Method and system for intercepting crank calls
By integrating blacklist data and incoming user feature data, calculating the stability coefficient X and constructing behavioral correspondence, and cross-comparing it with historical call records, the problem of difficult to identify camouflage harassing calls in existing systems is solved, significantly improving interception efficiency and preventing user harassment.
Patent Information
- Application Number
- CN202411576991.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-06
- Publication Date
- 2025-02-28
AI Technical Summary
The existing harassing phone interception system is difficult to effectively identify harassing phones disguised as legal phone numbers, resulting in reduced interception efficiency.
By logging in to the platform that records public labeling information, integrating blacklist data and characteristic data of incoming users, calculating the stability coefficient X of incoming users, and building a behavioral correspondence relationship, judging the second state of incoming users, and cross-comparing them with historical call records to ensure the correctness of state judgment.
Effective identification and interception of disguised harassing phone calls is achieved, significantly improving the identification capability of the harassing phone call interception system, and preventing called users from being subjected to unnecessary harassment.
Smart Images
Figure CN119544865A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of harassing call blocking, and in particular to a harassing call blocking method and system. Background Art
[0002] In order to deal with the problem of harassing calls more effectively, harassing call blocking systems are emerging. The blocking system technology combines advanced technologies such as big data analysis, cloud computing, and artificial intelligence. It can more accurately identify harassing calls, thereby blocking harassing calls and adding the intercepted harassing calls to the blacklist of the platform that records public annotation information.
[0003] At present, the principle of blocking harassing calls is mainly to identify the caller's place of origin, blacklist, call frequency and other characteristics, and identify multiple callers who do not meet the preset conditions as malicious harassing calls. When the caller is identified as a malicious harassing call, the system will automatically block it to prevent the user from being harassed. However, criminals may use special software to disguise the caller's number as a seemingly legitimate phone number, including a public service number, to increase the answering rate, and imitate the voice prompts of regular institutions after the call is connected, so that the victim mistakenly believes that it is a legitimate call, thereby evading the identification of the harassing call blocking system, thereby achieving the purpose of harassment, reducing the interception efficiency of the harassing call blocking system, and having certain defects. Therefore, the present invention proposes a harassing call blocking method and system, aiming to effectively identify disguised harassing calls. Summary of the invention
[0004] To this end, the present invention provides a method and system for blocking harassing calls to solve the above-mentioned problems in the prior art.
[0005] In order to achieve the above object, the present invention provides the following technical solutions:
[0006] According to a first aspect of the present invention, a method for blocking harassing calls specifically comprises the following steps:
[0007] S1: Log in to the platform that records public annotation information, and export the blacklist data marked by the public in the platform, and obtain the feature data of the caller, and integrate the blacklist data and the feature data to form a unified data set. The feature data includes the caller's phone number, marking status, call time, and call frequency;
[0008] S2: Determine the first state of the number based on the data set in S1, and query the calling user number location information, wherein the determination of the first state is based on the phone number and the mark state in the data set;
[0009] S3: calculating the stability coefficient X of the incoming caller according to the data set of the incoming caller and the first state, constructing a behavior correspondence relationship, and determining the second state of the incoming caller according to the behavior correspondence relationship;
[0010] The calculation formula of stability factor X is:
[0011]
[0012] Among them, t is the total call duration of the caller in one month, n i is the number of calls made by the calling user, and y is the mean;
[0013] S4: Filter the blacklist data in S1, select the historical call records of the harassing calls that are in contact with the calling user in the blacklist data, and perform cross-comparison to obtain the final status to ensure the correctness of the judgment status;
[0014] S5: Summarize and analyze the final results, and take corresponding measures to allow normal calls and intercept calls based on the final status of the output.
[0015] Furthermore, the marking status includes the marking times and marking type of the calling user.
[0016] Furthermore, the calculation formula of the mean y is:
[0017]
[0018] Among them, t is the total call duration of the caller in one month, n i is the number of calls made by the calling user, and y is the mean.
[0019] Furthermore, the stability coefficient X of the incoming caller within half a year is calculated, and the stability coefficient X-time t is generated. i (i=1, 2, 3...6), and determine the fluctuation range of the stability coefficient X based on the curve graph.
