An anti-telecom fraud system based on big data cloud platform

Abstract

The present invention discloses an anti-telecom network fraud system based on a big data cloud platform, which relates to the technical field of anti-telecom network fraud. The present invention includes a user information acquisition module, a preliminary screening module, an information leakage degree analysis module, an information leakage judgment module, a leakage address analysis module and a display terminal. According to the call information of the registered user, it is screened to each target website through the preliminary screening module, and then the call recording corresponding to each harassing call obtained is used to analyze the information leakage corresponding to each target website. Through the address corresponding to each harassing call in each leaked website, the leakage situation of each website is analyzed, thereby avoiding the economic loss that may be caused to the user due to data leakage, improving the user experience, protecting the rights and interests of the user, improving network security, maintaining the healthy development of the network community, and promoting a positive social atmosphere.

Description

Technical Field

[0001] The present invention relates to the technical field of anti-telecom network fraud, and in particular to an anti-telecom network fraud system based on a big data cloud platform. Background Art

[0002] In recent years, with the rapid development of network and communication technologies, the integration of websites and social media platforms has become increasingly common. Users can log in to websites using their personal accounts. However, user accounts may be compromised by websites, which may lead to users being victims of telecommunications network fraud. Therefore, it is necessary to monitor the security of user accounts on websites.

[0003] Current anti-telecom fraud monitoring mainly focuses on monitoring users and ignores monitoring of websites, thus failing to guarantee the protection of user information by websites. This in turn increases the risk of user information leakage and the probability of users falling victim to telecommunications fraud, thereby reducing user experience, infringing user rights, reducing network security, destroying the healthy development of online communities, and hindering a positive social atmosphere. Summary of the Invention

[0004] In response to the above-mentioned technical deficiencies, the purpose of the present invention is to provide an anti-telecom network fraud system based on a big data cloud platform to comprehensively identify and prevent telecom network fraud. By analyzing abnormal situations on the network end, it can better protect users from telecom network fraud, purify the network environment, enhance the user experience, protect the user's rights and interests, improve network security, maintain the healthy development of the online community, and promote a positive social atmosphere.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides an anti-telecom fraud system based on a big data cloud platform.

[0006] The user information acquisition module is used to obtain the corresponding registered users of each website within a specified time period, and then monitor the registered users on each website to obtain the corresponding call information of each registered user on each website, where the call information includes the phone number of each caller, the call recording of each caller, and the corresponding address of each caller;

[0007] A preliminary screening module is used to determine whether the registered users on each website receive a harassing number from the calling party's phone number corresponding to each registered user on each website. If it is determined that the registered user on a certain website receives a harassing number, the website is recorded as a target website, thereby obtaining each target website;

[0008] The information leakage degree analysis module is used to obtain the number of registered users who received harassing calls on each target website, the number of harassing calls on each target website, and the corresponding call recordings of each harassing call, and then analyze the information leakage assessment coefficient corresponding to each target website;

[0009] The information leakage judgment module judges the leakage of each target website information according to the information leakage assessment coefficient of each target website, and then obtains each leaked website;

[0010] The leaked address analysis module is used to obtain the addresses corresponding to the harassing calls on each leaked website, and then analyze the leakage situation corresponding to each leaked website;

[0011] The display terminal is used to display each leaking website and the leakage situation corresponding to each leaking website.

[0012] Preferably, the specific judgment process of judging whether a registered user on each website receives a harassing number is as follows:

[0013] Compare the calling number of each registered user on each website with the calling number of the registered user. If the calling number of a registered user on a website is the same as the calling number of a registered user on a certain website, then determine that the calling number is the suspicious harassment number corresponding to the website. In this way, the suspicious harassment numbers corresponding to each website are obtained. Then, compare the calling number of each registered user on each website with the suspicious harassment numbers corresponding to each website. Count the number of registered users on each website that were called by each suspicious harassment number.

[0014] Compare the number of registered users dialed by each suspicious spam number on each website with the set number of registered users allowed to call. If the number of registered users dialed by a suspicious spam number on a website exceeds the set number of registered users allowed to call, mark the suspicious spam number on that website as a spam number.

