Telephone fraud early warning system based on artificial intelligence

Abstract

The invention provides a telephone fraud early warning system based on artificial intelligence, which relates to the field of electric digital data processing, and comprises a communication channel monitoring module, a parameter acquisition and analysis module, an intelligent fraud analysis module and an early warning prompt module, the parameter acquisition and analysis module is used for acquiring call information of a calling number communication channel and performing preliminary analysis, the intelligent fraud analysis module performs deep analysis based on the acquired call information to judge whether the call is a telephone fraud, and the early warning prompt module sends out early warning information when detecting the telephone fraud; the system can intelligently analyze the communication relationship of the calling number, excavate the related communication subject information, and compare the formed network structure, so that the phone fraud can be effectively identified.

Description

Technical Field

[0001] The present invention relates to the field of electronic digital data processing, and in particular to an artificial intelligence-based telephone fraud early warning system. Background Art

[0002] With the popularity of mobile phones, telephone fraud has become a very common form of fraud. At the same time, the content of the fraud is becoming increasingly difficult to identify. Therefore, there is a need for a fraud warning system that does not require analysis of the content of the call but analyzes the communication data of the caller number to identify telephone fraud, effectively protect the user's property safety, and reduce the success rate of telephone fraud.

[0003] The foregoing discussion of the background art is intended only to facilitate an understanding of the present invention. This discussion does not acknowledge or admit that any of the material referred to is part of the common general knowledge.

[0004] Many fraud warning systems have been developed. After a lot of searching and reference, it is found that the existing fraud warning systems include the system disclosed in the publication number CN117202200A. These systems generally include establishing an anti-fraud application download data channel with the anti-telephone fraud server when the calling number is a fraudulent call. The anti-fraud application download data channel is a channel established by adding data channel SDP parameters in the session request signaling sent from the calling user end to the called user end. The data channel SDP parameters refer to the data channel negotiation SDP parameters of the anti-telephone fraud server; based on the anti-fraud application download data channel, telephone fraud warning is performed. However, this system is an early warning method under known fraudulent calls, but there is no specific method for identifying fraudulent calls, so it cannot effectively identify fraudulent calls and reduce the success rate of telephone fraud. Summary of the invention

[0005] The purpose of the present invention is to propose an artificial intelligence-based telephone fraud early warning system to address the existing deficiencies.

[0006] The present invention adopts the following technical solution:

[0007] An artificial intelligence-based telephone fraud early warning system, including a communication channel monitoring module, a parameter collection and analysis module, an intelligent fraud analysis module and an early warning prompt module;

[0008] The communication channel monitoring module is used to monitor the call information of the communication channel, the parameter collection and analysis module is used to collect the call information of the caller number communication channel and perform preliminary analysis, the intelligent fraud analysis module performs in-depth analysis based on the collected call information to determine whether it is a telephone fraud, and the early warning prompt module issues an early warning message when a telephone fraud is detected;

[0009] The communication channel monitoring module includes a session request recording unit and a call duration recording unit, wherein the session request recording unit is used to record the session request information issued by each communication subject, and the call duration recording unit is used to record the call duration information of each communication subject in a successfully connected state;

[0010] The parameter collection and analysis module includes an activation query unit, an information storage unit and a primary detection unit, wherein the activation query unit is used to query the communication channel monitoring module for call information of the calling number when receiving the session request information, the information storage unit is used to store the queried call information, and the primary detection unit is used to analyze the call information and determine whether to conduct in-depth analysis;

[0011] The intelligent fraud analysis module includes an object screening unit, an intelligent analysis unit and a structure judgment unit. The object screening unit is used to screen out communication subjects that are closely related to the calling number, the intelligent analysis unit is used to perform intelligent analysis on the screened communication subjects, and the structure judgment unit analyzes and judges the mesh structure formed by the calling number and the screened communication subjects;

[0012] Further, the primary detection unit includes a pre-statistical processor and a primary calculation processor, the pre-statistical processor is used to classify and count the collected call information, and the primary calculation processor is used to calculate and analyze the statistical results and determine whether to perform in-depth analysis on the calling number;

[0013] The items counted by the pre-statistics processor include the number of rejections n 1 、Accept quantity n 2 、Number of objects n 3 And the total call duration of each object {T i};