[0020] Furthermore, if the fluctuation amplitude is greater than <1, it means that the call behavior of the incoming caller remains stable within a certain period of time and has a certain regularity and stability. If the fluctuation amplitude is greater than ≥1, it means that the call behavior of the incoming caller shows a large volatility and is significantly different from the behavior pattern of normal users. The incoming caller is classified as a suspected call and marked as the second state.
[0021] Furthermore, the cross comparison requires real-time collection of the caller's most recent historical call content through the call center platform, and construction of a comparison framework, through which the historical call records are compared with the harassing call content marked in the blacklist.
[0022] The present invention also provides a system for implementing the above-mentioned harassing call blocking method:
[0023] Further, the system includes a data processing module, a behavior building module, a state judgment module, a verification module and a decision module, and the data processing module includes a data access module, a data integration module and a query module;
[0024] A data processing module, which allows the called user to connect to the data storage system of the platform that records the public annotation information, so that the user can obtain the blacklist data from the platform that records the public annotation information, and integrate the blacklist data and the feature data of the calling user into a data set;
[0025] The behavior building module is used to process and calculate the data according to the preset rules and algorithms, so as to calculate the stability coefficient X for the caller's data set and the first state through the machine learning algorithm, and after calculating the stability coefficient X, the behavior building module is responsible for building the behavior correspondence relationship according to the stability coefficient X and other relevant features;
[0026] State judgment module: It can receive the calculation results and curve graphs of the behavior building module, analyze and judge the current state or condition of the behavior building module, classify and mark the numbers according to preset rules and machine learning algorithms, and output the judged second state to the verification module.
[0027] Furthermore, the verification module is used to verify the status judgment of the incoming caller, randomly divide the real-time collected incoming caller data into data sets, and build a comparison framework through an algorithm model to verify the final status of the incoming caller data through the comparison framework;
[0028] The decision module includes an allowing submodule and an intercepting submodule. When the verification module determines that the incoming call is a normal call, the allowing submodule is responsible for executing the operation of allowing the call, and can establish a call connection and maintain the call status;
[0029] When the verification module determines that the incoming call is a harassing call, the interception submodule is responsible for executing the interception operation, hanging up the call, and sending an interception notification.
[0030] The present invention has the following advantages:
[0031] Through the behavior construction module and the state judgment module, a dual state recognition mechanism can be implemented for incoming callers, which can more effectively identify and intercept harassing calls disguised as seemingly legitimate phone numbers, significantly improving the recognition ability of the harassing call blocking system, thereby effectively preventing the called users from being unnecessary harassment. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1A flowchart of a method for blocking harassing calls provided in some embodiments of the present invention.
[0033] Figure 2 A schematic diagram of the module composition of a harassing call blocking system provided for some embodiments of the present invention. DETAILED DESCRIPTION
[0034] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the 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.
[0035] Example 1
[0036] like Figure 1 to Figure 2 As shown, a method and system for blocking harassing calls in the first embodiment of the present invention, the method for blocking harassing calls specifically comprises the following steps:
[0037] S1: Log in to the platform that records public annotation information, and export the blacklist data marked by the public in the platform, and obtain the feature data of the caller at the same time, integrate the blacklist data and feature data into a unified data set. Before integration, the feature data needs to be cleaned, duplicate records need to be removed, and erroneous data needs to be corrected. Then, the data structures of different sources in the feature data are uniformly adjusted and the field names are standardized and merged to form a complete data set, so as to achieve the purpose of data integration, ensure the accuracy and reliability of the data, and provide a basis for subsequent analysis; the feature data includes the caller's phone number, marking status, call time, and call frequency;
[0038] S2: Determine the first state of the number based on the data set in S1, and query the calling user's number location information, wherein the determination of the first state is based on the phone number and the mark state in the data set, and the mark state includes the number of marks and the mark type of the calling user;
[0039] If the number of marks is ≥5 and the mark type is a nuisance call, the caller will be classified as a nuisance call, and the interception module will directly intercept the caller. If there is no mark status, the caller will be classified as a call to be verified and marked as the first status. The first status is a simplified classification, which is mainly used to quickly identify the main features of the caller.
