sensitive word strength coefficients and a plurality of part-of-speech strength coefficients.

[0153] The sixth generating submodule 324 is used to calculate the average value of multiple significant intensity coefficients to generate an average significant intensity coefficient.

[0154] The seventh generating submodule 325 is used to generate a significant intensity coefficient variance according to a plurality of significant intensity coefficients and an average significant intensity coefficient.

[0155] In the embodiment of the present invention, the device further includes: a second setting module 38 .

[0156] The second setting module 38 is used to set the calling number corresponding to the media signaling message as a good telephone number if the judging module 33 judges that the significant intensity coefficient variance is not located in a closed interval.

[0157] In the embodiment of the present invention, the device further includes: a sending module 39.

[0158] The receiving module 35 is further used to stop receiving media signaling messages sent by the distribution units in the honeypot pool, where the media signaling messages are sent by the terminal devices to the distribution units.

[0159] The sending module 39 is used to send a blocking instruction to the honeypot pool so that the honeypot pool can block the terminal device corresponding to the media signaling message.

[0160] In an embodiment of the present invention, the fifth generation submodule 323 is specifically used to calculate multiple sensitive word strength coefficients and multiple part-of-speech strength coefficients through the formula θ=α*β to generate multiple significant strength coefficients, where θ is the significant strength coefficient, α is the part-of-speech strength coefficient, and β is the sensitive word strength coefficient.

[0161] The seventh generation submodule 325 is specifically used to generate Multiple significant intensity coefficients and average significant intensity coefficients are calculated to generate significant intensity coefficient variance, where θ i is the significant strength coefficient, is the average significant intensity coefficient, and n is the total number of significant intensity coefficients.

[0162] In a technical solution of a method for identifying bad phone numbers provided in an embodiment of the present invention, multiple keyword information and part-of-speech information corresponding to each keyword information are generated by parsing the generated media stream information; a significant intensity coefficient variance is generated based on the multiple keyword information and the multiple part-of-speech information; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number. In an embodiment of the present invention, a significant intensity coefficient variance is generated by calculating the acquired media signaling message; if it is determined that the significant intensity coefficient variance is in a closed interval, the calling number corresponding to the media signaling message is set as a bad phone number, thereby improving the accuracy of identifying bad phone numbers.

[0163] The bad phone number identification device provided in this embodiment can be used to achieve the above Figure 2 and Figure 3 For the method for identifying bad phone numbers in the embodiment of the method for identifying bad phone numbers, the specific description can be found in the above-mentioned embodiment of the method for identifying bad phone numbers, which will not be repeated here.

[0164] An embodiment of the present invention provides a storage medium, which includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the steps of the embodiment of the above-mentioned bad phone number identification method. For specific description, please refer to the embodiment of the above-mentioned bad phone number identification method.

[0165] An embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions. When the program instructions are loaded and executed by the processor, the steps of the embodiment of the above-mentioned bad phone number identification method are implemented. For a specific description, please refer to the embodiment of the above-mentioned bad phone number identification method.

[0166] Figure 8 A schematic diagram of a computer device provided by an embodiment of the present invention. Figure 8 As shown, the computer device 40 of this embodiment includes: a processor 41, a memory 42, and a computer program 43 stored in the memory 42 and executable on the processor 41. When the computer program 43 is executed by the processor 41, the method for identifying bad phone numbers in the embodiment is implemented. To avoid repetition, it is not described one by one here. Alternatively, when the computer program is executed by the processor 41, the functions of each model / unit in the device for identifying bad phone numbers in the embodiment are implemented. To avoid repetition, it is not described one by one here.

[0167] The computer device 40 includes, but is not limited to, a processor 41 and a memory 42. Those skilled in the art will appreciate that Figure 8It is only an example of the computer device 40 and does not constitute a limitation of the computer device 40. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, a network device may also include input and output devices, network access devices, buses, etc.

[0168] The processor 41 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0169] The memory 42 may be an internal storage unit of the computer device 40, such as a hard disk or memory of the computer device 40. The memory 42 may also be an external storage device of the computer device 40, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the computer device 40. Further, the memory 42 may also include both an internal storage unit of the computer device 40 and an external storage device. The memory 42 is used to store computer programs and other programs and data required by the network device. The memory 42 may also be used to temporarily store data that has been output or is to be output.