[0014] The primary calculation processor calculates the primary judgment value P according to the following formula:

[0015]

[0016] Among them, T0 is the time base;

[0017] When the primary judgment value is greater than the primary threshold, it is determined that in-depth analysis is required and the calling number information is sent to the intelligent fraud analysis module;

[0018] Further, the object screening unit includes an association calculation processor, a process control processor and a subject record processor, wherein the association calculation processor is used to calculate the degree of association between two communication subjects, the process control processor is used to control the screening process of the communication subjects, and the subject record processor is used to record the screened communication subject information;

[0019] Further, the intelligent parsing unit includes a detection calling processor and an attribute judgment processor, the detection calling processor is used to call the primary detection unit to calculate the primary judgment value of all communication subjects, and the attribute judgment processor parses and judges the attribute of each communication subject;

[0020] The attribute judgment processor calculates the command index Cm of the communication subject according to the following formula:

[0021]

[0022] Where m is the number of other communication entities associated with the communication entity, P i represents the primary judgment value of the i-th associated communication subject, P 0 Indicates the primary judgment value of the communication subject itself;

[0023] The attribute judgment processor uses (m, Cm) as the attribute vector of the communication subject, and assigns corresponding attributes according to the attribute vector, and the attributes include command, execution and commonality;

[0024] Furthermore, the early warning prompt module includes a text display unit and a voice prompt unit, the text display unit is used to display fraud prompt information on the terminal before the conversation is connected, and the voice prompt unit is used to issue a fraud prompt voice after the conversation is connected.

[0025] The beneficial effects achieved by the present invention are:

[0026] This system divides fraud identification into two parts: primary identification and in-depth identification. Primary identification can simply analyze and process the data, filter out a large number of normal calls, and reduce the amount of calculation. In-depth identification can dig out the communication relationship network behind the caller number, and perform calculations and analysis on the communication relationship network, determine the role of different communication subjects in the relationship network, and ultimately identify the type of fraud network. It can effectively identify telephone fraud and issue early warnings, thereby reducing the success rate of telephone fraud.

[0027] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only for reference and description and are not intended to limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a schematic diagram of the overall structural framework of the present invention;

[0029] Figure 2 This is a schematic diagram of the parameter acquisition and analysis module of the present invention;

[0030] Figure 3This is a schematic diagram of the intelligent fraud analysis module of the present invention;

[0031] Figure 4 This is a schematic diagram of the primary detection unit of the present invention;

[0032] Figure 5 It is a schematic diagram of the object screening unit of the present invention. DETAILED DESCRIPTION

[0033] The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted according to actual sizes. It is stated in advance. The following embodiments will further explain the relevant technical contents of the present invention in detail, but the disclosed contents are not intended to limit the scope of protection of the present invention.

[0034] Embodiment 1.

[0035] This embodiment provides a telephone fraud early warning system based on artificial intelligence, combined with Figure 1 , including communication channel monitoring module, parameter collection and analysis module, intelligent fraud analysis module and early warning prompt module;

[0036] The communication channel monitoring module is used to monitor the call information of the communication channel, the parameter collection and analysis module is used to collect the call information of the caller number communication channel and perform preliminary analysis, the intelligent fraud analysis module performs in-depth analysis based on the collected call information to determine whether it is a telephone fraud, and the early warning prompt module issues an early warning message when a telephone fraud is detected;

[0037] The communication channel monitoring module includes a session request recording unit and a call duration recording unit, wherein the session request recording unit is used to record the session request information issued by each communication subject, and the call duration recording unit is used to record the call duration information of each communication subject in a successfully connected state;

[0038] The parameter collection and analysis module includes an activation query unit, an information storage unit and a primary detection unit, wherein the activation query unit is used to query the communication channel monitoring module for call information of the calling number when receiving the session request information, the information storage unit is used to store the queried call information, and the primary detection unit is used to analyze the call information and determine whether to conduct in-depth analysis;

[0039] The intelligent fraud analysis module includes an object screening unit, an intelligent analysis unit and a structure judgment unit. The object screening unit is used to screen out communication subjects that are closely related to the calling number, the intelligent analysis unit is used to perform intelligent analysis on the screened communication subjects, and the structure judgment unit analyzes and judges the mesh structure formed by the calling number and the screened communication subjects;