[0040] S3: calculating the stability coefficient X of the incoming caller according to the data set of the incoming caller and the first state, constructing a behavior correspondence relationship, and determining the second state of the incoming caller according to the behavior correspondence relationship;
[0041] The calculation formula of stability factor X is:
[0042]
[0043] Among them, t is the total call duration of the caller in one month, n i is the number of calls made by the calling user, and y is the mean;
[0044] The calculation formula for the mean y is:
[0045]
[0046] Calculate the stability coefficient X of incoming callers within half a year and make stability coefficient X-time t i (i=1, 2, 3...6), and the fluctuation range of the stability coefficient X is determined according to the curve graph. If the fluctuation range is greater than <1, it means that the call behavior of the incoming caller remains stable within a certain period of time, with certain regularity and stability. If the fluctuation range is greater than ≥1, it means that the call behavior of the incoming caller shows a large fluctuation and is significantly different from the behavior pattern of normal users. The incoming caller is classified as a suspected call and marked as the second state.
[0047] S4: Filter the blacklist data in S1, select the historical call records of the harassing calls that are in contact with the calling user in the blacklist data, and perform cross-comparison to obtain the final status to ensure the correctness of the judgment status;
[0048] Cross-comparison requires selecting the historical call records of harassing calls that are connected to the caller in the blacklist data and the historical call records of the caller, and building a comparison framework. The comparison framework is used to compare the call duration and call frequency of the historical call records of the harassing calls and the historical call records of the caller. The comparison framework can clearly show the connection and difference between the call start time, hang-up time, and connection time, so as to carry out effective comparison and verification. The comparison range threshold is preset. If the range threshold of the connection time of the historical call record of the caller and the connection time corresponding to the harassing call is ≥5 seconds, it is judged that there is a difference between the caller and the harassing call, and the caller is judged to be a normal call. If the range threshold of the connection time of the historical call record of the caller and the connection time corresponding to the harassing call is <5 seconds, it is judged that the caller is similar to the harassing call, and the caller is judged to be a harassing call and blocked;
[0049] When processing the historical call records of callers, only necessary information is collected and used, and personal information is desensitized to protect the privacy of callers;
[0050] S5: Summarize and analyze the final results, and take corresponding measures to allow normal calls and intercept calls based on the final status of the output.
[0051] The harassing call blocking system includes a data processing module, a behavior building module, a state judgment module, a verification module and a decision-making module, wherein the data processing module includes a data access module, a data integration module and a query module;
[0052] The data access module is used to verify the identity information of the called user. After the verification is completed, the data access module allows the called user to connect to the data storage system of the platform that records the public annotation information, which is convenient for obtaining blacklist data from the platform that records the public annotation information. The data integration module can integrate the blacklist data and the feature data of the calling user and merge them into a unified data set for comprehensive analysis and mining; the query module is the core part of the data processing module. It queries the database or API according to the input phone number to obtain the number's location information, and then outputs preliminary decision suggestions;
[0053] The behavior building module is used to process and calculate the data according to the preset rules and algorithms, so as to calculate the stability coefficient X for the caller's data set and the first state through the machine learning algorithm, and after calculating the stability coefficient X, the behavior building module is responsible for building the behavior correspondence relationship according to the stability coefficient X and other relevant features;
[0054] State judgment module: It can receive the calculation results and curve graphs of the behavior building module, analyze and judge the current state or condition of the behavior building module, classify and mark the numbers according to preset rules and machine learning algorithms, and output the judged second state to the verification module.
[0055] Verification module: used to verify the status judgment of the incoming caller, randomly divide the real-time collected incoming caller data into data sets, and build a comparison framework through the algorithm model to verify the final status of the incoming caller data.
[0056] The decision module includes an allowing submodule and an intercepting submodule. When the verification module determines that the incoming call is a normal call, the allowing submodule is responsible for executing the operation of allowing the call, and can establish a call connection and maintain the call status; Intercepting submodule: When the verification module determines that the incoming call is a harassing call, this submodule is responsible for executing the intercepting operation, hanging up the call, and sending an intercepting notification; thereby, the calling user can be dually identified in terms of status, and disguised harassing calls can be effectively identified and intercepted, thereby improving the recognition of the harassing call intercepting system, thereby achieving the purpose of preventing harassment by the called user.