[0015] Preferably, the information leakage assessment coefficient corresponding to each target website is analyzed, and the specific analysis process is as follows:

[0016] Obtain the number of registered users who received harassing calls on each target website and the number of harassing calls on each target website, and analyze the harassment assessment coefficient corresponding to each target website, recorded as α 1j , where j represents the number corresponding to each target website, j = 1, 2...n;

[0017] According to the call recordings corresponding to each harassing call on each target website, the risk assessment coefficient corresponding to the registered users on each target website is obtained, which is recorded as α 2j ;

[0018] By calculating the formula Get the information leakage assessment coefficient α corresponding to each target website j , where ε1 and ε2 are the weight factors corresponding to the set harassment assessment coefficient and danger level assessment coefficient respectively.

[0019] Preferably, the harassment assessment coefficient corresponding to each target website is obtained through analysis. The specific analysis process is as follows:

[0020] By calculating the formula Get the harassment assessment coefficient corresponding to each target website, where β is the number of registered users who are allowed to receive harassing calls. is the number of permitted nuisance calls, β j is the number of registered users who received harassing calls on the jth target website, is the number of harassing calls in the jth target website, κ1 is the weight factor of the set number of registered users, and κ2 is the weight factor corresponding to the number of harassing calls.

[0021] Preferably, the analysis obtains the risk assessment coefficient corresponding to the registered users in each target website. The specific analysis process is as follows:

[0022] Extract the call recordings corresponding to each harassing call on each target website, extract the keywords in the call recordings corresponding to each harassing call on each target website using keyword extraction technology, compare the keywords in the call recordings corresponding to each harassing call on each target website with a set of dangerous keywords, and obtain the number of dangerous keywords in the call recordings corresponding to each harassing call on each target website. Count the number of keywords in the call recordings corresponding to each harassing call on each target website, and then obtain the proportion of dangerous keywords in the call recordings corresponding to each harassing call on each target website.

[0023] By calculating the formula Get the risk assessment coefficient α corresponding to the registered users in each target website 2j , where A is the proportion of reference dangerous keywords, A j is the proportion of dangerous keywords in the call recording corresponding to the harassing calls on the jth target website, T is the reference call recording duration, T j is the call recording duration corresponding to the harassing calls in the jth target website, ξ1 is the weight factor of the proportion of dangerous keywords, and ξ2 is the weight factor of the call recording duration.

[0024] Preferably, the analysis obtains the leaked websites, and the specific analysis process is as follows:

[0025] Compare the information leakage assessment coefficient of each target website with the set information leakage assessment coefficient threshold. If the information leakage assessment coefficient of a website is greater than the set information leakage assessment coefficient, then the target website is recorded as a leaked website, thereby obtaining each leaked website;

[0026] Preferably, the analysis of the leakage degree corresponding to each leaked website is carried out in the following specific analysis process:

[0027] The addresses corresponding to the harassing calls on each leaked website are recorded as the leaked addresses corresponding to each leaked website, and then the number of leaked addresses corresponding to each leaked website is obtained. At the same time, the leaked distances corresponding to each leaked website are obtained. The leaked distances corresponding to each leaked website are compared to obtain the leaked distances corresponding to each leaked website. If a leaked distance on a leaked website is longer than other leaked distances, then the leaked distance is used as the leaked distance corresponding to the leaked website. In this way, the leaked distances corresponding to each leaked website are obtained.

[0028] By calculating the formula Get the severity coefficient of the leakage situation corresponding to each leaked website, where i represents the number corresponding to each leaked website, i=1,2...m, R i Indicates the severity coefficient of the leakage corresponding to the i-th leaked website, L i represents the leakage distance corresponding to the i-th leaked website, L represents the set reference leakage distance, S i represents the number of leaked addresses corresponding to the i-th leaked website, S represents the number of reference leaked addresses, is the weight factor of the number of leaked addresses corresponding to the leaked website, is the weight factor of the leakage distance corresponding to the leaked website;

[0029] The leakage severity coefficient corresponding to each leaked website is compared with the set leakage severity coefficient threshold. If the leakage severity coefficient corresponding to a leaked website is greater than the set leakage severity coefficient threshold, the leakage of the website is judged to be serious. Otherwise, the leakage of the website is judged to be not serious. In this way, the leakage severity of each website is obtained.