[0040] The primary detection unit includes a pre-statistical processor and a primary calculation processor, wherein the pre-statistical processor is used to classify and count the collected call information, and the primary calculation processor is used to calculate and analyze the statistical results and determine whether to conduct in-depth analysis on the calling number;

[0041] The items counted by the pre-statistics processor include the number of rejections n 1 、Accept quantity n 2 、Number of objects n 3 And the total call duration of each object {T i};

[0042] The primary calculation processor calculates the primary judgment value P according to the following formula:

[0043]

[0044] Among them, T0 is the time base;

[0045] When the primary judgment value is greater than the primary threshold, it is determined that in-depth analysis is required and the calling number information is sent to the intelligent fraud analysis module;

[0046] The object screening unit includes an association calculation processor, a process control processor and a subject record processor, wherein the association calculation processor is used to calculate the degree of association between two communication subjects, the process control processor is used to control the screening process of the communication subjects, and the subject record processor is used to record the screened communication subject information;

[0047] The intelligent parsing unit includes a detection calling processor and an attribute judgment processor, wherein the detection calling processor is used to call the primary detection unit to calculate the primary judgment values ​​of all communication subjects, and the attribute judgment processor performs parsing and judgment on the attributes of each communication subject;

[0048] The attribute judgment processor calculates the command index Cm of the communication subject according to the following formula:

[0049]

[0050] Where m is the number of other communication entities associated with the communication entity, P i represents the primary judgment value of the i-th associated communication subject, P0 Indicates the primary judgment value of the communication subject itself;

[0051] The attribute judgment processor uses (m, Cm) as the attribute vector of the communication subject, and assigns corresponding attributes according to the attribute vector, and the attributes include command, execution and commonality;

[0052] The early warning prompt module includes a text display unit and a voice prompt unit. The text display unit is used to display fraud prompt information on the terminal before the conversation is connected, and the voice prompt unit is used to issue a fraud prompt voice after the conversation is connected.

[0053] Embodiment 2.

[0054] This embodiment includes all the contents of the first embodiment, and provides an artificial intelligence-based telephone fraud early warning system, including a communication channel monitoring module, a parameter collection and analysis module, an intelligent fraud analysis module, and an early warning prompt module;

[0055] The communication channel monitoring module is used to monitor the call information of the communication channel, the parameter collection and analysis module is used to collect the call information of the caller number communication channel and perform preliminary analysis, the intelligent fraud analysis module performs in-depth analysis based on the collected call information to determine whether it is a telephone fraud, and the early warning prompt module issues an early warning message when a telephone fraud is detected;

[0056] The communication channel monitoring module includes a session request recording unit and a call duration recording unit, wherein the session request recording unit is used to record the session request information issued by each communication subject, and the call duration recording unit is used to record the call duration information of each communication subject in a successfully connected state;

[0057] Combination Figure 2 The parameter collection and analysis module includes an activation query unit, an information storage unit and a primary detection unit. The activation query unit is used to query the communication channel monitoring module for call information of the calling number when receiving the session request information. The information storage unit is used to store the queried call information. The primary detection unit is used to analyze the call information and determine whether to conduct in-depth analysis;

[0058] Combination Figure 3 The intelligent fraud analysis module includes an object screening unit, an intelligent analysis unit and a structure judgment unit. The object screening unit is used to screen out communication subjects that are closely related to the calling number, the intelligent analysis unit is used to perform intelligent analysis on the screened communication subjects, and the structure judgment unit analyzes and judges the mesh structure formed by the calling number and the screened communication subjects;

[0059] The early warning prompt module includes a text display unit and a voice prompt unit. The text display unit is used to display fraud prompt information on the terminal before the conversation is connected, and the voice prompt unit is used to issue a fraud prompt voice after the conversation is connected;

[0060] The session request information recorded by the session request recording unit includes the request object, request time and request result, and the request result is divided into rejection and acceptance;

[0061] The session request recording unit includes a request number processor and a request information register, wherein the request number processor is used to set a number for each session request information, and the request information register is used to store each numbered session request information;