[0170] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0171] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0172] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0173] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0174] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (Processor) to perform some steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0175] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying bad phone numbers, characterized in that: Applied to a working honeypot server in a honeypot pool, the method comprises: Parsing the generated media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each of the keyword information; Calculate based on the plurality of keyword information and the plurality of part-of-speech information to generate a significant intensity coefficient variance; Determine whether the significant intensity coefficient variance is in a closed interval, where the closed interval includes a numerical range between a first set threshold and a second set threshold; If it is determined that the significant intensity coefficient variance is within the closed interval, the calling number corresponding to the media signaling message is set as a bad phone number, where the bad phone number includes a harassing phone number, a fraudulent phone number, or a toll-charging phone number; The calculation based on the plurality of keyword information and the plurality of part-of-speech information to generate the significant intensity coefficient variance comprises: Calculate the multiple keyword information by using a text ranking algorithm to generate multiple sensitive word strength coefficients; Calculating the plurality of part-of-speech information by an expert algorithm to generate a plurality of part-of-speech strength coefficients; Generating a plurality of significant intensity coefficients according to the plurality of sensitive word intensity coefficients and the plurality of part-of-speech intensity coefficients; Calculating an average value of a plurality of the significant intensity coefficients to generate an average significant intensity coefficient; A significant intensity coefficient variance is generated based on the plurality of significant intensity coefficients and the average significant intensity coefficient.

2. The method according to claim 1, characterized in that The step of parsing the generated media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information includes: Receiving a media signaling message sent by a distribution unit in the honeypot pool, wherein the media signaling message is sent by a terminal device to the distribution unit; Extracting multiple media content information of the calling direction from the media signaling message; Arrange the plurality of media content information in chronological order to generate media stream information.

3. The method according to claim 1, characterized in that The parsing of the media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information includes: Calculating the media stream information through speech recognition technology to generate media text information; The media text information is parsed by word segmentation technology to generate a plurality of keyword information and part-of-speech information corresponding to each keyword information.

4. The method according to claim 1, characterized in that: Also includes: If it is determined that the significant intensity coefficient variance is not located in the closed interval, the calling number corresponding to the media signaling message is set as a good telephone number.

5. The method according to claim 1, characterized in that The step of setting the calling number corresponding to the media signaling message to a bad phone number includes: Stop receiving a media signaling message sent by a distribution unit in the honeypot pool, where the media signaling message is sent by a terminal device to the distribution unit; A blocking instruction is sent to the server so that the server can block the terminal device corresponding to the media signaling message.

6. The method according to claim 1, characterized in that The generating a plurality of significant intensity coefficients according to the plurality of sensitive word intensity coefficients and the plurality of part-of-speech intensity coefficients comprises: By formula Calculating the multiple sensitive word strength coefficients and the multiple part-of-speech strength coefficients to generate multiple significant strength coefficients, wherein θ is the significant strength coefficient, α is the part-of-speech strength coefficient, and β is the sensitive word strength coefficient; Generating a significant intensity coefficient variance according to the plurality of significant intensity coefficients and the average significant intensity coefficient comprises: By formula The significant intensity coefficient variance is generated by calculating a plurality of the significant intensity coefficients and the average significant intensity coefficient, wherein is the significant intensity coefficient, is the average significant intensity coefficient, and n is the total number of the significant intensity coefficients.

7. A device for identifying bad phone numbers, characterized in that: include: A first generating module is used to parse the acquired media stream information to generate a plurality of keyword information and part-of-speech information corresponding to each of the keyword information; A second generating module, configured to generate a significant intensity coefficient variance by performing calculations based on the plurality of keyword information and the plurality of part-of-speech information; A judging module, used for judging whether the significant intensity coefficient variance is in a closed interval, wherein the closed interval includes a numerical range between a first set threshold and a second set threshold, and the first set threshold and the second set threshold are endpoints of the closed interval; A first setting module is used to set the calling number corresponding to the media signaling message to a bad phone number if the judgment module determines that the significant intensity coefficient variance is located in the closed interval, and the bad phone number includes a harassing phone number, a fraudulent phone number or a toll-charging phone number; The second generation module includes: a third generation submodule, a fourth generation submodule, a fifth generation submodule, a sixth generation submodule and a seventh generation submodule; The third generating submodule is used to calculate multiple keyword information through a text ranking algorithm to generate multiple sensitive word strength coefficients; The fourth generating submodule is used to calculate the multiple part-of-speech information by an expert algorithm to generate multiple part-of-speech strength coefficients; The fifth generating submodule is used to generate a plurality of significant intensity coefficients according to a plurality of sensitive word intensity coefficients and a plurality of part-of-speech intensity coefficients; The sixth generation submodule is used to calculate the average value of multiple significant intensity coefficients to generate an average significant intensity coefficient; The seventh generating submodule is used to generate a significant intensity coefficient variance according to a plurality of significant intensity coefficients and an average significant intensity coefficient.

8. A storage medium, characterized in that: include: The storage medium includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the bad telephone number identification method according to any one of claims 1 to 6.

9. A computer device comprising a memory and a processor, wherein the memory is used to store information including program instructions, and the processor is used to control the execution of the program instructions, characterized in that: When the program instructions are loaded and executed by the processor, the steps of the bad telephone number identification method according to any one of claims 1 to 6 are implemented.

Images