[0030] The beneficial effects of the present invention are as follows: the anti-telecom fraud system based on the big data cloud platform provided by the present invention screens the call information of registered users to various target websites through a preliminary screening module, and then analyzes the information leakage of each target website through the call recording corresponding to each harassing call obtained; through the address corresponding to each harassing call in each leaked website, the leakage situation of each website is analyzed; thereby avoiding the economic losses that may be caused to users due to data leakage; improving the user experience, protecting the rights and interests of users, improving network security, maintaining the healthy development of the online community, and promoting a positive social atmosphere. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 paying any creative work.

[0032] Figure 1 This is a schematic diagram of the system structure connection of the present invention. DETAILED DESCRIPTION

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 making creative efforts are within the scope of protection of the present invention.

[0034] See also Figure 1 As shown, the present invention provides an anti-telecom fraud system based on a big data cloud platform, including: a user information acquisition module, a preliminary screening module, an information leakage degree analysis module, an information leakage judgment module, a leakage address analysis module and a display terminal.

[0035] The user information acquisition module is connected to the preliminary screening module and the leakage address analysis module respectively, the information leakage degree analysis module is connected to the preliminary screening module and the information leakage judgment module respectively, and the leakage address analysis module is also connected to the information leakage judgment module and the display terminal;

[0036] The user information acquisition module is used to obtain the corresponding registered users of each website within a specified time period, and then monitor the registered users on each website to obtain the corresponding call information of each registered user on each website, where the call information includes the phone number of each caller, the call recording of each caller, and the corresponding address of each caller;

[0037] It should be noted that the corresponding registered users of each website within a specified time period and the corresponding call information of each registered user on each website are obtained from the backend of each website;

[0038] A preliminary screening module is used to determine whether the registered users on each website receive a harassing number from the calling party's phone number corresponding to each registered user on each website. If it is determined that the registered user on a certain website receives a harassing number, the website is recorded as a target website, thereby obtaining each target website;

[0039] As an optional implementation method, it is determined whether the registered users of each website receive a harassing number. The specific determination process is as follows:

[0040] Compare the calling number of each registered user on each website with the calling number of the registered user. If the calling number of a registered user on a website is the same as the calling number of a registered user on a certain website, then determine that the calling number is the suspicious harassment number corresponding to the website. In this way, the suspicious harassment numbers corresponding to each website are obtained. Then, compare the calling number of each registered user on each website with the suspicious harassment numbers corresponding to each website. Count the number of registered users on each website that were called by each suspicious harassment number.

[0041] Compare the number of registered users dialed by each suspicious spam number on each website with the set number of registered users allowed to call. If the number of registered users dialed by a suspicious spam number on a website exceeds the set number of registered users allowed to call, mark the suspicious spam number on that website as a spam number.

[0042] The information leakage degree analysis module is used to obtain the number of registered users who received harassing calls on each target website, the number of harassing calls on each target website, and the corresponding call recordings of each harassing call, and then analyze the information leakage assessment coefficient corresponding to each target website;

[0043] As an optional implementation method, the information leakage assessment coefficient corresponding to each target website is analyzed. The specific analysis process is as follows:

[0044] Obtain the number of registered users who received harassing calls on each target website and the number of harassing calls on each target website, and analyze the harassment assessment coefficient corresponding to each target website, recorded as α 1j , where j represents the number corresponding to each target website, j = 1, 2...n;

[0045] In the above, the number of registered users who received harassing calls on each website and the number of harassing calls on each website are obtained. The specific acquisition process is as follows:

[0046] Obtain the harassing numbers corresponding to each target website, and then count the number of harassing numbers on each target website, and use this as the number of harassing calls corresponding to each target website;

[0047] Compare the calling phone numbers of each registered user on each target website with the harassing phone numbers on each target website. If a calling phone number in the call information corresponding to a registered user on a target website is the same as the harassing phone number on the target website, then record the registered user as the registered user on the target website who received the harassing call. In this way, the number of registered users who received harassing calls on each website is counted.