[0062] The call duration record unit number query processor and call information register, the number query processor retrieves the corresponding number from the session request record unit when the call is connected, and the call information register is used to store the call duration information of the retrieved number;

[0063] The activation query unit includes an activation response processor and a communication subject query processor, the activation response processor is used to respond to the session request information and obtain the caller number information, and the communication subject query processor is used to query the corresponding call information from the communication channel monitoring module using the caller number information as the communication subject;

[0064] The communication subject query processor only queries the call information within a fixed time, where the fixed time refers to a fixed time period before the current time point, and the specific value of the fixed time period is set by the user;

[0065] Combination Figure 4 The primary detection unit includes a pre-statistical processor and a primary calculation processor, wherein the pre-statistical processor is used to classify and count the collected call information, and the primary calculation processor is used to calculate and analyze the statistical results and determine whether to perform in-depth analysis on the calling number;

[0066] The items counted by the pre-statistics processor include the number of rejections n 1 、Accept quantity n 2 、Number of objects n 3 And the total call duration of each object {T i};

[0067] The primary calculation processor calculates the primary judgment value P according to the following formula:

[0068]

[0069] Among them, T0 is the time base;

[0070] When the primary judgment value is greater than the primary threshold, it is determined that in-depth analysis is required and the calling number information is sent to the intelligent fraud analysis module;

[0071] Combination Figure 5 , the object screening unit includes an association calculation processor, a process control processor and a subject record processor, the association calculation processor is used to calculate the degree of association between two communication subjects, the process control processor is used to control the screening process of the communication subjects, and the subject record processor is used to record the screened communication subject information;

[0072] The workflow of the process control processor includes the following steps:

[0073] S1. Acquire a communication subject that has valid communication with the calling number;

[0074] S2, sending the calling number and one of the communication subjects obtained in step S1 as input parameters to the association calculation processor;

[0075] S3, receiving the result fed back by the association calculation processor, and sending the communication subject that meets the requirements to the subject record processor;

[0076] S4, repeating step S2 and step S3 until all the communication subjects obtained in step S1 are processed;

[0077] S5. Obtain a communication subject in the subject record processor as an extended subject;

[0078] S6. Acquire a communication subject that has valid communication with the extended subject;

[0079] S7, sending the extended subject and one of the communication subjects obtained in step S6 as input parameters to the association calculation processor;

[0080] S8, receiving the result fed back by the association calculation processor, and sending the communication subject that meets the requirements to the subject record processor;

[0081] S9, repeating step S7 and step S8 until all the communication subjects obtained in step S6 are processed;

[0082] S10, repeating steps S5 to S9 until all communication subjects acquired from step S3 in the subject record processor are processed;

[0083] The association calculation processor calculates the association index As between the two communication entities according to the following formula:

[0084] As = lg (n·T);

[0085] Where n is the number of communications between two communication entities, and T is the total communication duration between the two communication entities;

[0086] When the correlation index is greater than the correlation threshold, it means that the correlation degree of the two communication subjects meets the requirements and is correlated;

[0087] The intelligent parsing unit includes a detection calling processor and an attribute judgment processor, wherein the detection calling processor is used to call the primary detection unit to calculate the primary judgment values ​​of all communication subjects, and the attribute judgment processor performs parsing and judgment on the attributes of each communication subject;

[0088] The communication subjects processed by the intelligent analysis unit are all communication subjects recorded in the subject record processor;

[0089] The attribute judgment processor calculates the command index Cm of the communication subject according to the following formula:

[0090]

[0091] Where m is the number of other communication entities associated with the communication entity, P i represents the primary judgment value of the i-th associated communication subject, P 0 Indicates the primary judgment value of the communication subject itself;

[0092] The attribute judgment processor uses (m, Cm) as the attribute vector of the communication subject and assigns corresponding attributes according to the attribute vector. The attribute judgment processor has built-in attribute vector intervals corresponding to different attributes, and the attributes include command, execution and commonality;

[0093] The structure judgment unit includes a mesh generation processor and a structure comparison processor, wherein the mesh generation processor generates corresponding mesh connectivity information based on the attribute judgment result, and the structure comparison processor compares the generated mesh connectivity information with the fraudulent mesh structure;

[0094] The mesh generation processor retains the communication subjects with command and execution attributes, and connects the communication subjects based on relevance to obtain a mesh structure;

[0095] The structure comparison processor records multiple patterns of fraudulent mesh structures and compares them with the generated mesh structure one by one;

[0096] The i appearing in the above text is an ordinal number used to indicate a sequence number.