Claims
1. A method for blocking harassing calls, characterized in that: The specific steps include: S1: Log in to the platform that records public annotation information, and export the blacklist data marked by the public in the platform, and obtain the feature data of the caller, and integrate the blacklist data and the feature data to form a unified data set. The feature data includes the caller's phone number, marking status, call time, and call frequency; S2: Determine the first state of the number based on the data set in S1, and query the calling user number location information, wherein the determination of the first state is based on the phone number and the mark state in the data set; S3: calculating the stability coefficient X of the incoming caller according to the data set of the incoming caller and the first state, constructing a behavior correspondence relationship, and determining the second state of the incoming caller according to the behavior correspondence relationship; The calculation formula of stability factor X is: Among them, t is the total call duration of the caller in one month, n i is the number of calls made by the calling user, and y is the mean; S4: Filter the blacklist data in S1, select the historical call records of the harassing calls that are in contact with the calling user in the blacklist data, and perform cross-comparison to obtain the final status to ensure the correctness of the judgment status; S5: Summarize and analyze the final results, and take corresponding measures to allow normal calls and intercept calls based on the final status of the output.
2. A method for intercepting harassing calls according to claim 1, characterized in that: The marking status includes the marking times and marking type of the calling user.
3. A method for blocking harassing calls according to claim 1, characterized in that: The calculation formula of the mean y is: Among them, t is the total call duration of the caller in one month, n i is the number of calls made by the calling user, and y is the mean.
4. A method for blocking harassing calls according to claim 1, characterized in that: The stability coefficient X of the incoming caller within half a year is calculated, and the stability coefficient X-time t is generated. i (i=1, 2, 3...6), and determine the fluctuation range of the stability coefficient X based on the curve graph.
5. A method for blocking harassing calls according to claim 4, characterized in that: If the fluctuation amplitude is greater than <1, it means that the call behavior of the incoming caller remains stable within a certain period of time and has a certain regularity and stability. If the fluctuation amplitude is greater than ≥1, it means that the call behavior of the incoming caller shows a large volatility and is significantly different from the behavior pattern of normal users. The incoming caller is classified as a suspected call and marked as the second state.
6. A method for blocking harassing calls according to claim 1, characterized in that: The cross comparison requires selecting the historical call records of the harassing calls that are connected to the calling user in the blacklist data and the historical call records of the calling user, and constructing a comparison framework to compare the call duration and call frequency of the historical call records of the harassing calls and the historical call records of the calling user through the comparison framework.
7. A nuisance call blocking system, characterized in that: The system includes a data processing module, a behavior building module, a state judgment module, a verification module and a decision module, and the data processing module includes a data access module, a data integration module and a query module; A data processing module, which allows the called user to connect to the data storage system of the platform that records the public annotation information, so that the user can obtain the blacklist data from the platform that records the public annotation information, and integrate the blacklist data and the feature data of the calling user into a data set; The behavior building module is used to process and calculate the data according to the preset rules and algorithms, so as to calculate the stability coefficient X for the caller's data set and the first state through the machine learning algorithm, and after calculating the stability coefficient X, the behavior building module is responsible for building the behavior correspondence relationship according to the stability coefficient X and other relevant features; State judgment module: It can receive the calculation results and curve graphs of the behavior building module, analyze and judge the current state or condition of the behavior building module, classify and mark the numbers according to preset rules and machine learning algorithms, and output the judged second state to the verification module.
8. A nuisance call blocking system according to claim 7, characterized in that: The verification module is used to verify the status judgment of the incoming caller, randomly divide the real-time collected incoming caller data into data sets, and build a comparison framework through an algorithm model to verify the final status of the incoming caller data through the comparison framework; The decision module includes an allowing submodule and an intercepting submodule. When the verification module determines that the incoming call is a normal call, the allowing submodule is responsible for executing the operation of allowing the call, and can establish a call connection and maintain the call status; Interception submodule: When the verification module determines that the incoming call is a harassment call, this submodule is responsible for executing the interception operation, hanging up the call, and sending an interception notification.