[0048] According to the call recordings corresponding to each harassing call on each target website, the risk assessment coefficient corresponding to the registered users on each target website is obtained, which is recorded as α 2j ;

[0049] By calculating the formula Get the information leakage assessment coefficient α corresponding to each target website j , where ε1 and ε2 are the weight factors corresponding to the set harassment assessment coefficient and danger level assessment coefficient respectively.

[0050] As an optional implementation, the harassment assessment coefficient corresponding to each target website is obtained through analysis. The specific analysis process is as follows:

[0051] By calculating the formula Get the harassment assessment coefficient corresponding to each target website, where β is the number of registered users who are allowed to receive harassing calls. is the number of permitted nuisance calls, β j is the number of registered users who received harassing calls on the jth target website, is the number of harassing calls in the jth target website, κ1 is the weight factor of the set number of registered users, and κ2 is the weight factor corresponding to the number of harassing calls.

[0052] As an optional implementation method, the risk assessment coefficient corresponding to the registered users of each target website is analyzed and obtained. The specific analysis process is as follows:

[0053] Extract the call recordings corresponding to each harassing call on each target website, extract the keywords in the call recordings corresponding to each harassing call on each target website using keyword extraction technology, compare the keywords in the call recordings corresponding to each harassing call on each target website with a set of dangerous keywords, and obtain the number of dangerous keywords in the call recordings corresponding to each harassing call on each target website. Count the number of keywords in the call recordings corresponding to each harassing call on each target website, and then obtain the proportion of dangerous keywords in the call recordings corresponding to each harassing call on each target website.

[0054] By calculating the formula Get the risk assessment coefficient α corresponding to the registered users in each target website 2j , where A is the proportion of reference dangerous keywords, A j is the ratio of dangerous keywords in the call recordings corresponding to harassing calls on the j-th target website, T is the reference call recording duration, Tj is the call recording duration corresponding to harassing calls on the j-th target website, ξ1 is the weight factor of the ratio of dangerous keywords, and ξ2 is the weight factor of the call recording duration.

[0055] In the above, the number of dangerous keywords in the call recordings corresponding to each harassing call on each target website is obtained. The specific implementation process is as follows:

[0056] Extract the call recordings corresponding to each harassing phone call on each target website, extract the keywords in the call recordings corresponding to each harassing phone call on each target website using keyword extraction technology, compare the keywords in the call recordings corresponding to each harassing phone call on each target website with a set set of dangerous keywords, if a keyword in the call recording corresponding to a harassing phone call on a target website is the same as a dangerous keyword in the set set of dangerous keywords, then determine that the keyword in the call recording corresponding to the harassing phone call on the target website is a dangerous keyword, in this way, extract the dangerous keywords in the call recordings corresponding to each harassing phone call on each target website, and then obtain the number of dangerous keywords in the call recordings corresponding to each harassing phone call on each target website.

[0057] The information leakage judgment module judges the leakage of each target website information according to the information leakage assessment coefficient of each target website, and then obtains each leaked website;

[0058] As an optional implementation, the analysis obtains the leaked websites. The specific analysis process is as follows:

[0059] The information leakage assessment coefficient of each target website is compared with the set information leakage assessment coefficient threshold. If the information leakage assessment coefficient of a website is greater than the set information leakage assessment coefficient, the target website is recorded as a leakage website, thereby obtaining each leakage website.