[0097] The contents disclosed above are only preferred feasible embodiments of the present invention, and do not limit the protection scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention specification and drawings are included in the protection scope of the present invention. In addition, the elements therein can be updated as technology develops.

Claims

1. An artificial intelligence-based telephone fraud early warning system, characterized in that: It includes communication channel monitoring module, parameter collection and analysis module, intelligent fraud analysis module and early warning prompt module; The communication channel monitoring module is used to monitor the call information of the communication channel, the parameter collection and analysis module is used to collect the call information of the caller number communication channel and perform preliminary analysis, the intelligent fraud analysis module performs in-depth analysis based on the collected call information to determine whether it is a telephone fraud, and the early warning prompt module issues an early warning message when a telephone fraud is detected; The communication channel monitoring module includes a session request recording unit and a call duration recording unit, wherein the session request recording unit is used to record the session request information issued by each communication subject, and the call duration recording unit is used to record the call duration information of each communication subject in a successfully connected state; The parameter collection and analysis module includes an activation query unit, an information storage unit and a primary detection unit, wherein the activation query unit is used to query the communication channel monitoring module for call information of the calling number when receiving the session request information, the information storage unit is used to store the queried call information, and the primary detection unit is used to analyze the call information and determine whether to conduct in-depth analysis; The intelligent fraud analysis module includes an object screening unit, an intelligent analysis unit and a structure judgment unit. The object screening unit is used to screen out communication subjects that are closely related to the calling number, the intelligent analysis unit is used to perform intelligent analysis on the screened communication subjects, and the structure judgment unit analyzes and judges the network structure formed by the calling number and the screened communication subjects.

2. The artificial intelligence-based telephone fraud early warning system according to claim 1, characterized in that: The primary detection unit includes a pre-statistical processor and a primary calculation processor, wherein the pre-statistical processor is used to classify and count the collected call information, and the primary calculation processor is used to calculate and analyze the statistical results and determine whether to conduct in-depth analysis on the calling number; The items counted by the front-end statistical processor include the number of rejections n1, the number of acceptances n2, the number of objects n3, and the total call duration of each object {T i }; The primary calculation processor calculates the primary judgment value P according to the following formula: Among them, T0 is the time base; When the primary judgment value is greater than the primary threshold, it is determined that in-depth analysis is required and the calling number information is sent to the intelligent fraud analysis module.

3. The artificial intelligence-based telephone fraud early warning system as claimed in claim 2, characterized in that: The object screening unit includes an association calculation processor, a process control processor and a subject record processor. The association calculation processor is used to calculate the degree of association between two communication subjects, the process control processor is used to control the screening process of the communication subjects, and the subject record processor is used to record the screened communication subject information.

4. The artificial intelligence-based telephone fraud early warning system as claimed in claim 3, characterized in that: The intelligent parsing unit includes a detection calling processor and an attribute judgment processor, wherein the detection calling processor is used to call the primary detection unit to calculate the primary judgment values ​​of all communication subjects, and the attribute judgment processor performs parsing and judgment on the attributes of each communication subject; The attribute judgment processor calculates the command index Cm of the communication subject according to the following formula: Where m is the number of other communication entities associated with the communication entity, P i represents the primary judgment value of the i-th associated communication subject, and P0 represents the primary judgment value of the communication subject itself; The attribute judgment processor uses (m, Cm) as the attribute vector of the communication subject, and assigns corresponding attributes according to the attribute vector, and the attributes include command, execution and commonality.

5. The artificial intelligence-based telephone fraud early warning system as claimed in claim 4, characterized in that: The early warning prompt module includes a text display unit and a voice prompt unit. The text display unit is used to display fraud prompt information on the terminal before the conversation is connected, and the voice prompt unit is used to issue a fraud prompt voice after the conversation is connected.

Images