[0060] The leaked address analysis module is used to obtain the addresses corresponding to the harassing calls on each leaked website, and then analyze the leakage situation corresponding to each leaked website;

[0061] As an optional implementation, the leakage level of each leaked website is analyzed. The specific analysis process is as follows:

[0062] The addresses corresponding to the harassing calls on each leaked website are recorded as the leaked addresses corresponding to each leaked website, and then the number of leaked addresses corresponding to each leaked website is obtained. At the same time, the leaked distances corresponding to each leaked website are obtained. The leaked distances corresponding to each leaked website are compared to obtain the leaked distances corresponding to each leaked website. If a leaked distance on a leaked website is longer than other leaked distances, then the leaked distance is used as the leaked distance corresponding to the leaked website. In this way, the leaked distances corresponding to each leaked website are obtained.

[0063] By calculating the formula Get the severity coefficient of the leakage situation corresponding to each leaked website, where i represents the number corresponding to each leaked website, i=1,2...m, R iIndicates the severity coefficient of the leakage corresponding to the i-th leaked website, L i represents the leakage distance corresponding to the i-th leaked website, L represents the set reference leakage distance, S i represents the number of leaked addresses corresponding to the i-th leaked website, S represents the number of reference leaked addresses, is the weight factor of the number of leaked addresses corresponding to the leaked website, is the weight factor of the leakage distance corresponding to the leaked website;

[0064] The leakage severity coefficient corresponding to each leaked website is compared with the set leakage severity coefficient threshold. If the leakage severity coefficient corresponding to a leaked website is greater than the set leakage severity coefficient threshold, the leakage of the website is judged to be serious. Otherwise, the leakage of the website is judged to be not serious. In this way, the leakage severity of each website is obtained.

[0065] The display terminal is used to display each leaking website and the leakage situation corresponding to each leaking website.

[0066] The present invention provides an anti-telecom fraud system based on a big data cloud platform. The system screens the call information of registered users to target websites through a preliminary screening module, and then analyzes the information leakage of each target website through the call recording corresponding to each harassing call obtained; and analyzes the leakage of each website through the address corresponding to each harassing call on each leaked website. This avoids the economic losses that may be caused to users due to data leakage, improves the user experience, protects the rights and interests of users, improves network security, maintains the healthy development of the online community, and promotes a positive social atmosphere.

[0067] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. An anti-telecom fraud system based on a big data cloud platform, characterized in that: include: The user information acquisition module is used to obtain the corresponding registered users of each website within a specified time period, and then monitor the registered users on each website to obtain the corresponding call information of each registered user on each website, where the call information includes the phone number of each caller, the call recording of each caller, and the corresponding address of each caller; A preliminary screening module is used to determine whether a registered user on each website has received a harassing number based on the calling phone number corresponding to each registered user on each website. If it is determined that a registered user on a website has received a harassing number, the website is recorded as a target website, thereby obtaining each target website; The information leakage degree analysis module is used to obtain the number of registered users who received harassing calls on each target website, the number of harassing calls on each target website, and the corresponding call recordings of each harassing call, and then analyze the information leakage assessment coefficient corresponding to each target website; The information leakage judgment module judges the leakage of each target website information according to the information leakage assessment coefficient of each target website, and then obtains each leaked website; The leaked address analysis module is used to obtain the addresses corresponding to the harassing calls on each leaked website, and then analyze the leakage situation corresponding to each leaked website; A display terminal is used to display each leaking website and the leakage situation corresponding to each leaking website; The specific process of determining whether a registered user on each website receives a harassing number is as follows: Compare the caller phone numbers in the call information of each registered user on each website. If a registered user on a website has the same caller phone number as a registered user on another website, then determine that the caller phone number is the suspicious harassment number corresponding to the website. This will determine the suspicious harassment numbers corresponding to each website. Then, compare the caller phone numbers in the call information of each registered user on each website with the suspicious harassment numbers corresponding to each website. Count the number of registered users on each website that were called by each suspicious harassment number. Compare the number of registered users dialed by each suspicious spam number on each website with the set number of registered users allowed to call. If the number of registered users dialed by a suspicious spam number on a website exceeds the set number of registered users allowed to call, mark the suspicious spam number on that website as a spam number. The information leakage assessment coefficient corresponding to each target website is analyzed, and the specific analysis process is as follows: Obtain the number of registered users who received harassing calls on each target website and the number of harassing calls on each target website, and analyze the harassment assessment coefficient corresponding to each target website, which is recorded as , where j represents the number corresponding to each target website, ; According to the call recordings corresponding to each harassing call on each target website, the corresponding risk assessment coefficient of the registered users on each target website is obtained, which is recorded as ; By calculating the formula , get the information leakage assessment coefficient corresponding to each target website ,in 、 They are the weight factors corresponding to the set harassment assessment coefficient and danger level assessment coefficient respectively; The analysis yields the harassment assessment coefficient corresponding to each target website. The specific analysis process is as follows: By calculating the formula , get the harassment assessment coefficient corresponding to each target website, where The number of registered users who are allowed to receive harassing calls is set. The number of permitted nuisance calls is set. is the number of registered users who received harassing calls on the jth target website, is the number of harassing calls in the jth target website, is the weight factor of the number of registered users, is the weight factor corresponding to the number of harassing calls; The analysis obtains the risk assessment coefficient corresponding to the registered users on each target website. The specific analysis process is as follows: Extract the call recordings corresponding to each harassing call on each target website, extract the keywords in the call recordings corresponding to each harassing call on each target website using keyword extraction technology, compare the keywords in the call recordings corresponding to each harassing call on each target website with a set of dangerous keywords, and obtain the number of dangerous keywords in the call recordings corresponding to each harassing call on each target website. Count the number of keywords in the call recordings corresponding to each harassing call on each target website, and then obtain the proportion of dangerous keywords in the call recordings corresponding to each harassing call on each target website. By calculating the formula , get the corresponding risk assessment coefficient of registered users in each target website ,in For reference, the proportion of dangerous keywords. is the proportion of dangerous keywords in the call recordings corresponding to harassing calls on the jth target website, For reference call recording duration, is the recording duration of the harassing calls on the jth target website, is the weight factor of the proportion of dangerous keywords, The weight factor for call recording duration.

2. The anti-telecom fraud system based on a big data cloud platform according to claim 1, characterized in that: The leaked websites are determined. The specific judgment process is as follows: The information leakage assessment coefficient of each target website is compared with the set information leakage assessment coefficient threshold. If the information leakage assessment coefficient of a website is greater than the set information leakage assessment coefficient, the target website is recorded as a leakage website, thereby obtaining each leakage website.

3. The anti-telecom fraud system based on a big data cloud platform according to claim 1, characterized in that: Analyze the leakage of each leaked website. The specific analysis process is as follows: The addresses corresponding to the harassing calls on each leaked website are recorded as the leaked addresses corresponding to each leaked website, and then the number of leaked addresses corresponding to each leaked website is obtained. At the same time, the leaked distances corresponding to each leaked website are obtained. The leaked distances corresponding to each leaked website are compared to obtain the leaked distances corresponding to each leaked website. If a leaked distance on a leaked website is longer than other leaked distances, then the leaked distance is used as the leaked distance corresponding to the leaked website. In this way, the leaked distances corresponding to each leaked website are obtained. By calculating the formula , get the leakage severity coefficient corresponding to each leaked website, where i represents the number corresponding to each leaked website, , Indicates the severity coefficient of the leakage corresponding to the i-th leaked website, represents the leakage distance corresponding to the i-th leaked website, Indicates the set reference leakage distance, Indicates the number of leaked addresses corresponding to the i-th leaked website, Indicates the number of addresses of the set reference leak, is the weight factor of the number of leaked addresses corresponding to the leaked website, is the weight factor of the leakage distance corresponding to the leaked website; The leakage severity coefficient corresponding to each leaked website is compared with the set leakage severity coefficient threshold. If the leakage severity coefficient corresponding to a leaked website is greater than the set leakage severity coefficient threshold, the leakage of the website is judged to be serious. Otherwise, the leakage of the website is judged to be not serious. In this way, the leakage severity of each website is obtained.

Images